Williamson, Jamie. The Evolution Of Modern Fantasy - From Antiquarianism To The Ballantine Adult Fantasy Series

The Evolution of Modern Fantasy

This page intentionally left blank

The Evolution of Modern Fantasy From Antiquarianism to the Ballantine Adult Fantasy Series Jamie Williamson

Palgrave

macmillan

the evolution of modern fantasy Copyright © Jamie Williamson, 2015. Softcover reprint of the hardcover 1st edition 2015 978-1-137-51808-8 All rights reserved. First published in 2015 by PALGRAVE MACMILLAN® in the United States—a division of St. Martin’s Press LLC, 175 Fifth Avenue, New York, NY 10010. Where this book is distributed in the UK, Europe and the rest of the world, this is by Palgrave Macmillan, a division of Macmillan Publishers Limited, registered in England, company number 785998, of Houndmills, Basingstoke, Hampshire RG21 6XS. Palgrave Macmillan is the global academic imprint of the above companies and has companies and representatives throughout the world. Palgrave® and Macmillan® are registered trademarks in the United States, the United Kingdom, Europe and other countries.

ISBN 978-1-349-70433-0 DOI 10.1057/9781137515797

ISBN 978-1-137-51579-7 (eBook)

Library of Congress Cataloging-in-Publication Data Williamson, Jamie. The evolution of modern fantasy : from antiquarianism to the Ballantine adult fantasy series / by Jamie Williamson. pages cm Includes bibliographical references and index. 1. Fantasy literature— History and criticism. I. Title. PN56.F34W55 2015 809.3'8766—dc23

2015003219

A catalogue record of the book is available from the British Library. Design by Scribe Inc. First edition: July 2015 10 9 8 7 6 5 4 3 2 1

I would like to dedicate this to my mother, who flew this earth while this book was nearing completion

This page intentionally left blank

Contents

Preface

ix

1

1 2 7

Introduction: Charting the Terrain Assembling a Genre Criticism: Fantasy and Fantasy The BAFS Template and the Problems of Retrojected Homogeneity “Faery, or romance literature” A Time and a Place

12 23 39

2

The Eighteenth Century: The Forgotten Past The Canonical Narrative Revisions of the Classical Inheritance Quasi-Oriental Fictions Revisions in the English Poetic Tradition The Forgotten Past and Lost Antiquities Syncretism and William Blake’s Mythology Conclusion

47 47 49 51 55 58 65 66

3

Romantic Transformations Introduction Earlier Romantic Narrative Poetry Later Romantic Narrative Poetry: Neo-Elizabethan Processing Romantic Prose Fiction German Romanticism and Phantasmion

69 69 71 77 82 85

4

From Verse to Prose: The Victorian Period Introduction Victorian Verse Fantasy Victorian Fairy Tales for Children Victorian Fantasy for Adults The Prose Romances of William Morris

91 91 96 106 113 117

viii

CONTENTS

5

Twentieth Century: The Literary Canon Introduction Children’s Fantasy The Literary Branch of the BAFS Canon The Celtic Revival Before the War: Cabell and Some Others After the War: Eddison and Some Others Tolkien Conclusion: The Question of Influence

127 127 129 132 134 143 148 155 160

6

Twentieth Century: Popular Fantasy Introduction Dark Fantasy and Weird Tales: Lovecraft and Smith Sword and Sorcery: Howard, Leiber, and Some Others Unknown After Unknown to the Beginnings of the Genre Conclusion

167 167 169 172 178 180 186

7

Conclusion: 1960–80 The “First” Fantasy Writers 1960–74 1975–80 Conclusion

191 191 192 195 198

Notes

201

Index

233

Preface

The original impetus behind the present study when it was originally conceived, now nearly two decades ago, was the need, as it seemed to me, for a more adequate historical framework for the critical discussion of modern fantasy, specifically “fantasy” as connoted by the core literary work that was assembled in the 1960s and 1970s in the wake of the Sword and Sorcery revival and Tolkien-mania. While critical and academic discussion of fantasy has proliferated in those nearly two decades, and there have been some excellent studies, the historical framework for conceptualizing what ultimately arrived at Tolkien has still not been notably rethought since the 1970s studies of Lin Carter and L. Sprague de Camp. While the Carter and de Camp studies have their points, they are equally inadequate in many ways. The present study takes as its core point of departure, and as the focus of its latter chapters, what I have termed the Ballantine Adult Fantasy Series (BAFS) canon, the body of reprinted work that was, during the crucial formative stage of the fantasy genre, “canonized” as the “tradition” behind Tolkien. While Carter and de Camp are now names of a bygone era, and much of this canon has only been sporadically in print since the mid-1970s, the basic construction of the “tradition” still more or less holds. In his 1997 study Fantasy: The Liberation of Imagination, what seems to be the last to center itself in the pregenre canon, Richard Mathews assumes Carter’s historical framework largely intact, although the actual critical discussion is engaging and a considerable advance on Carter’s. Among the issues explored in the following pages is the fact that the pregenre canonical writers did not compose any conscious literary movement, did not collectively see themselves as members of a “genre,” and in fact did not call their work “fantasy.” This is an issue that does not seem to have attracted much discussion, though it seems to me that it needs to be considered in looking at the BAFS canon. Fantasy is often effectively posited as a sort of timeless Platonic Form, involving magic and invented preindustrial worlds. The canon was constructed by way of this Form: track back from Tolkien until you find the “first one.” The results of this approach are not entirely without merit: it did collectivize a body of material that has

x

PREFACE

some cohesion. On the other hand, this approach tends toward oversimplification and breeds a kind of tunnel vision. One area which that tunnel vision has largely eliminated from consideration in histories of fantasy has been the narrative poetry, some quite long, of the eighteenth and nineteenth centuries: work that engaged similar subject matter, identified itself with similar areas of premodern and traditional narrative, and was widely read by many of the writers of the BAFS canon. Another area, not neglected but needing some refinement of perspective, has to do with those “epics and romances and sagas”: they are generally alluded to rather indiscriminately as stuff from (vaguely) “way back then.” But modern access to these works is via scholarly editions, translations, epitomes, and retellings, themselves reflecting modern perspectives; to readers of two centuries ago, the medieval Arthurian romances seeing print for the first time were as new as Pride and Prejudice. My contention is that what we call modern fantasy was in fact a creative extension of the antiquarian work that made these older works available. The history here, then, begins in the eighteenth century. This is, obviously, a wide arc to cover, and the following, of necessity, treats individual authors and works with brevity; detailed close reading has been avoided. No doubt, some will take issue with what I’ve said about this author or that author, but regardless, I hope that the present framework will contribute to a much needed reformulation of the literary history extending back over two centuries prior to the Tolkien explosion of the 1960s. With the emphasis on the arc, a note on the bibliographic apparatus is in order. Taken to one extreme, a book with the broad canvas this one has could easily generate a webwork of reference longer than the work itself. In order to avoid undue intrusion, and also—particularly—with space limitations in mind, I have taken a sparer approach, as follows. Quotations, direct paraphrases, and facts/information that stem from one specific source I have cited. For more general background information, and facts/information that appear in multiple sources sufficient to be considered “common knowledge,” I have listed under “Secondary Work Consulted” those works that I frequently referred to over the span of my writing. I have not provided an extensive list of primary sources, insofar as in-text references provide sufficient information (author, title, sometimes publisher) for readers to track down copies. The listings under “Some Primary Source Editions” include anthologies that include material discussed as well as editions that, generally for scholarly reasons, stand above others that might be available. Finally, the apparatus appears on a chapter-by-chapter basis rather than in one cumulative listing.

1

Introduction Charting the Terrain

T

he coalescence of fantasy—that contemporary literary category whose name most readily evokes notions of “epic trilogies” with “mythic” settings and characters—into a discrete genre occurred quite recently and abruptly, a direct result of the crossing of a resurgence of interest in American popular “Sword and Sorcery” in the early 1960s with the massive commercial success of J. R. R. Tolkien’s The Lord of the Rings, the paperback editions of which had been motivated by the former, in the mid-1960s. Previously, there had been no identifiable genre resembling contemporary fantasy, and the work that is now identified as laying the groundwork for it (“pregenre” fantasy) appeared largely undifferentiated in widely dispersed areas of the publishing market. In the pulps between the wars, and in American genre book publishing between World War Two and the early 1960s, fantasy by writers like Robert E. Howard, Clark Ashton Smith, Fletcher Pratt, Fritz Leiber, and Jack Vance hovered between science fiction, horror, and action adventure fiction. On the other hand, work by Lord Dunsany, E. R. Eddison, James Branch Cabell, and Tolkien, who found “reputable” literary publishers, was not, in presentation, readily distinguishable from the work of Edith Wharton, D. H. Lawrence, E. M. Forster, and Ernest Hemingway, and it was apt to seem anomalous. Other work was absorbed by that modern catchall “Children’s Literature,” whether it reflected the authors’ intentions (as with C. S. Lewis’s Narnia series) or not (as with Kenneth Morris’s Book of the Three Dragons). It was a common perception that stories with the elements of content now associated with fantasy were, by their nature, suited especially to children. A differentiated genre did emerge quite rapidly on the heels of the Sword and Sorcery revival and Tolkien’s great commercial success, however— its form and contours most strongly shaped by Ballantine Books and its

2

THE EVOLUTION OF MODERN FANTASY

crucially influential “Adult Fantasy Series” (1969–74). By the early 1980s, fantasy had grown to a full-fledged sibling, rather than an offshoot, of science fiction and horror. By now, it has been around in more or less its present form long enough to be taken for granted. A brief account of the construction of fantasy as a genre, then, is an appropriate place to begin the present discussion. Assembling a Genre In 1960, there was no commercial fantasy genre, and when the term was used to designate a literary type, it did not usually connote the kind of material that came to typify the genre when it coalesced, particularly in the Ballantine Adult Fantasy Series (hereafter BAFS). But in the early 1960s, there was a swell of interest in what then became identified as “Sword and Sorcery” or, somewhat less pervasively, “Heroic Fantasy.” At the heart of this was reprinted material that had originally appeared in pulp magazines between the 1920s and the early 1940s,1 and occasionally later, or in hardcover book editions from genre publishers.2 Published as, functionally, a subcategory of science fiction, Sword and Sorcery rapidly became very popular. Newly identified and designated, there was not a huge amount of back material for competing publishers3 to draw on, and given the general unmarketability of such work during the preceding decade and before, it is not surprising that few writers were actively producing Sword and Sorcery.4 Demand soon overtook supply. In this context, Ace Books science fiction editor Donald Wollheim became interested in Tolkien’s The Lord of the Rings, which had generated something of a cult following among science fiction fans, though it had been released in hardcover in 1954–56 as a sort of prestige item by literary publishers (Unwin in the United Kingdom, Houghton Mifflin in the United States). The elements it had in common with the Sword and Sorcery that had been appearing were sufficient for Wollheim to suppose it would be popular with aficionados of the new subgenre. Duly described as “a book of sword-and-sorcery that anyone can read with delight and pleasure” on its first-page blurb, Ace Books published their unauthorized paperback edition in early 1965. The minor scandal attending the unauthorized status of the Ace Books edition, and its replacement later that year by the revised and authorized Ballantine Books edition, no doubt drew some crucial initial attention to the book, but that can scarcely account for the commercial explosion of the following year or two, which has now sustained itself for five decades. The Lord of the Rings sold quite well to Sword and Sorcery fans, but it also

INTRODUCTION

3

sold quite well to a substantial cross section of the remainder of the reading public, and it became a bona fide bestseller. The Tolkien craze in fact ballooned into something quite close to the literary equivalent of the thencontemporary Beatlemania. The result of this was something of a split phenomenon. There can be little doubt that the Tolkien explosion bolstered Sword and Sorcery to some degree and drew new readers to the subgenre who may otherwise have remained unaware of it. But Sword and Sorcery never became something that “everyone” was reading, as was the case with The Lord of the Rings, and its core readership remained centered in the audience that had grown up prior to the Tolkien paperbacks. In presentation, there was little to distinguish those Sword and Sorcery releases that followed the Tolkien explosion, through the second half of the 1960s and into the 1970s, and those that had preceded it. So there was Sword and Sorcery, and there was Tolkien. Ballantine Books clearly recognized this dichotomy. Not a major player in the Sword and Sorcery market, the firm was eager to strike out in a more Tolkien-specific direction. The initial results over the next few years were a bit halting and haphazard. The Hobbit followed The Lord of the Rings in 1965, and the remaining work by Tolkien then accessible was gathered in The Tolkien Reader (1966) and Smith of Wootton Major and Farmer Giles of Ham (1969). The works of E. R. Eddison, a writer Tolkien had read and enthused on, appeared from 1967 to 1969. The year 1968 saw the less Tolkienian Gormenghast trilogy of Mervyn Peake, as well as A Voyage to Arcturus by David Lindsay. Like The Lord of the Rings, these works were originally released by “reputable” literary publishers. The Last Unicorn, a newer work by young writer Peter S. Beagle published in hardcover by Viking the previous year, appeared in 1969. The more impressionistic cover artwork of these releases served to distinguish them from the Sword and Sorcery releases of Lancer, Pyramid, and Ace: no doubt Ballantine wished to attract Sword and Sorcery readers, but they were also attempting to attract that uniquely Tolkien audience that Sword and Sorcery did not necessarily draw. Enter Lin Carter. A younger writer who had begun to publish Sword and Sorcery, including Conan spin-offs in collaboration with de Camp, during the mid-1960s, Carter approached Betty and Ian Ballantine in 1967 with a proposed book on Tolkien. This was accepted and published as Tolkien: A Look behind the Lord of the Rings in early 1969. One of the chapters, “The Men Who Invented Fantasy,” gave a brief account of the nonpulp fantasy tradition preceding Tolkien, which dovetailed with what Ballantine had been attempting with their editions of Eddison, Peake, Lindsay, and Beagle. Sensing a good source for editorial direction, Ballantine contracted

4

THE EVOLUTION OF MODERN FANTASY

Carter as “Editorial Consultant” for their subsequent Ballantine Adult Fantasy Series, which commenced in spring 1969 (see 1Carter 269). The importance of the BAFS in the shaping of the fantasy genre cannot be overestimated. It was the first time that fantasy was presented on its own terms as a genre in its own right. Though the volumes were inevitably destined for the science fiction sections in bookstores, the “SF” tag was gone, replaced by the “Adult Fantasy” Unicorn’s Head colophon;5 the garish, often lurid cover art became softer colored, drifting toward the impressionistic and the surreal; the muscle-bound swordsmen battling ferocious monsters (with the free arm around a scantily clad wench) were replaced by Faerie-ish landscapes. It was also the first time the peculiar cross section of work now considered seminal in the genre was drawn together under a unified rubric; to this day, it stands as the most substantial publishing project devoted to (mainly) pre-Tolkien fantasy. Sheer quantity also lent the BAFS indelible impact. With 66 titles in 68 volumes published between 1969 and 1974 (regularly one and sometimes two a month before a slowdown in late 1972), the BAFS rapidly became the dominant force in fantasy publishing (whether tagged “SF” or not). There was no real competition. The bully pulpit engendered by this dominance gave the BAFS far-reaching influence in two crucial respects. First, it gave the BAFS the power of defining the terrain. In Tolkien: A Look behind the Lord of the Rings, Imaginary Worlds (a study of the newly demarcated fantasy genre published in tandem with the BAFS in 1973), and in dozens of introductions to Series titles, Lin Carter repeated an operative definition of what was now simply termed “fantasy”: “A fantasy is a book or story . . . in which magic really works” and, in its purest form, is “laid in settings completely made up by the author” (1Carter 6–7). Carter further stipulates that fantasy circles around the themes of “quest, adventure, or war” (2Carter ix). Some four decades later, a wildly prolific body of work unambiguously reflects the terms of this template, then newly formulated under the aegis of the BAFS.6 Second, the quantity of titles, with primary emphasis on reprints,7 gave to the BAFS the power of determining a general historical canvas and implicitly shaping a “canon” of fantasy. Carter’s introduction to the 1969 BAFS edition of William Morris’s The Wood beyond the World begins with the portentous declaration: “The book you hold in your hands is the first great fantasy novel ever written: the first of them all; all the others, Dunsany, Eddison, Pratt, Tolkien, Peake, Howard, et al., are successors to this great original” (2Carter ix). This basic contention, like the aforementioned definition, was repeated over and over again in Carter’s commentaries and books, with Cabell, Clark Ashton Smith, de Camp, Leiber, Vance, and a few others rotating into the list of Morris’s followers, depending on

INTRODUCTION

5

which recitation you encountered. The dispersal by author of the BAFS titles suggests how the canon-shaping nature of Carter’s declarations were given body. The “major authors” were William Morris (four titles in five volumes), Lord Dunsany (six volumes), James Branch Cabell (six volumes), E. R. Eddison (four volumes), Clark Ashton Smith (four volumes), and Tolkien (six volumes).8 That the relevant work by Howard, Pratt and de Camp, Leiber, and Vance included in the BAFS was minimal in quantity9 reflects the fact that it was already available in editions by Ace, Lancer, and so on at the time, and Ballantine was not interested in issuing competing editions. On the basis of Carter’s oft reiterated “list,” however, those authors’ work should rightly be considered part of the BAFS canon, though little of it actually appeared in Series releases. Like the BAFS template, this informal canon has held through the succeeding decades. Despite its massive proportions and the breadth of the permutations of fantasy covered, John Clute and John Grant would declare in the introduction to The Encyclopedia of Fantasy (1997) that the late nineteenth- and twentieth-century authors representing “the heart of this enterprise” were “George MacDonald, William Morris, Lewis Carroll, Abraham Merrit, E.R. Eddison, Robert E. Howard, J.R.R. Tolkien, C.S. Lewis, L. Sprague de Camp and Fletcher Pratt, Fritz Leiber . . . and so on” (Clute and Grant viii). This is, more or less, the Carter/BAFS canon.10 Since the millennium, no doubt partly spurred by the renewed Tolkien boom following the Peter Jackson films, small publisher Wildside Press has mined the BAFS for titles for its classic fantasy series—even reprinting some Lin Carter introductions. It doesn’t always seem to be remembered that this “canon” was functionally constructed by Carter and Ballantine Books three to four decades ago, cobbled together from work of widely disparate publishing backgrounds. By 1974, then, a discrete genre, with a definition and a canon, had demonstrably emerged. Such a thing had not existed at all in 1960, and even in early 1969 it had consisted of a cross section of work appearing as a subbranch of science fiction (Sword and Sorcery) or as books for young readers,11 with a few titles presented as loosely “Tolkienian.” But while the discrete genre that emerged was predominantly shaped by Lin Carter and the BAFS, the series itself was not to last. BAFS releases decreased markedly through the latter part of 1972 and 1973, and in 1974 they ceased entirely.12 The degree of Ballantine’s dominance in the field can be seen in the partial vacuum left in its wake.13 No new BAFS appeared in the commercial market, though a small press, the Newcastle Publishing Company, followed Ballantine with its Forgotten Fantasy Library, augmenting but not repeating BAFS titles with 24 tradesized volumes between 1973 and 1979. Elsewhere, Bantam Books and Avon

6

THE EVOLUTION OF MODERN FANTASY

Books released a few newer books in BAFS style.14 But when a refurbished Ballantine reentered fantasy publishing in 1976, with a new look marked by the “fantasy realism” of the Brothers Hildebrandt and Darrell Sweet, and now edited by Judy Lynn and Lester del Rey, initial reprints were as often culled from work previously revived by the now defunct Lancer and Pyramid15 as from the former BAFS, and the latter usually appeared in garb reflecting the new aesthetic, with the Carter introductions eliminated.16 These works, and particularly the pre-Tolkien titles, were clearly no longer the core focus of Ballantine’s fantasy-publishing agenda. The major shift in focus from the mid-1970s on was an increasing emphasis on new rather than “classic” titles. As noted, few writers had been actively producing such work in the 1960s. But by the mid-1970s, this had begun to change fairly rapidly. While newer work and first publications had been in the minority in the BAFS, the frequency of reprints suggests that they were among the bestselling titles. Unlike the bulk of the “classic” reprints, Peter S. Beagle’s The Last Unicorn, Katherine Kurtz’s initial three Deryni books, H. Warner Munn’s Merlin’s Ring, Joy Chant’s Red Moon and Black Mountain, and Evangeline Walton’s tetralogy based on The Mabinogion17 all continued to be reprinted frequently through the del Rey period, with Kurtz’s series spawning uncounted sequels. In a reversal of previous proportions, five of the final seven BAFS titles in 1973–74 had been new titles. With an established genre, new writers began more frequently to write to its specifications, and when the distillation of a more formulaic, Tolkien-derivative approach produced two major bestsellers a couple years later,18 the “classic” BAFS authors and titles were overshadowed. The BAFS volumes were less frequently reprinted and became progressively more difficult to find. Since the focus of this study is the “canon” that was assembled by the BAFS—augmented by the relevant work available at the time through Ace Books, Lancer Books, and so on and hence not included in the series—I will break off the chronology here. Suffice to say, the bestseller genre that mushroomed in the late 1970s and early 1980s was an outgrowth of (and dependent on) developments in publishing that stretched back to the early 1960s. At a point in time when this bestseller genre has been around long enough to be taken for granted, it is important to note that it is not simply a timeless, unchanging entity, but was constructed, quite deliberately, to meet a new demand. The basic stages of the genre’s construction can be summarized thus: (1) the revival of interest in American Sword and Sorcery and the sudden commercial explosion of Tolkien’s work in the 1960s; (2) the isolation, naming, definition, and canonization of fantasy as a discrete genre between the late 1960s and mid-1970s, accomplished largely through the BAFS; and (3) the distilling of a bestseller formula for fantasy

INTRODUCTION

7

in the late 1970s, which completed what Ballantine had initiated in turning fantasy into a sibling rather than a subcategory of science fiction. The BAFS serves as the crucial and necessary hinge between the first and third, consolidating the former into a fully articulated genre and laying the necessary conceptual groundwork for the latter. Criticism: Fantasy and Fantasy With the distillation of the genre, of course, came criticism. That fantasy was indeed constructed is borne out by the notable dearth of critical discussion of it in any capacity prior to the 1970s.19 Tolkien’s essay, “On Fairy-Stories,” given as a lecture in 1939 and first published in Essays Presented to Charles Williams (1947), edited by C. S. Lewis, is probably the most often cited essay from the pregenre period for critics laying out their conceptual groundwork. Tolkien’s discussion—unsurprisingly, given the role of The Lord of the Rings in shaping the genre—floats ideas that cross fairly well with the contours of the genre. However, the modern works he cites tend to be Victorian and Edwardian works published for young readers.20 And of course, the literary form that Tolkien discusses is the “Fairy-Story”: “Fantasy” vacillates between a power of the human mind often evident in fairy-stories and a literary quality discernable in work in various forms. It is not treated as a literary form or genre. (I will return to this.) C. S. Lewis begins to chart something akin to the contemporary genre in his essay “On Science Fiction,” given as a lecture in 1955 and published in Of Other Worlds (1966), edited by Walter Hooper. But here fantasy is presented as a subbranch of science fiction. L. Sprague de Camp’s aforementioned introductions (see note 6) to his Pyramid Books anthologies, Swords and Sorcery (1963) and The Spell of Seven (1965), sketch the subsequent BAFS ground fairly precisely, though de Camp more specifically terms his focus “heroic fantasy,” and his approach is brief and highly summary. The first full-length studies focusing on the genre as it emerged through the 1960s date to the 1970s. The first of these was the aforementioned Imaginary Worlds (1973) by Lin Carter. Penned by its editor, and included as a title in the BAFS, the focus is, quite naturally, the canon constituted by the Series and related releases by Ace, Lancer, and so on. The core genre definition, the BAFS template, is the center of gravity, and Carter outlines in detail his historical framework, beginning with William Morris (with a nod back to traditional epic and romance) and proceeding through Dunsany, Eddison, Howard, and so on. The closing chapters form a kind of “how to” for aspiring “fantasy writers”—presumably for those writers whose work now proliferates on bookstore shelves.

8

THE EVOLUTION OF MODERN FANTASY

Another work of popular criticism was, not surprisingly, by L. Sprague de Camp, whose Literary Swordsmen and Sorcerers: The Makers of Heroic Fantasy was published by Arkham House in 1976. Apart from continuing to designate the form “heroic fantasy” and his rather arbitrary terminal date of 1950, de Camp’s ground is virtually identical to that of Carter (who provided the introduction to the study). His characterization of his subject as “tales of swordplay and sorcery in imaginary settings, where magic works” (1de Camp 4) is functionally identical to Carter’s. At the same time, de Camp takes a bit more of a “major authors” approach,21 includes more detailed (and often more reliable) biographical information, and demonstrates rather more critical acumen. A more academic study, C. N. Manlove’s Modern Fantasy: Five Studies, was published by Cambridge University Press in 1975. Unlike de Camp and Carter, Manlove was not an active player in the shaping of the genre, but his focus is, on the whole, amenable to the BAFS template. Manlove’s operative definition of fantasy, while couched in distinct terms, foregrounds similar concerns: “[A] fantasy is: A fiction evoking wonder and containing a substantial and irreducible element of supernatural or impossible worlds, beings or objects with which the mortal characters in the story or the readers become on at least partly familiar terms” (1Manlove 1; italics are Manlove’s). Of Manlove’s five authors (Charles Kingsley, MacDonald, Lewis, Tolkien, and Peake), only Kingsley was an addition to the core BAFS canon, and while some of the works discussed hover on the borderline of the BAFS template,22 they do not dramatically depart from it. Over the subsequent years, other studies followed the essential parameters of these: Manlove’s The Impulse of Fantasy Literature appeared in 1983; the first version of another popular study, Michael Moorcock’s Wizardry and Wild Romance: A Study of Epic Fantasy, appeared in 1987 and has been revised several times since;23 Richard Mathews’s Fantasy: The Liberation of Imagination was published in 1997. And again, the “heart” of John Clute and John Grant’s massive The Encyclopedia of Fantasy (1997) situates itself in the BAFS canon. However, some other studies of “fantasy” that appeared close on the heels of the genre’s emergence were apt to seem confusing to readers whose idea of it had largely been shaped by Tolkien, Sword and Sorcery, and the BAFS. W. R. Irwin’s The Game of the Impossible: A Rhetoric of Fantasy appeared in 1976. While many of the names associated with the BAFS canon are liberally sprinkled throughout its pages (Tolkien, Morris, Dunsany, Eddison, Lewis, MacDonald), most are simply passing references, and sustained discussions of their work or ideas are few. Conversely, works by Anthony Burgess, G. K. Chesterton, William Golding, David Garnett, and others not

INTRODUCTION

9

generally associated with the popular genre are discussed frequently and at length. It is quite clear that Irwin’s focus is not that of Carter, de Camp, and so on. Not surprisingly, Irwin’s delineation of fantasy as “that kind of extended narrative which establishes and develops an antifact, that is, plays the game of the impossible” (Irwin ix) carries a range of suggestion that is not particularly keyed to Morris or Eddison or Tolkien. Ironically, Irwin states that after 1957 the “spate [of fantasy] has all but run dry” (Irwin x)—implicit evidence that his focus is not what Ballantine dubbed fantasy. Rosemary Jackson’s Fantasy: The Literature of Subversion appeared in 1981 and moves more explicitly further from the BAFS canon and template: “The best-selling fantasies of Kingsley, Lewis, Tolkien, LeGuin, or Richard Adams are not discussed at great length . . . because they belong to that realm of fantasy which is more properly defined as faery, or romance literature” (Jackson 9). Closer to Jackson’s concerns are Maturin, Hawthorne, Kafka, and Pynchon. Jackson’s definition of fantasy reflects her key texts: “[F]antastic literature points to or suggests the basis upon which cultural order rests, for it opens up, for a brief moment, on to disorder, on to illegality, on to that which lies outside the law, that which is outside dominant value systems. The fantastic traces the unsaid and the unseen of culture: that which has been silenced, made invisible, covered over and made ‘absent’” (Jackson 4). This is clearly afield from the territory staked by Carter, de Camp, and so on; though developed in more overtly psychocultural terms than Irwin’s “game,” phenomenally it points in a quite similar direction. Again, these studies were apt to seem confusing to readers whose ideas about fantasy literature were shaped by the genre constructed in the wake of Sword and Sorcery and the Tolkien explosion.24 It is tempting to ascribe this discrepancy to the conventional popularversus-academic rift, with the former flawed for sloppy and imprecise terminology, while the latter, ensconced in the Ivory Tower, stubbornly ignores the fact of common usage. And there may be some substance to this: both Jackson’s and Irwin’s books are emphatically academic works of literary theory and criticism; all three of the nonacademic studies (Carter, de Camp, Moorcock) reflect the popular genre. In the introduction to his The Oxford Book of Fantasy Stories (1994), Tom Shippey remarks that “current academic definitions [of fantasy] . . . leave one wondering whether those who produce them ever stray into an ordinary bookshop at all” (Shippey xi).25 However, to simply ascribe the confusion, such as it is, to academic pique centers attention on who is disagreeing with who rather than on the actual substance of the disagreement. The essence of the issue circles on the

10

THE EVOLUTION OF MODERN FANTASY

use of the signifier fantasy: when Carter uses the term, he means one thing; when Jackson uses it, she means something quite different. Aptly, Shippey muses, “It is possible these arguments over definition are caused simply by reference to different things” (Shippey xii). The academic definitions of Irwin and Jackson in fact represent a continuation of the term’s usage prior to the 1960s. While the peripheral status of “fantasy” of any kind in the literary world prior to the 1960s meant that it bred little criticism, and there are no book-length studies, there are anthologies, the contents of which suggest what types of stories “fantasy” connoted at the time. Probably the two most widely circulated anthologies of “fantasy” stories between the latter years of World War Two and 1960 are Philip van Doren Stern’s The Moonlight Traveler: Great Stories of Fantasy and Imagination (1943) and Ray Bradbury’s Timeless Stories for Today and Tomorrow (1953).26 Both present their stories as, without any qualification, “fantasy.” Neither contains any stories in keeping with the BAFS template; nearly all the stories would respond well to the theoretical frameworks of Irwin or Jackson.27 In the BAFS sense, the term was appropriated. Note that during the initial push during the earlier 1960s, the protogenre material was dubbed “Sword and Sorcery” and, secondarily, “Heroic Fantasy.” The qualifier Heroic suggests that some qualification was needed, that just Fantasy was not enough.28 When Carter dubbed this work simply “fantasy” (with “Adult” qualifying audience, not form or content) in the late 1960s, it was the first time it had been collectively so designated. However, coming with an association with as widely read a book as The Lord of the Rings, while the earlier usage came attached to a fringe literary entity, common usage was bound to be affected considerably. Fantasy, later abetted by such extraliterary phenomena as Dungeons and Dragons, came to predominantly connote stories set in preindustrial invented worlds where magic works. The two are essentially different things. Sometimes they jockey for claim to the term; sometimes attempts are made to define the term so as to be inclusive of both. In the latter case, the bonding consideration is that both contain elements that are contradictory to our post-Enlightenment consensus reality. But this runs dangerously close to simply dividing literature in half, with one half adhering, in content and presentation, to the “real” (or, perhaps more appropriately, the demonstrable) and the other half not. “Fantasy,” in this case, becomes so broad as to be virtually useless as a term indicating anything about narrative form. This bonding consideration, however, is also the point of distinction. In the case of Irwin and Jackson, the most representative work focuses on a character who shares our post-Enlightenment consensus reality, and the narrative tension tends to emerge from that character being confronted with some phenomenon that

INTRODUCTION

11

contradicts the basic tenets of what he or she considers to be “reality.” This may come in the form of an inexplicable déjà vu that turns an emphatically common scene into something with a feeling of the “unreal”; it may be a radio that inexplicably begins to “tune in” to other residences in an apartment complex;29 it may be something that, deceptively, seems to connect it to the Carter/de Camp end of things—say, a water gnome appearing in mid-twentieth-century Long Island.30 In such cases, the dislocation resulting from the rationally inexplicable is precisely the point: for Irwin, this is the “Game of the Impossible”; for Jackson, it manifests something “outside” dominant cultural perceptions. The result may, at one extreme, be the dissolution of the protagonist’s perceptual center, loss of identity, and madness.31 At the other, the use of hardheaded, rationally (and scientifically) based common sense may, humorously, negate the seemingly impossible. But the “impossibilities” (what Carter and de Camp simply term “magic”) in Morris or Dunsany or Tolkien do not exist as such within the contexts of the worlds of their stories. The magical powers of the Lady of The Wood Beyond the World, of Ziroonderel in The King of Elfland’s Daughter, or of Gandalf in The Lord of the Rings are not “impossibilities” in their fictional contexts, and they do not open up any rifts in the fabric of the protagonists’ fundamental conceptions of reality. If Golden Walter, Alveric, or Frodo are at all shocked by such powers, it is by their degree (as a 747 might shock one used to propeller airplanes), not by the fact that they exist. If there is any “dislocation” caused, it is restricted to the reader, and the occurrence of magic does not so much challenge our consensus reality as disregard it: we must simply accept that magic is a part of the fictional reality.32 Though there do not seem to be any widely circulated discussions delving into the background of the mildly contradictory use of the term fantasy as signifier of a literary form, this does not mean that the essential phenomenal distinction is new. In the introduction to his anthology Black Water: The Book of Fantastic Literature (1983), Alberto Manguel distinguishes between “fantasy” and “the fantastic”: “Unlike tales of fantasy (those chronicles of mundane life in mythical surroundings such as Narnia or Middle-earth), fantastic literature deals with what can best be defined as the impossible seeping into the possible” (Manguel xvii). Jane Mobley divided her 1977 anthology, Phantasmagoria: Tales of Fantasy and the Supernatural, into two sections, “The Wondrous Fair: Magical Fantasy” and “The Passing Strange: Supernatural Fiction,” the first containing work amenable to the BAFS template, the second work more in keeping with the frameworks of Irwin and Jackson. Robert H. Boyer and Kenneth J. Zahorski, in their series of fantasy anthologies from the same period, distinguished

12

THE EVOLUTION OF MODERN FANTASY

between fundamentally the same things with their tags “High Fantasy” and “Low Fantasy.”33 The terminology is, perhaps, confusingly various, but the phenomenal distinction is strikingly consistent. Since the focus of the proceeding discussion will be on fantasy in the sense of the BAFS template, there is no need to labor this discussion beyond distinguishing, for purposes of clarity, between the term’s conflicting uses as a signifier and giving some account of that conflict. And my use of the term is strictly as a signifier: I have no intention of attempting to “define” fantasy beyond using it to point to a particular body of writing that (appropriately and inappropriately) was used to build a genre. Nor is there any need to argue about which usage of the term is “really” correct: the Irwin/ Jackson vein holds seniority, and perhaps keys itself more strongly to the psychological ramifications of the term; the Carter/de Camp vein reflects the popular genre and hence common usage. The BAFS Template and the Problems of Retrojected Homogeneity So fantasy, as the focus of the balance of this study, will be used to refer to work that is, more or less, amenable to the BAFS template—the type of narrative corollary to, or including, what has at various points been termed “high fantasy,” “epic fantasy,” “heroic fantasy,” or “sword and sorcery.” Using de Camp’s and Carter’s formulations as the most straightforward articulations of a functional lowest common denominator, this means narratives set in worlds in which the supernatural or magical are part of the fabric of reality and that center on the themes of quest, war, and adventure. To adapt that slightly for my own purposes, the worlds in which the supernatural or magical are part of the fabric of reality are not necessarily strictly “invented” and may include the world of the Arthurian legends (White’s The Once and Future King), the world of the Welsh Mabinogion (Kenneth Morris’s Fates of the Princes of Dyfed), the world of Scandinavian myth (Anderson’s Hrolf Kraki’s Saga), or romanticized regions (the Spain of Dunsany’s Don Rodriguez or the China of Ernest Bramah’s Kai Lung stories). Carter’s “quest, adventure, and war” I take to refer to the types of themes and tropes endemic to traditional heroic romance, epic, saga, and so on. De Camp’s stipulation of “preindustrial” with regard to settings I will take to implicitly mean settings modeled on those endemic to the preindustrial narrative forms just mentioned. However, as noted, the focus here will be on the pregenre material canonized by Lin Carter and the BAFS34 and on tracing the ancestry of this material back to the pre-Romantic period of the eighteenth century. It bears reiterating that this canon was a product of particular impeti (the

INTRODUCTION

13

early 1960s Sword and Sorcery revival and the mid-1960s Tolkien explosion) at a particular period in time (from c. 1960 to c. 1975), with a particular motivation (creating a viable publishing category) with particular results (among other things, a generic based criticism). In other words, the terms of the genre are largely retrojected onto the body of work that constitutes the canon, which was produced in contexts in which those terms had not been formulated and in which the collectivity presupposed by the genre did not exist. As mentioned, the paperback phenomenon of the 1960s and 1970s, with the BAFS as its centerpiece, was the first time this material had been drawn together under a unified rubric. What is the significance, and the degree of significance, of this fact? On the one hand, it may not seem particularly significant. Something previously without a collective name and identity was given a name and an identity with the advent of the genre. Certainly the relevant works of William Morris, Lord Dunsany, Robert E. Howard, and Fletcher Pratt fit the terms of the BAFS template well enough, and certainly their works stand as the natural forerunners of the genre as we now have it. There is certainly a logic in treating their work collectively—as I am doing in the present study. On the other hand, while the terms of the genre may serve as a convenient pointer to prior work “of interest,” and work that influenced later writers consciously writing fantasy, a more detailed and totalizing framework may easily become misleading. For example, on the simple level of original publication context, a cursory look at the canon reveals an immediate rift. As I have noted, Morris, Dunsany, Cabell, Eddison, and Tolkien emerged in the world of “literary” book publishing. By contrast, Lovecraft, Smith, Howard, Pratt and de Camp, Leiber, and Vance emerged in the ephemeral world of popular publishing—pulp and early genre-book publishing. The literary writers, with the exception of Dunsany, wrote primarily book-length narratives, while long works by the popular authors, with the exception of Pratt, rarely reached the length of the shortest of Morris’s romances. When the literary writers published in periodicals, it was in Harper’s rather than Weird Tales; when the popular writers published books, they were published by Arkham House rather than Houghton Mifflin Co. The work by the literary writers on the whole stands apart from other contemporary forms of fiction, whether literary or popular; the tropes and conventions of science fiction, horror, and adventure fiction actively inform the work of the popular writers. The prose styles of the popular writers (excepting Smith) tended to be unadorned and contemporary; the literary writers often developed carefully nuanced, archaized, poetic styles.35 An unreflecting jump, assuming the contours of the genre as a timeless constant, can lead quickly to the contention that, though the details of

14

THE EVOLUTION OF MODERN FANTASY

publication were different, the two groups of writers were, all told, doing the same type of thing: narratives set in worlds where magic works and so on. The stylistic and formal differences are simply the result of adapting this same type of thing to different publishing venues. But were they really trying to do the same type of thing? The terms of the “same type of thing” are essentially the terms of the BAFS template: both the literary and popular writers include elements— the magical worlds with their attendant wizards, magic swords, dragons, warriors, and so on—which became, essentially, generic trademarks when the genre coalesced. But again, the foregrounding of these elements as definitive in pregenre work comes predominantly with hindsight. There is little evidence suggesting that our pregenre writers considered these elements in themselves to collectively signal a particular literary form. On the other hand, the differences noted previously, taken collectively and without consideration of the later genre, are really rather substantial and underline the fact that the two groups of writers existed in very different literary worlds. Discernable “crossover” influence is limited. Simple chronology, of course, renders any influence of the popular writers on the literary largely impossible. However, it might be countered, some earlier popular writers who strongly influenced the popular fantasy of Weird Tales and Unknown were also read enthusiastically by some of the literary writers. H. Rider Haggard is perhaps the most noteworthy of these. But the nature of Haggard’s influence on, say, Howard, is broader and more generic than his influence on (to choose a literary writer who conceded his influence) Tolkien.36 In the latter case, the influence is predominantly evident in the atmosphere surrounding certain places: the Gates of Argonath, the ancient twin cities of Minas Tirith and Minas Morgul, the Paths of the Dead. It is not particularly evident in Tolkien’s prose and only occasionally in the textures of his narrative. In contrast, Howard’s prose and the actionbased fabric of his storytelling are quite reminiscent of Haggard: both did write “swashbucklers.” From the other angle, some influence might be discerned moving from certain of the literary writers to the popular writers. But even in the most notable case, that of Lord Dunsany, this is as limited as Haggard’s discernable influence on Tolkien. While the evocative names and fabulous atmosphere of many of Dunsany’s tales are clearly echoed in Lovecraft, Smith, and Leiber, his narrative distance, mannered irony, and the King James Bible–infused rhythms of his prose are not. Much of this will be discussed in greater detail in Chapter 6, so I will not elaborate further here. Suffice to say, the fantasy canon is not, from the standpoint of the contexts of the writers themselves, anything like a unified tradition that descends in linear fashion from the romances of William

INTRODUCTION

15

Morris to the 1960s and beyond. The retrojection of the BAFS template that underscores the contemporary genre can suggest otherwise. But rather than saying that the literary and popular writers adapted the same type of thing to their respective contexts, it would perhaps be better to say that the two moved out of distinct literary territories (with, as I shall develop, distinct literary ancestries), developed vocabularies of content similar in notable respects, and converged in the 1960s. Out of this convergence emerged the present day genre. But from the standpoint of the pregenre context, we should really speak of two traditions. A significant factor in the tendency to treat these distinct strands as a single literary tradition was the abrupt and subsequently widespread adoption of the unifying label fantasy. While, again, I do not oppose the use of the term as a convenient shorthand, the ramifications of its use with regard to pregenre work, and its limitations, have not on the whole received much consideration. For example, little attention has been paid to the fact that adoption of the generic term entails a shift in the consolidating point of reference from terms that may have been used by the authors in question, such as “romance” or “fairy-story,” to a word that the writers before the 1960s rarely used to classify their work. The comparative lack of attention, in turn, suggests that this is not particularly significant. Just as I contend that the differing contexts of the popular and literary branches of the pregenre fantasy canon are rather more significant than generally affirmed, I would contend that this shift in the terminological point of reference is more significant than might be supposed. First, a study beginning, naturally enough, with a close discussion and definition of “fantasy” (both Carter and Manlove do so) thereby consolidates itself around the etymological and semantic associations of that term. But if the authors in question did not classify their work by the term, but by other terms with (however subtly) different associations, and a critic professes to be concerned to a significant degree with “what they were trying to do” (1Manlove 258), this practice must be, to some degree, problematic. Second, the associations are not only etymological and semantic: studies of fantasy have, as I have noted, proliferated only since the genre was constructed. That is, they have emerged in a context where such a genre is widely recognized. In consequence, contours and emphases, not to mention a strongly articulated sense of collective affinity, are, with little question, imposed retroactively. My distinction between the popular and literary branches of the pregenre canon indicates that this, too, is problematic. I will pursue this in terms of the two key features on which critical discussions of fantasy tend to hinge: invented worlds and magic. The critical preoccupation with these features and issues contingent on them is mirrored in the concerns of many, perhaps most, contemporary fantasy

16

THE EVOLUTION OF MODERN FANTASY

writers working more or less within the BAFS template—Lin Carter’s series of “how to” chapters in Imaginary Worlds can be seen to provide a fairly accurate, if reductionist, sketch of these concerns. Certainly the maps, formulated background “mythologies,” attention to issues of sociopolitical history (often in the form of quasi-historical appendices), and so on that so frequently form a part of the apparatus of work from Le Guin’s initial Earthsea trilogy (1968–72), to Guy Gavriel Kay’s Fionavar Tapestry (1985– 86), to Robert Jordan’s Wheel of Time series (1990–2013) underscore the notion that the practical mechanics of developing invented worlds that will generate the illusion of existing independently of the author’s story are an explicit, and prior, concern of “the fantasy writer.” Likewise, the frequent overt preoccupation with the practical structure and nature of the magic in those worlds, from the Eleven Kingdoms of Katherine Kurtz’s Deryni series (1970–) to the Dales of Jane Yolen’s Great Alta trilogy (1988–97) to the recently concluded Harry Potter series (1997–2007), underscores the notion that this, too, is an explicit, and prior, concern of “the fantasy writer”: the magic itself must be given a clear, conceptual base and framework to be “credible.” As noted, both the work and the criticism that have followed the emergence of the genre can be said to reflect each other in lifting these crucial, defining issues into relief. But what of the earlier work that was written without the generic publishing and critical apparatus? Regarding “invented worlds,” most of the pregenre writers canonized during the BAFS period, both literary37 and popular,38 did employ some form of invented world (or, at least, country) in their relevant work.39 It scarcely takes Dunsany’s “we have new worlds here” (Dunsany Preface) in The Book of Wonder (1912) or Pratt’s introductory reference to “this other world” (Pratt xi) in The Well of the Unicorn (1948) to be persuaded that they were doing so quite consciously. At the same time, there is a great deal less development of the kinds of practical mechanics that are now commonplace. Maps appear in only the last of Morris’s romances (The Sundering Flood), in the Pape illustrated edition of The Silver Stallion but no other Cabell volumes, in Eddison’s Zimiamvia trilogy but not The Worm Ouroboros, in Pratt’s The Well of the Unicorn, and in Tolkien’s The Hobbit and The Lord of the Rings, but not in the pre-1960s editions of any of the other relevant work. Apart from Tolkien,40 only Dunsany, and to some extent Lovecraft,41 articulated anything much resembling a “mythology”42—and the “mythology” of Dunsany’s first two books, The Gods of Pegana (1905) and Time and the Gods (1906), has no integrated dramatic structure, unfolding in an almost whimsical fashion, and it had no connection to Dunsany’s subsequent work. Apart from Tolkien,43 none of the above developed sociohistorical contexts that extended beyond the immediate issues related

INTRODUCTION

17

to their stories, and even this only Eddison and Pratt did in considerable detail. In some cases (most notably Morris and Dunsany), “one has,” in C. S. Lewis’s words, “an uneasy feeling that the worlds . . . weren’t there at all before the curtain rose” (1Lewis 86). Regarding “magic,” in virtually all the pregenre work, magic is, of course, present in some form. But it is only rarely that we find a clear attempt to elaborate a conceptual framework with regard to its structure or function. Fletcher Pratt does so in The Blue Star, abetted by its rather Time Machine–ish prologue. The magic in Pratt’s Harold Shea collaborations with L. Sprague de Camp is constantly subjected to logical scrutiny, as is the magic of Poul Anderson’s Three Hearts and Three Lions. But elsewhere this does not tend to occur.44 For example, Eddison describes King Gorice’s “conjuring” in considerable detail in Chapter IV of The Worm Ouroboros, but the description, while an aesthetically powerful evocation of magic working, offers nothing of substance in terms of the conceptual nature of magic as an isolated phenomenon in Eddison’s world. The same may be said of Gandalf ’s magic working in both The Hobbit and The Lord of the Rings. It is difficult, in either case, to see magic in itself as a thematic concern of the author. Magic does arguably emerge as a theme in some of the work of Lord Dunsany, but the interest rarely becomes practical, moving rather in the direction of magic’s relation to Dunsany’s romantic conception of the imagination. When practical issues connected to magic arise, as in The Charwoman’s Shadow, Dunsany tends to become tongue in cheek. In Morris, magic is often quite muted and in some cases does not substantially affect the story at all.45 But even in those works where it does play a major, foregrounded role, as in The Wood beyond the World, there is even less extrapolation than with Tolkien and Eddison. This is not to say that magic is random or insignificant in this pregenre work but that its presence reflects a congruity with the styles and settings of the stories, and with the characters who wield it, rather than a “thematic” concern as such. The point here is not, obviously, that these defining elements of contemporary fantasy do not occur in the earlier, pregenre material. As these examples affirm, they do. Nor is it to say that there is anything amiss, in itself, in the critical discussion of these elements in the earlier work, with an eye on the background of the genre. However, the preceding examples do suggest that, with the possible exception of work connected (whether through publication, author association, or influence) to Unknown,46 invented worlds and magic were not accorded the same isolated significance, in degree or kind, by the pregenre writers as has come to be assumed over the past several decades. It is questionable in virtually all the cases (again, with the possible exception of the Unknown-connected work) whether the practical mechanics of “world making” formed an explicit and

18

THE EVOLUTION OF MODERN FANTASY

prior concern. There is, naturally, a concern with internal consistency, but this in itself can be seen as simply an extension of the concerns of fiction writing in general. But even in the cases where there seems to be a more involved preoccupation with such mechanical issues, there is room to suppose that other concerns than fantasy “world making” are at work. For example, Eddison’s thematic interest in Machiavellian intrigue and “High Politic,” particularly in the Zimiamvia trilogy, would provide sufficient explanation for his focused attention to political nuance and historical context. Nowhere in his fairly lengthy letters of introduction to either A Fish Dinner in Memison or The Mezentian Gate does he so much as mention a specific concern with “world making” (or magic). Fletcher Pratt’s meticulous attention to political and historical context in both The Well of the Unicorn and The Blue Star may simply be seen as an extension of his personal interest in history itself. A crucial paragraph in the “Author’s Note” to the former work expands on repeating patterns in “histories real or imagined (and this is not to draw a line between the two)” (Pratt xi). It should come as little surprise that the freelancing Pratt was, in fact, a popular historian, more well-known at the time of his death in 1956 as a historical writer. Magic, as I have noted, is lifted into consistent practical thematic relief only in Unknown and following. On this basis, then, I would contend that to profess a concern with “what they were trying to do,” while simultaneously adhering to the notion of invented worlds and magic as elements to be isolated as definitive to pregenre work, is problematic—just as working out of a definition built on the etymological and semantic associations of the word fantasy, if the authors did not categorize their work as fantasy, is problematic. The problems this practice engenders are often evident in the criticism, and it is particularly unfortunate when the criticism becomes evaluative. For example, C. N. Manlove states that the goal of his Modern Fantasy: Five Studies is “to take a range of modern ‘imaginative’ fantasists and show what they were trying to do and how well they do it” (1Manlove 258). “What they were trying to do” finds its bottom-line summary in the definition cited earlier, which thereby becomes the basic yardstick for determining “how well they do it.” But the works Manlove discusses, by the five authors also noted earlier, are extremely heterogeneous, even without considering the absence, in their respective time periods, of any collective genre resembling that which Manlove articulates. To what degree can Tolkien’s The Lord of the Rings, Kingsley’s The Water Babies, MacDonald’s Phantastes, and Mervyn Peake’s Titus Groan be meaningfully measured against an identical yardstick? Yet Manlove uses his uniform framework to conclude that “not one of the people we have looked at sustains his original vision”

INTRODUCTION

19

(258) and that the willingness to tolerate the failings must be left “to the cultists” (261). It should be noted that Manlove is not, in practice, simplistically reductionist, and he does affirm that there is a substantial “variety” in his choices of texts and authors—that there are among them “a range of approaches to the supernatural” and to the “fantastic worlds”—but he nevertheless persists in classifying them as a “kind” (12). He does begin with, and returns to as the basis of his evaluation, his definition, and he clearly assumes a generic affinity sufficient to support the definition as such. And the features distinguishing the works he discusses as a “kind” are the “fantastic worlds” and “supernatural” that mark the core BAFS template. The individual “original vision” of each author is subsidiary to the generic collective. But if one really takes the works he discusses on their own terms, these features must be seen as limited, and rather misleading, lowest common denominators,47 not definitive features indicating sufficient commonality that a uniform evaluative framework might be derived from them. This certainly does not mean that there are no aspects of these authors’ work that may be taken as flaws: this is not the indignant rejoinder of a “cultist.” But it does challenge Manlove’s framework as an apt basis for articulating those flaws, and in places there is arguably a subtle manipulation of the texts that serves to give credence to his essentially categorical judgment.48 Needless to say, Manlove is retrojecting the contours of a formulated genre onto work written before that genre had become a discrete collective entity.49 The BAFS template can be used to expose “flaws” in pregenre material even more reductionist. Morris’s failure to develop his invented settings in more detail can be taken as a flaw. Lack of attention to application of practical anthrohistorical “knowledge” can lead to the idea that the lack of organized religion in Tolkien’s The Lord of the Rings undermines the “believability” of Middle-earth (see 1Carter 122–24). In general, lack of explicit attention to the mechanics of magic, or failure to provide a systematic conceptual structure within which magic operates, may be seen as injurious to the supposed need of readers to have the seemingly impossible given a “credible” framework, if not explanation. And so on. The underlying assumption that begins to take shape is that the pregenre writers did not really know what they were doing (writing fantasy) and must be seen as anticipating in (at best pardonably) primitive terms what became more clearly understood later. I will return here to the question of generic signifiers. That the pregenre writers did not class their work as fantasy (replete with the post-BAFS web of associations)50 does not mean that they did not use other signifying terms. Though there was no systematic, single designation, the two most

20

THE EVOLUTION OF MODERN FANTASY

frequently recurring were fairy tale/story and romance. That fantasy is in fact a substitution, a shift in the signifying referent away from the authors’ chosen term(s), does not seem to have been widely remarked on, and there has not been in the criticism much to suggest that it is of any significance. Manlove, for example, opens his study noting the widely various ways the term fantasy is used, and he states in response, “[A]ll that matters ultimately is the isolation of a particular kind of literature . . . the name is relatively unimportant” (1Manlove 1). That he does not devote any focused attention to such generic terms as may have been used by the authors he discusses implicitly suggests that those terms are interchangeable in their significance with fantasy. But are they in fact interchangeable? That this is not merely semantic hairsplitting can be seen by examining a passage from another critical study of fantasy, Ann Swinfen’s In Defense of Fantasy: A Study of the Genre in English and American Literature since 1945 (1984).51 She is one of the few critics to even note that there has been a terminological shift, but she sidesteps it in a curious passage. Noting the importance of Tolkien’s understanding of “fantasy” to an understanding of “the genre,” she writes, Tolkien’s lecture [“On Fairy-Stories”] is concerned with the nature, origins, and purpose of “fairy-stories,” a term which proves not easy to define. A study of the genre reveals that such stories are rarely concerned with fairies. “Most good ‘fairy-stories’ are about the aventures of men in the Perilous Realm or upon its shadowy marches” (Tolkien 113). The nature of the fairystory thus depends on “the nature of Faerie: the Perilous Realm itself, and the air that blows in that country” (Tolkien 114). The term “fairy-story” is thus misleading, and though Tolkien continues to use it in his lecture, it will not be used in this study. Instead, the term “fantasy” has been preferred, as having perhaps a wider currency now than in the 1930s. (Swinfen 4–5)

The essence of this suggests that, to Swinfen, the shift from fairy-story to fantasy as the consolidating point of reference is little more than incidental, a simple replacing of one word with an equivalent with “wider currency.” Implicitly, her suggestion is the same as Manlove’s: the name is “unimportant”; the “isolation of a particular kind of literature” is. Of course, the question, once again, is whether or not the two terms are in fact sufficiently equivalent as to be treated interchangeably. Certainly Tolkien, a meticulous linguist particularly absorbed in the issue of names, would have contended that a change in name entails at least some change in meaning. “Fairy-story” is the narrative form he discusses, however expansive his sense of it may be; “fantasy” is, alternately, a power of the human mind and a literary effect, often evident in fairy-stories (as well as

INTRODUCTION

21

in other literary forms), subsidiary to it. Swinfen says the term fairy-story is “misleading,” but Tolkien’s qualification of the term is meticulously clear: “Fairy-stories in normal English usage are not stories about fairies or elves, but stories about Fairy, that is Faerie, the realm or state in which fairies have their being” (Tolkien 113). What there is about the term that may potentially be misleading, Tolkien clarifies: the fairy of fairy-story refers to setting, not character. Ironically, the preceding suggests that fantasy has the potential to be considerably more misleading. But to return to my question, are the two terms in fact equivalent? Some of Swinfen’s subsequent discussion is interesting in this regard: a few pages after she adopts fantasy in lieu of fairy-story, she writes, “In the subcreative art of fantasy, Tolkien detects three faces: ‘the Mystical towards the Supernatural; the Magical towards Nature; and the Mirror of scorn and pity towards Man’ (Tolkien 125). Unfortunately, by deliberately choosing to exclude two types of tale—the beast fable and the Lilliputian story— Tolkien largely excludes the mirror of scorn and pity” (Swinfen 6). Swinfen’s objection here does not accurately reflect Tolkien’s text. According to Tolkien, the “three faces” are not aspects of fantasy but of the fairy-story: Swinfen’s citation is immediately preceded by “fairy-stories as a whole have three faces” (Tolkien 125). Tolkien’s chief discussion of “fantasy” comes later in the essay, where, again, it does not refer to a literary genre but to a power of the human mind and a literary effect. Nor does he say that “the beast fable and the Lilliputian story” are not fantasy: he says they are not fairy-stories. His reasoning in both cases is clear enough: while fairy-stories concern themselves with “the aventures of men in the Perilous Realm,” human beings are peripheral or absent in the beast fable, and the beast form itself is simply “a mask upon a human face” (Tolkien 117). The human interface with Faerie, which he contends is a key feature of the fairy-story, is largely absent in the beast fable (see Tolkien 117–18).52 The “Lilliputian story,” Tolkien says, “belongs to the class of traveller’s tales,” and the travels of Gulliver do not lead to Faerie but remain “in this mortal world in some region of our own time and space” (Tolkien 115). By eliminating these tale types from consideration as fairy-stories, he does not exclude “the Mirror of scorn and pity towards Man,” or satire, as a possible element in the fairy-story: “I do not rule this story [“A Voyage to Lilliput”] out because of its satirical intent: there is satire, sustained or intermittent, in undoubted fairy-stories, and satire may often have been intended in traditional tales where we do not now perceive it” (Tolkien 115). If Tolkien had rejected Andrew Lang’s Prince Prigio (which he accepts as a fairy-story, though he is critical of it) or his own Farmer Giles of Ham, Swinfen’s contention that he had excluded satire might have some validity. As it is, she has simply misrepresented Tolkien’s text.

22

THE EVOLUTION OF MODERN FANTASY

It is clear that this interpretive confusion and misrepresentation stems, to a considerable degree, from terminology. Swinfen has, with virtually no qualification, substituted the word fantasy for Tolkien’s term fairy-story, implicitly suggesting that the two are functionally identical. But, though she does use Tolkien substantially in developing her working definition of fantasy (she also uses Dante, Coleridge, and Aristotle), it is quite evident that, though what she means by fantasy may overlap with what Tolkien means by fairy-story, the two are not identical. The beast fable and the “Lilliputian story” may not be fairy-stories according to Tolkien’s framework and yet be fantasy according to Swinfen’s. While Tolkien contends that these two tale types are out of place in Andrew Lang’s The Blue Fairy Book, it does not follow that he would have contended that they were out of place in Swinfen’s discussion. Indeed, he does affirm that tales such as Swift’s “report many marvels” (Tolkien 115) and that the beast fable is a “type of marvellous tale” (Tolkien 117), while Swinfen states that the “marvelous element . . . lies at the heart of all fantasy” and characterizes the marvelous primarily as “what can never exist in the world of empirical experience” (Swinfen 5). Needless to say, Tolkien does not suggest that Lilliputians or talking beasts “exist in the world of empirical experience” but that where the marvelous may be sufficient to make a story fantasy to Swinfen, it is not sufficient to make a fairy-story to Tolkien. In a sense, Swinfen’s discussion of Tolkien may be seen as a sort of microcosm: in quite condensed form, we can see the essential nature of the problems that arise when “fantasy,” with all its contemporary generic associations, is equated with the form—in this case, “fairy-story”—with which a pregenre writer consciously identified his or her work. Misrepresentations and misperceptions become inevitable. If one is to discuss the pregenre work and its origins with some emphasis on “what they were trying to do,” there is a need to divest that material from a too ready (and rigid) association with the terms of the contemporary genre. This is not to invalidate extant studies (such as Manlove’s) but to call attention to a common underpinning whose significance has largely gone unnoticed: contemporary terms and a contemporary framework are being projected backward onto work by writers who did not share precisely those terms and framework. If the interest is restricted to earlier manifestations of what became definitive elements of the fantasy genre when it coalesced,53 this is fine. However, if the interest is in how those writers saw their own work with regard to then-recognized narrative forms, and with what earlier strands of literary tradition they allied their work, the terms of the contemporary genre may be relevant in identifying who/what is “of interest,” but little more. Instead, one must centralize the associations those writers attached, insofar as can be ascertained, to the narrative forms they

INTRODUCTION

23

recognized and identified their work with: forms such as “fairy-story” and “romance.” “Faery, or romance literature” As the foregoing suggests, not only did the pregenre writers whose work collectively constitutes the de facto canon that emerged during the period of the BAFS not write with the specifications of our contemporary fantasy genre in mind; their work cannot on the whole be said to have comprised, in the consciousness of the authors, a singular genre under any other name either. Yet I am treating them collectively here. Can some feature common to the work as a whole, apart from the retrojected generic features of invented worlds and magic, be discerned that would not simultaneously impose an undue sense of homogeny? One thing that can be said is that the work characterizing the core, pregenre canon all draws to a substantial degree on themes and subject matter ultimately derived from what might be termed nonmodern, or traditional, narrative forms: myth, legend, epic, saga, romance, and fairy-story. The link is apparent not in the form of allusion (as in Joyce’s Ulysses or Eliot’s “The Waste Land”) but in the elements informing the actual narratives as narratives directly. This connection is implicitly made when Lin Carter asserts, rather portentously, that “fantasy is no less than the original form of narrative literature itself ” (1Carter 4) and then moves on to cite Gilgamesh, The Odyssey, Mahabharata and Ramayana, Shah-Nameh, Beowulf, The Kalevala, Spenser’s The Faerie Queene, and others to support the assertion (see 1Carter 13 and following).54 A quarter of a century later, Richard Mathews makes the same implicit connection when he states in the opening pages of his Fantasy: The Liberation of Imagination (1997) that “fantasy is . . . pervasive in the early literature of every culture” (Mathews 2) and precedes his text with a “Chronology” that begins with the same works that Carter cites (Mathews xv–xvi). Needless to say, these assertions need considerable qualification, and I will return to this. There is, nevertheless, a certain validity to Carter’s and Mathews’s point, and the generic terms used most frequently by our pregenre authors, fairy-story and romance, by their nature suggest something of this connection, while fantasy does not. Both of these terms, and particularly the latter, have proven rather resistant to set definition and have (like fantasy) been treated in enormously various ways. I do not propose to define them here. However, something might be said of some of the associations the terms carried for those of our authors who used them. The persistence with which they appear during the pregenre period in reference to works by Morris, Dunsany, Eddison,

24

THE EVOLUTION OF MODERN FANTASY

and Tolkien seems to suggest that they would be more apt signifying terms than fantasy: they would at least more clearly reflect which narrative traditions the authors connected their work with. Fairy-story is, in some senses, problematic: for more than two centuries, fairy-stories have been popularly classified as “children’s literature” in Britain and North America. Given this, it is not surprising that a writer like E. R. Eddison, whom many adults find difficult, would distinguish his work from fairy-stories. Eddison’s assertion in his “Letter of Introduction” to The Mezentian Gate that the work is “not a fairy-story” is followed immediately by “not a book for babes and sucklings” (Eddison xiv). That he is assuming the common association of the two is obvious. Later, when fantasy was emerging as a genre, Lin Carter would assert that a fantasy was “not a fairy-tale, not a story written for children” (1Carter 6). This longstanding and deeply rooted association was no doubt why the term was not that frequently used, even when a given work—say, Dunsany’s The Charwoman’s Shadow—played on motifs derived quite plainly from the traditional fairy tale. Tolkien, of course, was not shy of the term, and his use of it is a major reason for discussing it here. It is clear that he associated both The Hobbit (a work for children, though he later regretted it) and The Lord of the Rings (a work for adults) with the subject of his essay “On Fairy-Stories.” James Stephens’s collection of short narratives based on traditional Irish tales—akin to Evangeline Walton’s interpretations of The Mabinogion or T. H. White’s of the Arthurian legends—is titled Irish Fairy Tales. Tolkien’s friend and contemporary C. S. Lewis was unapologetic in his enthusiasm for fairy-stories, and his adult work, That Hideous Strength, was subtitled “A Modern Fairy Tale for Grown-Ups”—a subtitle that implicitly affirms the common associations. In the previous century, George MacDonald had also written unapologetically of the fairy-story in his essay “The Fantastic Imagination,” stating somewhat ambiguously of his own work, “For my part, I do not write for children, but for the childlike, whether of five, or fifty, or seventy-five” (MacDonald 317). But the pervasiveness of the idea that children were the natural audience for fairy-stories is evident in the fact that Tolkien, Lewis, and MacDonald all addressed it as a matter of course, and all three saw at least some of their fairy-story work published for children. Tolkien’s accounting for this has often been quoted: [T]he association of children and fairy-stories is an accident of our domestic history. Fairy-stories have in the modern lettered world been relegated to the “nursery,” as shabby or old fashioned furniture is relegated to the playroom, primarily because the adults do not want it, and do not mind if it is misused.

INTRODUCTION

25

It is not the choice of the children which decides this. Children as a class— except in a common lack of experience they are not one—neither like fairystories more, nor understand them better than adults do; and no more than they like many other things. (Tolkien 130)

The reference to “domestic history” points to a fact that Lewis and MacDonald were also aware of: previous to the nineteenth century, fairy-stories, whether literary or oral, were not composed specifically for children. The French contes de fees of the late seventeenth and early eighteenth centuries, the work from which the modern literary fairy-story form emerged as a distinct entity, were composed in the salons of Louis XIV’s France by adults for adults, and the process of making them amenable to children in later periods frequently involved extensive abridgement and bowdlerization, even full rewriting, so that later versions of the tales, for children, often bore the same relation to the originals that “A Voyage to Lilliput” as adapted in Andrew Lang’s The Blue Fairy Book bears to Swift’s work. The literary märchen of the German Romantics (by Novalis, Hoffmann, Tieck, and others), a particularly strong influence on MacDonald, were also for adults, sufficiently so that few of these were even adapted for children. The oral tales, such as began to appear adapted to written form in the collections of the Grimms, were composed and told in contexts that included adults as much as children. The fact that Eddison and Carter both distance themselves from fairy-stories does not mean that they were necessarily unaware of this: they were clearly speaking of fairy-stories as perceived in their literary historical contexts and dispensing with the issue in as efficient a manner as possible. Indeed, the fact that they felt constrained to mention it indicates awareness that at least some of the external features of the works in question had some connection to the fairy-story. That modern fantasy is strongly allied with the tradition of the fairystory (which more often than not is set in an invented world in which magic works) has, of course, been widely recognized. New interpretations of traditional tales and original fairy tales, from Robin McKinley’s Beauty (1978) to the tales collected by Ellen Datlow and Terri Windling in The Faery Reel: Tales from the Twilight Realm (2004), abound in the fantasy sections of bookstores. Criticism has not neglected the connection: C. N. Manlove, for example, opens his second study of fantasy, The Impulse of Fantasy Literature (1983), with the statement, “Modern fantasy owes its existence in large part to the traditional fairy tale” (2Manlove 1). This is true: the prolific output of literary fairy tales in France three centuries ago, in Germany two centuries ago, and in England during the Victorian period is not simply closely allied to but in fact a crucial part of the development

26

THE EVOLUTION OF MODERN FANTASY

of modern fantasy. And all three clusters of literary fairy tales drew directly on the narrative conventions of the traditional fairy tale for their substance. But Manlove’s claim, while along the right lines, is not entirely sufficient. Two of the authors he devotes full chapters to, Charles Williams and Mervyn Peake, owe a marginal debt at the very most to the fairy tale, traditional or literary.55 Eddison, to whom he devotes an unflattering nine pages in his curious chapter about what he calls “anaemic fantasy,” owes little beyond some surface features, even in The Worm Ouroboros, to the traditional fairy tale as Manlove demarcates it—that is, as seen in the collections of the Grimms and their folkloristic followers. The insufficiency of the claim is not that it is wrong but that it is unduly restrictive: one emerges from the opening chapter of The Impulse of Fantasy Literature with the sense that such traditional tales are the, rather than a, determining factor in the development of modern fantasy. He does not discuss, beyond suggestion in some of the titles he cites,56 any of the other types of narrative that also began to proliferate in print during the first half of the nineteenth century, interest in which was akin to the interest in the traditional fairy tale but which were not, formally, fairy tales. Tolkien’s discussion of the fairy-story is of more use here. In “On FairyStories,” Tolkien refers to literary tales (such as those of MacDonald and Lang) and to traditional tales in keeping with Manlove’s main thrust (“The Frog Prince” and “The Juniper Tree” from Grimm, for example). But he also refers to Sir Gawain and the Green Knight, the Arthurian legends generally, Volsunga Saga, poems from The Poetic Edda, and the ancient Egyptian “The Tale of the Two Brothers”—narratives that, in form, resemble the folk fairy tale to some degree but are formally distinct from it. Of course, all these works date to periods long before the emergence of the fairy tale proper.57 While Tolkien’s sense of the fairy-story is highly expansive, and he is retroactively imposing the (relatively) modern term on these latter works as much contemporary criticism has imposed the term fantasy on pregenre work, he does open up the field of inquiry to types of narrative, the modern interest in which dates to the period just preceding and contemporary with the Grimms, that Manlove neglects. These other types of narrative are as important to the development of modern fantasy as the traditional fairy tale. Tolkien’s passing equivocation, “fairy-story (or romance)” (Tolkien 155), at the beginning of the “Epilogue” to his essay invokes the second term. In its context, it also partly accounts for, in a general sense, Tolkien’s expansive use of the term fairy-story. But unlike fairy-story, which hovers only occasionally invoked despite its obvious connection to the pregenre material, romance was much more commonly used. Though William

INTRODUCTION

27

Morris generally did not discuss his work, and usually referred to his “fantasies” of the 1880s and 1890s simply as “tales,” contemporary reviewers and editors/critics from May Morris58 on tended to unqualifiedly refer to them as “romances.” George MacDonald’s Lilith is subtitled “A Romance,” and Phantastes “A Faerie Romance.” In the ersatz scholarly apparatus to Domnei and Jurgen, Cabell refers to his work in terms of romance. Eddison’s The Worm Ouroboros is subtitled “A Romance.” Kenneth Morris’s preface to The Fates of the Princes of Dyfed, in its concluding section, implicitly classes his work as a “Romance.” Poul Anderson’s “Forward” to the original edition of The Broken Sword begins, “This is frankly a romance.” Romance is also, in its own way, problematic. As I have mentioned, it is, unqualified, at least as vague and inclusive a term as fantasy. In the past two centuries, it has been used to designate work as various as Walpole’s The Castle of Oranto, Hawthorne’s The House of Seven Gables, Wells’s The Time Machine, and Mitchell’s Gone with the Wind, not to mention its current degraded association with “Candlelight” and “Harlequin.” C. S. Lewis, in his “Preface to Third Edition” of The Pilgrim’s Regress (1933/43), writes of the cognate “romantic(ism),” “I now believe it to be a word of such varying senses that it has become useless and should be banished from our vocabulary” (3Lewis 5). However, specific details, such as Tolkien’s equivocation of romance with fairy-story, MacDonald’s designation of Phantastes as a “Faerie Romance,” and Anderson’s immediately subsequent qualification of romance as “a story of admittedly impossible events and completely non-existent places” (Anderson Foreword), do suggest somewhat more precise associations. Cabell’s mock comparison of the tales of Manuel the Redeemer to the “several cycles” of King Arthur and Charlemagne in the foreword to Figures of Earth (Cabell xvii) goes further, implicitly connecting romance to the work to which it was originally applied: medieval romance. Of course, romance was not being used in an entirely restricted sense formally, and there is an implied extension to include other traditional narrative forms that bore some relation to medieval romance. For example, of the older works Tolkien mentions in his essay, only Sir Gawain and the Green Knight is technically a romance. Volsunga Saga belongs to the Scandinavian genre of the sogur, or “saga,” specifically the “Sagas of Ancient Times.” The poems of The Poetic Edda are mythic and heroic lays. The narratives of King Arthur and Charlemagne, which Cabell refers to, are not restricted to romance but exist in quasi-historical chronicles, such as Geoffrey of Monmouth’s Historia Regum Britannia, and in the chanson de geste, heroic poems such as The Song of Roland.59 More recently collected tales, such as those of the Grimms, were not romances in the technical sense.

28

THE EVOLUTION OF MODERN FANTASY

So if we take romance, as used by these pregenre writers, as something sufficiently narrow as to exclude Gone with the Wind and The Time Machine, but sufficiently broad as to not be restricted to its technical medieval application, we at least have a general notion as to the term’s implied associations. We also open up the field to narrative forms that were often of greater significance to the writers than the traditional fairy-tale form Manlove discusses. By taking this web of association as a general “glue” for discussing the pregenre canon, we will come closer to “what they were trying to do” than by centering discussion on the terms of the BAFS template. We also have a viable collective that does not misleadingly suggest consciousness of a contemporary genre per se. But these “traditional narrative forms” warrant some further discussion. What of the common practice that designates them fantasy? Certainly there are in common with contemporary fantasy elements of content that do not, shall we say, conform to a post-Enlightenment consensus reality: dragons, monsters such as Grendel, gods and goddesses appearing as dramatic characters, and so on. The attraction to such narrative material quite undeniably overlaps with the attraction of works like The Lord of the Rings. In this sense, it is not amiss to apply the term fantasy to these older narrative forms. But again, this necessarily centralizes a modern perspective on such narrative material. In their original contexts, Gilgamesh, Sir Gawain and the Green Knight, The Mabinogion, and so on were not conceived as fantasy. This has, of course, been duly noted. But the main distinction, implied or noted specifically by Carter, Mathews, Lewis, and Tom Shippey, has centered rather simply on the issue of belief. Citing the reference to dragons in the generally sober Anglo-Saxon Chronicle, Shippey notes that “in earlier periods they [dragons] may have been regarded as unusual rather than impossible,” whereas for us, living in a world with “no space left . . . for proper fire-breathing dragons to exist . . . they have accordingly moved firmly into the realm of fantasy” (Shippey x). In discussing a permutation of romantic that corresponds fairly closely to our fantasy, Lewis writes, “The marvellous is ‘romantic,’ provided it does not make part of the believed religion” (3Lewis 6). And Carter, implicitly suggesting the element of belief in the writers of “ancient fantasy,” asserts that “the earlier writers . . . wrote from a naïve and wide-eyed Weltanschauung” (4Carter 4). There is certainly validity to this basic point, though one may well posit that individual “belief ” in all eras was various, as it is in our own, and speculate that there were probably Anglo Saxons of a thousand years ago who did not believe in dragons, just as there are now people who believe in fairies; most likely, Homer and the Gawain poet were not “naïve and wideeyed.” I believe a more apt distinction would be that, for the earlier authors,

INTRODUCTION

29

the narrative matter generally formed a part of a living, received tradition, which may or may not have been literally “believed,” whereas the narrative matter of what we now call fantasy is not, or was not, part of a living tradition as received by its authors. The relationship between the fantasy of the pregenre canon and Carter’s “ancient fantasy” is predicated on not continuity but discontinuity and distance: it is through the salvaged relics of forgotten times, ancient or collected oral texts transformed into modern books, that the modern writers found their inspiration. Their choice of subject matter was decidedly against the grain in terms of what was deemed proper practice in the literary world (during the pregenre period) and founded, to a substantial degree, on an aesthetic in which the remote, the miraculous, and that which conforms to an order counter to the postEnlightenment consensus reality, are key. In contrast, the “ancient” authors’ choice of subject matter was congruent with their contexts (i.e., there was nothing “against the grain” in a poet of the Middle Ages turning to the Arthurian legends) and affirmed the orders, if not necessarily in a literal manner, of their respective worlds. A look at the particular traditions that those of our writers who wrote modern interpretations of extant stories (as opposed to creating original “mythos”) engaged can underline the idea of a relation founded on discontinuity with and distance from the source tradition. Among the writers of the pregenre canon, we have Kenneth Morris and Evangeline Walton reworking parts of The Mabinogion; James Stephens reworking old Gaelic tales; and T. H. White reworking the Arthurian legends, specifically Malory’s redaction. At a remove, we have Pratt/de Camp dropping Harold Shea into the worlds of Scandinavian myth, the Irish Ulster cycle, the Finnish Kalevala, Spenser’s The Faerie Queene, and Ariosto’s Orlando Furioso;60 at a greater remove, we have the Scandinavian/Celtic/European faerie cosmos of Poul Anderson’s original tale, The Broken Sword, and the extrapolation on the Carolingian-rooted world of the legends of Ogier the Dane in his Three Hearts and Three Lions. Needless to say, none of these works represent the passing on of living, received traditions: Morris was not a bard who derived his narrative matter from a living bardic tradition, White was not a medieval troubadour passing on his variation on what he had heard or read in handwritten manuscripts of King Arthur, and so on. All these authors were engaging creatively the written remains of narrative traditions long dead—remains that were processed by antiquarians and scholars after having been largely forgotten for centuries,61 transformed into mediated texts and translations, and then read in the form of a nineteenth- or twentiethcentury book by the authors. What the modern authors “knew” of the living contexts that bred the works that inspired them were the reconstructions (including mediated texts) of linguists, historians, anthropologists,

30

THE EVOLUTION OF MODERN FANTASY

and archaeologists. Again, the modern fantasy versions of these tales are predicated precisely on distance from the living contexts in which the matters originated. For example, White’s extrapolations on Malory, such as the assertion that “the central theme of Morte d’Arthur is to find an antidote to war” (White xvi), suggest not an organic development of tradition but the imposition of a twentieth-century pacifistic perspective on a work far removed from the twentieth century. In The Once and Future King, we see a modern writer using a body of story sufficiently remote as to be open and malleable advancing ideals that medieval authors such as Malory never would have dreamed of. We find a similar situation with Kenneth Morris and his imposition of the Theosophically hued notions of “the human soul on its evolutionary journey” (Morris xviii) on the matter of the “First Branch” of The Mabinogion. The distance implied by the fact that these works had to be recovered during the eighteenth century and following is the space that permits them to be creatively reimagined and drastically reconceptualized. By way of contrast, it is striking to note the virtually complete lack of fantasy interpretations of (suitable, it would seem) narrative material from the Bible, which does occupy a central place in the living, received traditions of European and Euro-American Christianity. This is not to say that the texts that compose the Bible are not, in fact, more ancient than most of those noted previously or that the world that bred the texts is not as or more remote. But it is to say that, since the introduction of Christianity, the Bible has had a continuous, unbroken presence in Western culture, its stories forming part of a living, received tradition, and there is attached to it a webwork of established exegesis that does not encourage the kind of creative reshaping we see in the work of White and Morris.62 But at least as significant as the hostility of certain orthodoxies to this kind of thing is the simple fact that the Bible—though it may be chronologically distant from the modern world, though the world from which it emerged may be culturally distant from the European and European-rooted cultures for which it became a sacred book—has been a constant presence for centuries up to nearly two millennia and a stable, known quantity that never had to be “recovered.” A stable, known quantity, I would suggest, is not what the modern imagination geared to “romance, or faery literature” most readily turns to for subject matter.63 The Greek and Roman classics stand in a somewhat ambiguous position here: Latin was, of course, the core language of educated culture (as well as the language in which the Bible as a text was known) in Western and Northern Europe throughout the medieval period, and many of the Roman classics, including mythic and legendary works such as Virgil’s The Aeneid and Ovid’s Metamorphoses, were, like the Bible, widely known in

INTRODUCTION

31

some form. Though Greek arrived in the West only during the sixteenth century, it was rapidly integrated into the educated culture, augmenting Latin. It was only at this point that the texts of Homer’s epics, for example, became known in the West, though they had been known of, and their subject matter familiar, throughout the Middle Ages. The Latin classics, therefore, a constant presence from the time of the late Roman Empire, were never lost and forgotten—never had to be “recovered.” Though Greek was only introduced in the sixteenth century, the recognition of its classic works by Latin authors, as well as the substantial derivation of Roman mythology from Greek, meant that it was not entirely an “unknown quantity.” Its integration into educated culture was decidedly different in nature from the halting retrieval and reconstruction by antiquarians of the works of the medieval period and of Celtic and Scandinavian antiquity— fundamentally unknown quantities circa 1750—several centuries later. At the same time, this work, not attached to the strictures of orthodox Christian belief, was more open to reinterpretation, and original works founded on its parts abound throughout the Middle Ages and the Renaissance. The relation of the Greek and Roman classics to modern fantasy is somewhat checkered: while it is less restricted than the relation of the Bible to fantasy, neither has it had the presence of medieval and Arthurian, Celtic, and Scandinavian material. For example, though figures from classical mythology appear in the works of Eddison, in Stephens’s The Crock of Gold, and elsewhere, there are no fantasy reinterpretations or expansions of classical narratives in the core pregenre fantasy canon.64 And while classical, and specifically Greek, material had a vigorous presence in some of the Romantic and Victorian work I will discuss in the following pages, the eighteenth-century push that led to it, where I will begin in Chapter 2, was marked by reaction against the classicism of the late seventeenth and early eighteenth centuries. I have suggested that a consolidating characteristic of pregenre fantasy is its peculiar relation to traditional narrative forms. The foregoing suggests that “traditional narrative forms” by itself is perhaps too specious and that certain traditions are more particularly significant. That those of Western and Northern Europe, and of Britain and Ireland, figure most prominently suggests that to many of our authors there was a more specific concern with the traditional narrative forms of their own perceived heritage. The work of James Stephens and Kenneth Morris can be said to carry a patriotic dimension; less explicitly, Tolkien’s work may be seen likewise, as well as, to a degree, William Morris’s. In a sense, these works may be partly seen as an imaginative recovery, an attempt to center the imaginative impulse in indigenous tradition, or an act of ethnic imagination—a creative answer to the scholarly process of antiquarian text recovery. While not universal

32

THE EVOLUTION OF MODERN FANTASY

among the pregenre authors, this apparent urge to connect to a perceived heritage is common enough to note as a significant force in the development of modern fantasy, and in fact it finds its initial manifestation in the Gothic Revival of the eighteenth century, which, I will argue in Chapter 2, is where this conscious attempt to engage archaic narrative forms from outside established tradition first began to manifest. While this underlying “patriotic” motivation was sufficiently common to cite it as a force in the development of modern fantasy, it was, again, not universal. James Branch Cabell engaged the conventions of medieval romance, in a manner not dissimilar to that of William Morris, with no apparent concern for any idea of “heritage” (he was, of course, American). With Eddison, whose British and Northern inspirations were crossed with Greek and Italian Renaissance inspirations, any conventional concern for heritage is remote. Yet, in literary terms, both were nevertheless engaging the vocabulary of “faery, or romance literature” in a sufficiently similar manner to the others as to be seen as part of fundamentally the same phenomenon. More entirely removed from any vestiges of a patriotic urge, yet still characterized by an imaginative preoccupation with the remote, the miraculous, and that which conforms to an order counter to the postEnlightenment consensus reality, is that branch of fantasy that first emerged in the eighteenth century following the appearance of Antoine Galland’s French translation, and the subsequent English translation from French, of The Thousand and One Nights: the pseudo-Oriental tale. Examples from the post–William Morris, pregenre period of the BAFS canon are comparatively thin: Ernest Bramah’s Kai Lung stories are the only major body of what is explicitly pseudo-Oriental work, to which one may add some of the short tales of Kenneth Morris and Donald Corley. However, many of the tales of Lord Dunsany, as well as the Zothique cycle of Clark Ashton Smith, are pervaded by an atmosphere indebted to the pseudoOriental tale; so, too, to a lesser degree, are the fantasies of Jack Vance and Fritz Leiber. There are occasional pseudo-Oriental touches in Tolkien and Eddison. And it is primarily in the pseudo-Oriental tale that Lin Carter found “forerunners” of William Morris: Beckford’s Vathek (1786), Meredith’s The Shaving of Shagpat (1855), and F. Marion Crawford’s Khaled (1890). The pseudo-Oriental tale, as I will discuss in Chapter 2, was in fact a far more prolific genre in the eighteenth century than Carter conveys, and it is arguably here that one finds the first substantial body of modern English narrative fantasy in prose. The “traditional narrative modes” underlying fantasy, then, are not simply the universal, leveling “myth, epic, saga, and romance” sometimes conveyed. We have a particular weight given to Celtic, Scandinavian, and

INTRODUCTION

33

medieval romance traditions; a less frequent recourse to classical tradition, and usually Greek rather than Roman; and a strong current derived from a specious “Oriental tradition,” itself largely a European invention,65 with The Thousand and One Nights serving as the most continuously influential constituent work. The Bible lurks as an indirect presence, the mytholegendary historical structure of the Old Testament sometimes influencing the shape of fantasy constructs, the King James version sometimes influencing prose styles, Old Testament narrative sometimes coloring the pseudo-Oriental tale. I have put a certain emphasis on those fantasy works that explicitly constitute interpretations of actual traditional narrative material or take place in the locale of some traditional narrative, as an easy way to identify what traditions figure most significantly in the BAFS canon. These centers are largely maintained in the work of those writers working within the “invented world/country mode” as well: Morris’s worlds align themselves clearly with Germanic antiquity, the Middle Ages, or an amalgamation of both, and his narrative techniques rely heavily on his readings of saga and romance; the substance of Dunsany’s worlds draw heavily on the pseudoOriental and Greek traditions, with an influx of medieval romance and the fairy tale, melded together by a style echoing the King James Bible; Cabell’s Poictesme is an imaginary French province, and his “medieval” narratives, such as Figures of Earth and The Silver Stallion, are presented as ersatz medieval French romances; Eddison, as noted before, drew from a cross section of Greek, Italian and English Renaissance, Celtic, and saga elements; Tolkien’s world emerged from the desire to create an imaginative English mythology, and his construct represents a syncretistic merging of elements drawn, most notably, from Germanic and Celtic legend. Pratt’s The Well of the Unicorn is strongly informed by saga narrative techniques. The work of these writers is, in a sense, no less dependent on, variously, Celtic, Scandinavian and Germanic, medieval, Greek, and “Oriental” sources than is the work of Kenneth Morris, Stephens, Walton, and White, though one might say the “source” material is more distinctly processed. I have noted that some of the distinct individual elements informing the work of Kenneth Morris and T. H. White betray a uniquely modern distance from the traditions they invoke and are in fact reliant on that distance. The inherent syncretism of the latter, “invented world” writers is absolutely predicated on distance: elements defining distinct, autonomous, living traditions become “literary” through the process that results in mediated texts, and these elements are mixed and matched by modern writers who wish to evoke an aesthetic associated with actual legendary, romance, and saga traditions while simultaneously not being constrained by those traditions. This results in, for example, Elizabethan-talking Homeric-cum-Celtic

34

THE EVOLUTION OF MODERN FANTASY

heroes existing in a world of Renaissance castles, Greek goddesses, and Jacobean revenge tragedy–style intrigues (i.e., E. R. Eddison). This body of background material is, then, what was connoted by the formal terms fairy-story and romance, and these were what our pregenre authors clearly allied their own work with and often looked to for models. But the models were not, as the preceding discussion suggests, rigidly deterministic: they were fluid and freely melded and adapted according to the intentions of the authors, even when the model was a single work (as with White and Malory). The models were not literary works of fantasy, in the sense connoted in our modern context; neither are the modern works organic developments of received tradition, even when there is an underlying strand of patriotism motivating them. But though this recourse to “traditional narrative forms” may be taken as a viable template for the discussion of the pregenre fantasy canon beginning with William Morris, taken by itself it can suggest something akin to Lin Carter’s, shall we say, reductionist representation of literary history: “Cervantes’ bravura lampoon of the chivalric romances and Spenser’s chaotic smorgasbord of the whole school, resulted in the death of the fantastic story for a couple of centuries. Then William Morris came along to draw it forth, like Lazrus, from the tomb” (2Carter xi). If one narrowly defines modern fantasy as a prose narrative form characterized by invented worlds and magic, something like this viewpoint may perhaps be viable—with a good bit of qualification. However, if we look at the canon of pregenre fantasy in terms of its recourse to traditional narrative forms, and do not restrict the two centuries preceding Morris’s romances to prose narrative, we find that, in fact, Morris was continuing a modern practice that extends back to the eighteenth century, the period when the retrieval of, and construction of modern mediated texts derived from, traditional literatures was inaugurated under the aegis of antiquarianism. But I should pause here to note an important distinction within the pregenre fantasy canon, proceeding from the divide between the “literary” and “popular” branches discussed earlier. I mentioned that the latter, unlike the former, developed in close proximity to other forms of popular genre fiction—popular forms of romance narrative, in fact—most notably horror, science fiction, and action adventure fiction. In the absence of any articulated fantasy genre along the lines of the BAFS template, it is probably more apt to see what I am calling popular fantasy as developing from those recognized genres. Each has a clear history—and the histories often overlap—stretching back through the nineteenth century: to the Victorian tales of ghosts and the supernatural, to the swashbuckling lost race romance, to the “scientific romances” of H. G. Wells; and from there to the Gothic novel and the romanticism bred by Hoffmann and Poe, to

INTRODUCTION

35

the historical adventure romance, to Mary Shelley and Jules Verne. While the popular strand of the canon does draw on themes and subject matter derived, ultimately, from older, traditional forms of “faery, or romance literature,” the relationship exists at a remove. For example, the tales of Conan the Barbarian may in some capacities be described as a “heroic cycle” and likened to traditional tales and cycles of tales concerning heroes such as Cuchulainn, Sigurd, or Jason. However, in their telling, Howard’s Conan tales betray little if any debt to The Tain, Volsunga Saga, or the Argonautica as texts. On the other hand, they do bear ample debt to Haggard and Burroughs, among others. What might be said here is that Howard adopted elements of content derived ultimately from works such as The Tain and so on but sublimated them to the conventions of swashbuckling adventure fiction. The primary determinant of narrative style and strategy is the popular genre of fiction, not the texts, or close translations of the texts, of the older, traditional works. What Howard has in common with those older texts he could as well have gotten from sources like Bulfinch. A similar remove marks virtually all the popular fantasy canon.66 The popular strand of the canon, then, connects very clearly to other recognized forms of prose fiction, contemporary and earlier, which mediate its connection to traditional narrative forms. Characteristically, the popular pregenre canonical writers adopt traditional elements of content, which are then sublimated to the narrative strategies of recognizable modern forms. The literary ancestry here is fairly clear and in fact includes those areas conventionally cited in studies of fantasy concerned with the pre-Morris roots of the genre.67 For this reason, while some attention will be given to this eighteenth- and nineteenth-century material—the Gothic novel, and so on—in the following chapters, the ancestry of the popular strand of the canon per se will be accounted for relatively briefly in Chapter 6. As noted previously, the literary branch of the canon, in contrast, was much more removed from other contemporary forms of prose narrative of the English-speaking world. The general absence of a speculative dimension in the literary work puts it in a different world from science fiction. While Gothic and Poe-esque touches are not infrequent, particularly in the auras surrounding evil characters and the atmospheres of the places associated with them, these darker effects are part of a larger fabric, not themselves the chief concern. While the plots are adventurous, and often build to “thrilling” conclusions, the literary works on the whole do not operate like stereotypical nineteenth- and early twentieth-century adventure fiction, whether of the historical variety or the lost race variety: they are not, on the whole, breathlessly paced “yarns,” a substantial part of whose attraction lies in the piling of thrilling event on thrilling event. Distant from

36

THE EVOLUTION OF MODERN FANTASY

these forms of romance narrative, the literary work is, needless to say, also distant from the dominant literary narrative form, the novel. This leaves the literary fairy tale, but, as we have seen, in the English-speaking world this was considered children’s literature, and the authors of the pregenre literary fantasy canon were not writing for children—as Eddison’s vociferous remarks in the “Letter of Introduction” to The Mezentian Gate indicate. So where does this leave us? First, it turns us to our traditional narrative forms. Where the popular writers largely adapted traditional elements of content to the conventions of modern narrative forms, the literary writers would be more aptly described as writing modern works actively engaging the conventions of traditional “faery, or romance literature.” From Morris to Tolkien, a common effect sought by the literary writers is the illusion that their work is traditional, the product of another age (and, often, place). There are numerous elements keying the reader to this intention. First, there is the “bardic” narrator persona, such as we find in Kenneth Morris’s The Fates of the Princess of Dyfed and Book of the Three Dragons, or scribal narrator persona, such as we find in William Morris’s The Sundering Flood: “I, who gathered this tale, dwell in the House of the Black Canons [on the Thames at Abingdon]” (my italics, 1Morris 2). There is a frequent allusion to received tradition as authority, as when William Morris’s narrator, after Ralph of Upmeads has told his tale to “the good Prior of St. Austin’s at Wulstead,” informs the reader that “it has been deemed not unlike that from this monk’s writing has come the more part of the tale above told” (2Morris 277). This is further cued in Morris’s conventional openings—“It is told . . . ,” “Long ago . . . ,” “Once upon a time . . . ,” “The tale tells . . . ,” and so on68—and in James Branch Cabell’s oft repeated variations on “It is a tale they narrate in Poictesme . . .”69 Both Cabell and Tolkien extend this to suggest whole manuscript and oral traditions, which are often discussed in quasi-scholarly apparatus,70 implicitly placing Cabell and Tolkien both in the role not of author but of editor and translator. But the literary writers do not restrict themselves to these, what could be, sleight-of-hand flourishes that share with the “letter in a bottle” ploy used in some popular fantasy the creation of an authorial fiction. But whereas, in the latter case, we begin reading the firsthand account of the end of Atlantis, or of the end of Arthur’s Britain,71 only to find that the ancient narrators have a solid command of Victorian and early twentiethcentury popular adventure fiction conventions, with the literary writers we find narratives that actively echo the conventions and textures of traditional romance and saga—the kind of narratives that formed some part of the content of medieval and ancient manuscripts.72 In many instances,

INTRODUCTION

37

this included the crafting of an archaized prose to emphasize the medieval/ ancient aesthetic. So the “traditional narrative forms” are on the whole much more directly significant as texts to the literary writers than to the popular writers. Where contemporary popular forms of romance shaped the popular work, more traditional forms of romance shaped the literary work. At the same time, it is important to reiterate that most of this material had been, as of about 1750, completely unknown, and its currency as of about 1900 was the result of “scientific” research. It is as post-Enlightenment phenomena that this material appears to modern readers: scholarly texts, translations, paraphrases and epitomes, adaptations for children. The framework for the understanding of the archaic literature that so informs the work of the pregenre literary fantasy writers was itself a tradition that had its origins in eighteenth-century antiquarianism. The literary writers also inherited a narrative tradition predicated in part on texts “recovered” and transformed into mediated texts of various sorts in the wake of eighteenth-century antiquarianism, as well as on a rethinking along Romantic lines of classical (particularly Greek) tradition. Discussions of the ancestry of fantasy, particularly as pertains to the pregenre literary canon (Tolkienian fantasy, we might loosely term it), have been hampered by an almost exclusive emphasis on prose narrative. William Morris, an unarguably important author, nevertheless looms a little too largely as a result. He was the first author to produce a large body of the kind of narrative I am discussing here in prose. If we look strictly to prose for predecessors to Morris, we are restricted to scattered, single works: MacDonald’s visionary Phantastes (1858), Sara Coleridge’s fairytale romance Phantasmion (1837), Thomas Love Peacock’s Welsh-based The Misfortunes of Elphin (1821), Thomas Hogg’s eccentric cross-hatch of border legend and the legends of the medieval magician Michael Scott The Three Perils of Man (1822), and scattered shorter tales by Benjamin Disraeli, John Sterling, and Edward Bulwer-Lytton. However, the work of William Morris’s earlier literary career includes the work on which his considerable contemporary reputation rested: a large body of extended narrative poetry comprising versions of various classical, medieval, and Germanic/Scandinavian myths and legends.73 Though verse, this body of romance stands as a natural forerunner to the prose romances in terms of style and language: it is fantasy. If we thereby open up the field to other such extended fantasy verse narrative, we find a prolific tradition, beginning tentatively in the eighteenth century, flowering during the Romantic period, and reaching a somewhat baroque conclusion in the Victorian period—at which point, with Morris and the subsequent literary writers of the pregenre fantasy

38

THE EVOLUTION OF MODERN FANTASY

canon, such narrative moved predominantly to prose.74 Oddly, though the body of early and traditional material generally considered to be “ancient fantasy” includes both poetic75 and prose76 works, discussions of the postseventeenth-century fantasy narrative tradition have omitted any serious consideration of verse narrative. Yet Romantic and Victorian narrative poetry must be considered an important shaping force on the post-Morris literary writers. Having grown up in the late nineteenth and pre–World War twentieth centuries, their formative early encounters with the Arthurian cycle would have included the work of Tennyson and Swinburne. Formative early encounters with the legends of the Volsungs, and early Scandinavian narrative tradition generally, would include poetic versions of those legends by Morris, Matthew Arnold, and Longfellow (along with the libretto of Wagner’s Ring cycle). Encounters with Celtic tradition would have included the largely invented Ossian of MacPherson as well as Ferguson’s Lays of the Red Branch. Other widely read narrative poems that would have colored the literary writers’ conceptions of “faery, or romance literature” include, among others, Keats’s “Lamia,” Coleridge’s “The Rime of the Ancient Mariner,” Moore’s Lalla Rookh, Southey’s “epics,” Longfellow’s Hiawatha, and Shelley’s “Alastor.” That Morris produced a large corpus of such work before moving to prose narratives that similarly feature archaized language and (though no longer retellings) direct engagement of traditional romance conventions suggests a natural continuity. That he wrote tentative verse fragments of at least three of the late romances,77 before turning to prose, suggests that the initial story ideas were not inherently attached to the prose medium. Tolkien’s later “The Lay of the Children of Hurin” and “The Lay of Leithian,” substantial narrative verse fragments of tales that also exist in prose versions, suggests a similar fluidity in the presumed connection of what we might call “fantasy narrative matter” with the poetic and prose mediums. These two fragments find their obvious ancestors in the extended narrative verse fantasies of the immediately preceding eras, some of which (notably those of Morris and Longfellow) Tolkien greatly admired. Like the literary prose fantasies of Morris and his followers, this material—in most cases quite obviously, considering that its subject matter, like that of Kenneth Morris or James Stephens, was most often adapted rather than invented—follows quite clearly on the heels, and in fact accompanies, the “recovery” of the traditional material and its transformation into mediated texts. That the elevated, archaistic language of these Romantic and Victorian “epics” and “poetical romances” is of a piece with the elevated archaistic language characterizing most nineteenth-century translations of traditional material (including those of William Morris) is suggestive of how closely the two were associated in the nineteenth-century imagination.

INTRODUCTION

39

The cadenced, archaistic language of Morris’s followers, in addition to the romance- and saga-based narrative textures and the general nature of the subject matter, would suggest that the pregenre literary fantasy canon was an organic outgrowth of both the scholarly tradition that built the framework within which the traditional material could be understood, and thereby processed imaginatively, and its contemporary poetic narrative tradition, which existed in fundamentally the same relation to the traditional material as the later prose work. The following three chapters, on the eighteenth century, the Romantic period, and the Victorian period, respectively, will attempt a relatively detailed account of the parallel development of these mutually interconnected phenomena (along with related issues along the way), with the intent of creating a viable context for the discussion of the late romances of William Morris, “the man who invented fantasy,” at the end of Chapter 4 and then the literary work from Morris to Tolkien in Chapter 5. The bulk of the focus, therefore, will be on background specifically pertinent to the literary strand of the pregenre literary canon. A Time and a Place Before moving on to the main body of my discussion, a few further notes are perhaps in order concerning, first, the choice of the eighteenth century as a beginning date and, second, the implicit (at this point) restriction of emphasis to British, Irish, and Anglo-American literature. The reasons for my choice of the eighteenth century rather than the Victorian period and the prose romances of William Morris as a beginning point should be evident from the preceding discussion and do not need to be repeated here. However, something further might be said with regard to earlier possible beginning points. For example, why would a work like Edmund Spenser’s The Faerie Queene (1590–96), which does engage antiquarian (for its author) romance elements in an idiosyncratically individual manner and in fact takes place in a largely invented world, not provide an equally appropriate starting place? Or why not begin with the indubitably fantastic “artificial” Chivalric prose romances of the fifteenth and sixteenth centuries, such as Amadis of Gaul and Palmerin of England? It is with these latter that Michael Moorcock, in his Wizardry and Wild Romance: A Study of Epic Fantasy, suggests “popular fantasy fiction” begins, and he bluntly lumps them together with Tolkien’s The Lord of the Rings and Howard’s Conan the Conqueror rather than with the Nibelungenleid and La Chanson de Roland and Malory’s Le Morte d’Arthur, which are “not fantasy fiction.” The former are distinguished from the latter, which

40

THE EVOLUTION OF MODERN FANTASY

Moorcock classes as “myth, legend, and folk-tale,” and marked by “definite authorship and not genuinely purporting to be . . . true account(s) of historical or religious events.” Further, Moorcock asserts that, unlike the latter, the likes of Amadis and Palmerin “were fantasies in that their chief purpose was to amaze and shock. They are packed with wizards, magic weapons, cloaks of invisibility . . .” (Moorcock 25–26).78 In the terms he sets up, Moorcock does make sense, and there is a certain validity to his basic point. At the same time, there are a variety of considerations he does not take into account. Amadis, which provided the functional template for the other romances Moorcock refers to,79 is known through the Spanish text of Garci Rodriguez de Montalvo, written over the decades preceding its author’s death in 1505 and published in 1508. But Montalvo’s work was itself a reworking of an earlier Amadis, and scholarly opinion seems to situate a proto-Amadis in the mid-fourteenth century. In other words, the degree to which one might speak of “definite authorship” in relation to Amadis of Gaul as we have it is limited. Like Malory, Montalvo worked with sources. Though it is quite possible that the lost original (as is supposed) was a pure invention on the part of its author, it was developed by a series of hands over a century and a half, and it is apt to say that Montalvo (again, like Malory) is more a redactor than a “definite author.”80 In other words, the Amadis cycle as a literary entity evolved in much the same way as the stories attached to the “mythic” or “legendary” cycles of medieval romance, regardless of whether it was, in its written inception, invented. Lastly, with regard to style, Amadis as we have it is fully imbued with medieval romance narrative conventions81—conventions that, while beginning to be a bit old fashioned by Montalvo’s time, were still sufficiently part of the literary landscape to spur a flood of derivative texts throughout the sixteenth century (Place Preface). In terms of dissemination, development, and style, then, Amadis of Gaul is more aptly seen as signaling the last flowering of medieval romance than inaugurating the modern fantasy genre. And while the multitude of sequels and spin-offs,82 coinciding with the introduction of the printing press, may be seen as the sensational inventions of “definite authors,” their immediate debt to Amadis, as well as to the body of romance it drew on, tie them far more strongly to what immediately preceded them than to what followed several centuries later. A look at the subsequent history of Amadis of Gaul, its sequels and spin-offs, serves to underscore this. There is no doubt that these late chivalric romances were enormously popular through the sixteenth and into the seventeenth centuries and were widely translated, especially into other romance languages. But the excesses of length and the proliferation of “new” works founded in a limited and increasingly strained narrative

INTRODUCTION

41

vocabulary combined to create something of a surfeit by the early seventeenth century, and the vogue began to dwindle: it is perhaps indicative that the great work of seventeenth-century Spanish literature, Cervantes’s Don Quixote, is a satire of the sensibility associated with these romances. The vogue for Amadis and the chivalric romance stretched to England, though it was primarily reliant on French translations, and translations of Amadis into English did not appear until the late sixteenth and early seventeenth centuries,83 when their popularity had begun to decline. The influence of Amadis and the others can be seen to some degree in Spenser’s The Faerie Queene and in Sidney’s Arcadia, though it cannot really be seen as a model for either work,84 and that influence quickly becomes sparer from the early seventeenth century on.85 The English translations did not become perennial, oft-reprinted staples of literate reading, and though Amadis remained well-known enough to inspire an opera by Handel in 1715, it was clearly known more by reputation than by close textual familiarity.86 It was only with the editions of Southey in the opening decade of the nineteenth century that Amadis and Palmerin again had some currency in English in something approximating their original form. To return to Michael Moorcock’s contention, not only are Amadis and Palmerin, despite being inventions and despite the obvious sensationalizing of martial and magical content, far closer to the worlds of Arthurian and Carolingian romance as literary works than to J. R. R. Tolkien and Robert E. Howard,87 but it is difficult to ascribe to them much influence on the development of fantasy in English that is not indirect and at a remove. The imaginative sensibility of the eighteenth century that, together with the burgeoning of scientific-minded antiquarianism, initiated the continuum leading to William Morris and ultimately Tolkien was driven by, on the one hand, an often patriotic interest in “antiquities” indigenous to the British Isles and, on the other, a preoccupation with the exotic and the remote, catalyzed by the appearance of The Thousand and One Nights during the opening decades of the eighteenth century. Southey’s editions in the early nineteenth century did spur a new flurry of enthusiasm for Amadis and Palmerin, but it is difficult to discern an influence on either the poetic or the prose narratives of the period separable from generic romance influence.88 Nor did Southey’s editions become perennially available staples; by the early twentieth century, the two romances were available in English only in antiquarian volumes. Apart from Cabell, who borrowed some names from Amadis and whose Biography of Manuel no doubt found some inspiration in the Iberian material, there is little in the work of the pregenre canon that suggests any particular or direct debt at all. Amadis of Gaul and Palmerin of England are medieval chivalric romances, following the conventions of their day, not works built on an antiquarian

42

THE EVOLUTION OF MODERN FANTASY

knowledge of relics from a forgotten past crossed with a modern romantic sensibility. I would argue that these continuities with the literature of the world they were ultimately part of are more significant than the reasons Moorcock suggests for considering them the beginning of the modern fantasy genre. To turn to English literature, a rather better case could be made for Spenser and The Faerie Queene.89 Spenser is as much the poem’s “definite” author as Tolkien is the definite author of The Lord of the Rings, and while much of the content is traditional in rather a piecemeal sense, as much or more is invented, and the overall construct is very much Spenser’s invention. The poem is emphatically not the last form of a work that evolved from an exemplar preceding it by a couple centuries. At the same time, the romance narrative conventions that the poem may be said to have in common with modern fantasy were perhaps old fashioned by the late Elizabethan period, but the poem did not follow on or accompany an intellectual movement bent on recovering a lost past, nor was it founded in an aesthetic in which the remote and the miraculous were in themselves the central interest. Spenser’s Arthur, the legendary emperor from whom the House of Tudor claimed descent, was the byproduct of a very practical bit of political propaganda. The Faerie Queene is crisscrossed by a Byzantine web of often intractable allegory, sometimes historical, sometimes moral, and pervasive to a degree quite alien to modern fantasy. The traditional matter woven into the fictional construct includes elements derived from both the ancient epic and the Italian Renaissance epics of Boiardo and Ariosto, from classical myth and popular Christian legend (such as the tale of St. George and the Dragon), and from medieval English Arthurian romance.90 This was all part of the received tradition of Spenser’s time. Thomas Warton, in 1754, would implicitly typify the immensely complex and multilayered poem by saying of its author, “It was his business to engage the fancy, and to interest the attention by bold and striking images” (Woods 79). While The Faerie Queene may in fact be seen to do these things, it is doubtful that Spenser would have described his poetic goals in such terms. There is a large distance between Spenser and the post-Augustan eighteenth century, and The Faerie Queene did not emerge from the same aesthetic world as Warton’s “The Grave of King Arthur” or Ossian.91 There was a wealth of romance narrative aside from The Faerie Queene that appeared in the late sixteenth and early seventeenth centuries, from poetic treatments of Greek myths92 to prose romances indebted to the (new to England) ancient Greek works of Heliodorus, Longus, and others,93 but from Sidney and Lyly forward, the romance trappings became increasingly a frame within which to explore political philosophy. Closer to the age of Warton, we have Bunyan’s The Pilgrim’s Progress (1667) and Swift’s

INTRODUCTION

43

Gulliver’s Travels (1726). But Bunyan’s work is a determined allegory, and Swift’s is a very specific political satire couched as a traveler’s tale. While the work I will be discussing here, from the eighteenth century on, often contains allegory or satire, individual works are not thorough allegories or satires. And while political philosophy may be explored, the works are not disguised essays on political philosophy. The choice to develop archaistic modes of narrative in the eighteenth century and following was a predominantly aesthetic choice, though that choice might have inherently religious or political implications. This I would take to be the major point of transformation in the eighteenth century, the pertinent result of which, here, is a trajectory that leads directly to William Morris and J. R. R. Tolkien. As I have said, the focus will be on fantasy, as I have loosely delineated it, in British, Irish, and Anglo-American literature specifically. This choice is, foremost, motivated by the need for containment: to extend the discussion in any comprehensive way to the literatures of the continental European languages, much less to the literatures of other parts of the world, would make this study impossibly large and substantially remove the possibility of any cohesive (if loose) organizational frame related to the emphatically British/Irish/Anglo-American canon established by the BAFS. Of course, this tradition did not develop in a vacuum, and certainly (for example) the contes de fees and quasi-Oriental tales of late seventeenthand early eighteenth-century France, and the literary märchen of the German Romantics a century later, had both direct and indirect shaping effect on the pertinent areas of the British/Irish/Anglo-American literary traditions. For that reason, they will not be ignored here. But it is, at the same time, interesting that while work bearing a clear relation to British/ Irish/Anglo-American fantasy (with regard to both form and influence) appears in the continental European literatures, the kind of tradition that ultimately culminates in the establishment of a separable literary phenomena, a separable genre, does not. Perhaps this is due to the fact that “fantasy” was, on the continent, generally less taboo: for example, most of the “great” German writers right through the middle of the twentieth century wrote märchen—sometimes quite a few—at some point during their writing careers, with no apology and no “for children” qualification. However it may be, neither in Germany nor France nor elsewhere on the continent is there the kind of “countertradition” of romance/fairy-story/ fantasy, becoming, in the twentieth century, a body of work by authors largely ignored by the literary establishment(s). As we move outside of the European sphere, some of the crucial literaryhistorical circumstances that produced fantasy as it culminated in the BAFS evaporate. For instance, the traditional Laguna Pueblo stories, told in

44

THE EVOLUTION OF MODERN FANTASY

verse, which constitute a large portion of Leslie Marmon Silko’s Storyteller (1981), may be read as fantasy on the basis of their surface content, but these stories are Silko’s individual articulation of a living oral tradition she grew up in the midst of. Invented worlds, magic, and an aesthetic founded on the remoteness of lost ages are simply not concerns of the author. Likewise, the writings of the late Yoruba writer, Amos Tutuola, may also be read as fantasy on the basis of their surface content.94 But the aesthetic associated with modern fantasy writing is as strikingly absent as it is with Silko: he is quoted in Michael Thelwell’s 1984 introduction to The Palm-Wine Drinkard (1952), his first published work, saying his intention was to “tell of my ancestors and how they lived in their days” (Tutuola xvii). His work is an outgrowth of the oral Yoruba culture that, however beset by European colonialism, Tutuola grew up in. Neither writer chose their subject matter from the standpoint of the same circumstances that led Marion Zimmer Bradley to the Arthurian legends or O. R. Melling to the Ulster Cycle.95 Even in the context of a culture with a long written tradition, where a modern writer engages a work far older, in its first written form, than Beowulf or The Tain, the aesthetic associated with fantasy is largely absent. Of the subject of his Ramayana (1972), R. K. Narayan wrote, “It may sound hyperbolic, but I am prepared to state that almost every individual among the five hundred millions living in India is aware of the story of the Ramayana in some measure or other . . . The Ramayana pervades our cultural life in one form or another at all times . . . Everyone knows the story but loves to listen to it again” (Narayan xi). Narayan’s treatment of the Ramayana, in its cultural context of twentieth-century India, is more akin to the nonfantasy-related twentieth-century biblical adaptations of Kazantzakis and Thomas Mann than to Walton’s treatment of The Mabinogion. To attempt to adequately cover continental fantasy here, much less that of the rest of the world, would be hopelessly unwieldy, not just in terms of the bulk of material, but in terms of the questions of cultural and literary context that would need to be addressed. For these reasons, references and discussions of works from outside Britain, Ireland, and Englishspeaking North America will reflect the period, circumstances, and form in which they became accessible to English readers rather than the cultural background of the original texts. In other words, The Thousand and One Nights will be considered primarily as an eighteenth-century (and subsequent) literary phenomenon in English rather than as a medieval Arabic work.

INTRODUCTION

45

Works Cited Anderson, Poul. The Broken Sword. New York: Abelard-Schuman, 1954. Cabell, James Branch. Figures of Earth. New York: Robert M. McBride and Co., 1925. 1Carter, Lin. Imaginary Worlds. New York: Ballantine Books, 1973. 2Carter, Lin. “The Fresh, Scrubbed Morning World of William Morris,” in Morris, William. The Wood beyond the World. New York: Ballantine Books, 1969. 3Carter, Lin. “The Dubious Land,” in MacDonald, George. Evenor. Edited by Lin Carter. New York: Ballantine Books, 1972. 4Carter, Lin, ed. Dragons, Elves, and Heroes. New York: Ballantine Books, 1969. Clute, John, and John Grant, eds. The Encyclopedia of Fantasy. New York: St. Martin’s Press, 1997. 1de Camp, L. Sprague. Literary Swordsmen and Sorcerers: The Makers of Heroic Fantasy. Sauk City, WI: Arkham House, 1976. 2de Camp, L. Sprague, ed. The Spell of Seven. New York: Pyramid Books, 1965. Dunsany, Lord. Book of Wonder. Boston: John W. Luce and Co., 1912. Eddison, E. R. The Mezentian Gate. New York: Ballantine Books, 1969. Irwin, W. R. The Game of the Impossible: A Rhetoric of Fantasy. Urbana: U of Chicago P, 1976. Jackson, Rosemary. Fantasy: The Literature of Subversion. London: Methuen, 1981. 1Lewis, C. S. “Tolkien’s The Lord of the Rings,” in Lewis, C. S. On Stories and Other Essays on Literature. Edited by Walter Hooper. New York: Harcourt Brace Jovanovich, 1982. 2Lewis, C. S. That Hideous Strength. New York: Macmillan Publishing Co., 1946. 3Lewis, C. S. The Pilgrim’s Regress. Grand Rapids, MI: Wm. B. Eerdmans Publishing Co., 1958. MacDonald, George. “The Fantastic Imagination,” in MacDonald, George. A Dish of Orts. Whitethorn, CA: Johannesen, 1996. Manguel, Alberto, ed. Black Water: The Book of Fantastic Literature. New York: Clarkson N. Potter, 1983. 1Manlove, C. N. Modern Fantasy: Five Studies. New York: Cambridge UP, 1975. 2Manlove, C. N. The Impulse of Fantasy Literature. Kent, OH: Kent State UP, 1983. Matthews, Richard. Fantasy: The Liberation of Imagination. New York: Routledge, 2002. Moorcock, Michael. Wizardry and Wild Romance. Austin, TX: Monkey Brain Books, 2004. Morris, Kenneth. The Fates of the Princes of Dyfed. Point Loma, CA: Aryan Theosophical Press, 1914. 1Morris, William. The Sundering Flood. New York: Ballantine Books, 1973. 2Morris, William. The Well at the World’s End II. New York: Ballantine Books, 1970. Narayan, R. K. The Ramayana. New York: Penguin Books, 1977. Place, Edwin B., and Herbert C. Behm, ed. and tr. Amadis of Gaul Books I and II. Lexington, KY: U of Kentucky P, 1974. Pratt, Fletcher. The Well of the Unicorn. New York: William Sloane Associates, 1948. Shippey, Tom, ed. The Oxford Book of Fantasy Stories. New York: Oxford UP, 1994.

46

THE EVOLUTION OF MODERN FANTASY

Swinfen, Ann. In Defense of Fantasy. Boston: Routledge and Kegan Paul, 1984. Tolkien, J. R. R. “On Fairy-Stories,” in Tolkien, J. R. R. The Monsters and the Critics and Other Essays. Edited by Christopher Tolkien. Boston: Houghton Mifflin Co., 1984. Tutuola, Amos. The Palm-Wine Drinkard. New York: Grove Press, 1984. White, T. H. The Book of Merlyn. Austin: U of Texas P, 1977. Woods, George, ed. English Poetry and Prose of the Romantic Movement. New York: Scott, Foresman and Co., 1929.

2

The Eighteenth Century The Forgotten Past

The Canonical Narrative

T

he standard canonical narrative of eighteenth-century Englishlanguage literature during much of the twentieth century, which still figures as part of the lens through which the writing of that period is quantified, conveys little that would suggest important ties to fantasy as demarcated in this study. Poetry, according to that narrative, begins with Pope and the extreme formalism of the Augustans and closes with the appearance of Lyrical Ballads, which ushered in the Romantic period in 1798. The movement from one to the other involves a reaction against a poetry founded on objective rules, satiric in its nature, and tending to moralize, as well as an embracing of the notion of the poet as visionary and prophet. The resulting emphasis on the capacity of the poet, which set the poet apart from the nonpoet, transformed the individual’s making of poetry into a theme in its own right. The frequent preoccupation with the subjective, individual internal processes of the poet, which marks much work from the subsequent Romantic period, was a direct outgrowth of this. Prose fiction, through the work of Defoe, Fielding, and Richardson particularly, became marked by a content increasingly grounded in the familiar and the probable. The subsequent nineteenth-century work of Austen, Dickens, and Hardy descends from this. This summary is, of course, brutally terse. But even allowing for some reductionism,1 the key point is that the canonical narrative simply leaves a great deal of the eighteenth century out. An explanation for this is not difficult to find: the narrative itself solidified during the second quarter of the twentieth century, at a point when the major contemporary developments

48

THE EVOLUTION OF MODERN FANTASY

in literature and modernism were virtually synonymous. In constructing the literary historical narrative of the previous two to three centuries, what became most significant were those developments that seemed to anticipate modernism. I have said that this narrative conveys little that would connect to fantasy. However, there are some loose chinks in the armor. In poetry, there is William Blake, who began the twentieth century as a minor lyric poet but grew into a major figure, particularly with the studies of Northrup Frye and David Erdman in the 1940s and 1950s. But though Blake is generally grafted onto the Romantic period, most of his work precedes Lyrical Ballads, and what follows seems little affected by Romanticism. In fact, his work, and particularly his major work, the Prophetic Books, is more aptly seen as a culmination of currents in the “erased” eighteenth century. The Prophetic Books, ironically, tend to be skirted around by the canonical narrative; however, they embody elements that connect quite clearly to fantasy. In fiction, Samuel Johnson’s Rasselas is part of the standard narrative, and there is the sort of second-tiered presence of the Gothic novel inaugurated by Walpole’s Castle of Otranto. The former tends to be seen in isolation, and as a virtual “antifantasy,” it would seem of negligible significance to fantasy. But seen in the context of the aesthetic engendered by the appearance of The Arabian Nights Entertainments early in the century and the proliferation of quasi-Oriental fictions by both French and English writers that followed, Rasselas assumes a different dimension. Yet the quasiOriental fiction of the eighteenth century, significant in the development of fantasy, was largely erased in the formulation of the canonical narrative. The only other work in this mode given any recognition is Beckford’s Vathek, which has tended to be categorized as a Gothic. The Gothic, as I said, has a sort of second-tiered presence in the canonical narrative. Initiated by Otranto in 1765, the mode saw a huge outpouring during the last decade and a half of the century. But while it may be seen to represent a countercurrent to the novel, part of its validation has rested in the appropriation of indubitably “Gothic” elements by later canonical novelists, from Dickens to Faulkner. The Gothic has been often cited as a natural forerunner of fantasy, but I would argue that its significance to the BAFS mode that is the focus of this study is limited. Walpole’s Otranto, Matthew Lewis’s The Monk, and other works representing the “masculine” Gothic include supernatural elements and are often given medieval settings. However, the generally demonic quality of the supernatural links more strongly to the later horror genre than to the contents of the BAFS. The image of the Middle Ages is sensational and in fact reinforces then-stereotyped views of the period, founded in superstition, corruption, and decadence, and is close to the inverse of

THE EIGHTEENTH CENTURY

49

William Morris’s Middle Ages. The “feminine” Gothic, most definitively embodied in the works of Anne Radcliffe, has more in common with the BAFS mode with regard to the atmospheric evocations of lush, beautiful landscapes and the use of plot structures echoing chivalric romance, but actual supernatural elements are generally eschewed, and the time periods in which the stories are set tend to be closer to the authors’. Beyond these mild chinks, the canonical treatment of the eighteenth century, reflecting a perspective that sees the twentieth century primarily in terms of modernism, offers little to someone interested in the roots of fantasy. However, if one looks to the whole of eighteenth-century literature from the standpoint of the BAFS-centered genre, a different picture emerges: not all reaction against the Augustans points to modernism. For the balance of this chapter, I will discuss a number of pertinent developments. First, I will look to material external to the English-speaking world in two capacities: (1) a shift in the perceived relationship between the Greco-Roman classics and present poetic endeavor and (2) the effects of the influx of previously inaccessible literature from outside the European matrix, specifically The Arabian Nights. Second, I will look closer to home in two capacities: (1) a rethinking of the canon of English poetry and (2) the growing interest in then-inaccessible work reflecting an indigenous past. Last, I will look at the practice of syncretism, the attempt to link disparate traditions, on both a scholarly and a creative level. Revisions of the Classical Inheritance As a staple in educated culture, the Greco-Roman classics were, of course, not “forgotten.” What is pertinent here is the shift from a perspective that saw them serving a primarily instructional role for present poets, a necessary intermediary between the poet and the muse, to a perspective where the role was primarily inspirational, a door whose opening permits the poet direct access to the muse. With this shift came changes in the weight of emphasis given to individual writers, genres, and periods. A Roman poet such as the satirist Horace, so important to the Augustans, becomes eclipsed by a Greek lyric poet such as Pindar. This shift can be illustrated by a brief look at some poems from both sides of the divide whose subjects are the classical inheritance. Anticipating the Augustans, John Dryden’s “To Oedipus,”2 prologue to his adaptation of Sophocles’s tragedy (1679), sees Greece cast as lawgiver to “all the world beside.” Sophocles and Socrates stand at the top of an implied hierarchical order (“Supreme in wisdom one, and one in wit”), where poetry and philosophy are equivocated (“wit from wisdom

50

THE EVOLUTION OF MODERN FANTASY

differ’d not”) and situated symbolically at the center of law-giving Greece. Greece’s law-giving function extends to poets of all ages: “[E]very critic of each learned age, / By this just model [Sophocles’s play] has reform’d the stage.” The implied relationship is eminently practical, like that of teacher and pupil, the purpose to instruct in the proper methods of harnessing the poetic impulse (“With some respect to ancient wit proceed”), without which the “private spirit” becomes the sole point of reference and “[y]ou turn fanatics in your poetry.” Alexander Pope’s passage on Greece and Homer in “An Essay on Criticism” (1709)3 evinces a similar perspective. Greece plays the role of instructor, providing the poet with “useful rules” for determining “[w]hen to repress, when to indulge our flights.” Pope depicts Homer as a basis of “judgment” and “maxims”—an instructor providing the poet with the necessary intellectual equipment to then “trace the muses upward to their spring.” The instructional value ascribed by both Dryden and Pope to the Greek classics is also evident in their own poems: Pope’s is an “Essay” in verse; Dryden’s is expository. Both posit a distance between the poet and the muse, who are separated by mediating restrictions that are necessary if poetry is not to become “fanaticism.” Both, in fact, are more preoccupied with the mediating elements than with the poetic impulse itself and even seem to view the muse with some apprehension. Within a few decades, this would change. In Liberty (1735),4 James Thomson depicts Homer speaking in a “deep-searching voice” to which “heart,” “passion,” and the “varied soul” respond directly: Homer is not an intermediary, furnishing the poet with what is needed prior to approaching the muse, but the voice of the muse itself. Further, the poem itself is not an “essay.” Rather, it takes “the form of a Poetical Vision” in which the chief voice is of the Goddess of Liberty (Thomson 197) and hence assumes an oracular quality: the muse is invoked directly. In 1757, Thomas Gray published “The Progress of Poesy: A Pindaric Ode.”5 The poem traces the migration of “Poesy” from Greece to Italy to England. Unlike Dryden, Gray casts Greece not in the role of lawgiver but as receptor of poetic inspiration: Awake, Aeolian lyre, awake, And give to rapture all thy trembling strings, From Helicon’s harmonious springs A thousand rills their mazy progress take . . .

Poesy migrates from Greece to, ultimately, England, where the poetic spirit manifests itself anew: “Oh! Lyre divine, what daring Spirit / Wakes thee

THE EIGHTEENTH CENTURY

51

now?” While Dryden allies poetry with philosophy, Gray allies it to the lyre, to music: philosophy suggests instruction, music inspiration. One may extend this to suggest that Gray’s ode is “Pindaric” not in the sense that Gray slavishly followed a set of rules abstracted from Pindar’s practice (though Gray had studied Pindar closely) but in that Gray sought to manifest the same spirit of poesy evident in Pindar. Though the poems of Thomson and Gray are not narrative, they do evince a significant shift in the perception of received classical tradition that will underlie some of the poetic narrative fantasies of Keats, to which I will return in Chapter 3. They also begin to sketch, more generally, a relationship with “sources,” classical or otherwise, founded more on intuition than mastery of a fixed body of rules. Though Gray’s poem is informed by an intimate familiarity with Pindar (and others), the relationship implied is a sibling relationship, in which both ancient and modern speak with the voice of the same transcendent muse. This is echoed in the way “sources” inform William Morris, Eddison, and Tolkien. Quasi-Oriental Fictions While the classical inheritance was being reconceptualized, there was an influx of significant works from further afield than the Mediterranean, previously unknown in the West. The first, and most important, of these came by way of France. Antoine Galland’s Mille et une Nuit, translated from a fourteenth- or fifteenth-century manuscript of the Arabic Alf Layla Wa Layla, appeared in a series of volumes from 1704 to 1717.6 The popularity of Galland’s work spurred a host of original quasi-Oriental imitations by French writers. Translated anonymously from Galland’s French, The Arabian Nights Entertainments appeared in English from 1706 to 1721 and was followed by many of the imitation French tales in the succeeding decades, all of which proved quite popular. But despite the popularity of the translations from the French, the appearance of corollary work written in English was a bit halting during the first half of the century. Vignette-length imitations of the Nights surfaced, including some pieces written by Addison for The Spectator (1711– 14), but the sole extended quasi-Oriental work to appear prior to Rasselas in 1759 was Eliza Haywood’s The Adventures of Eovaai, Princess of Ijaveo (1736). This last, little-known work deserves attention here. That it is substantially cut from eighteenth-century quasi-Oriental cloth is readily evident: many of the magical elements, including angelic Genii, demonic Ypres, and flying carpets, derive directly from The Arabian Nights and the French

52

THE EVOLUTION OF MODERN FANTASY

imitations; themes such as the evil magician, Ochihatou, posing as a beneficent vizier are similarly common in the Nights and the imitations; and the romantic atmosphere echoes English and European perceptions of the Nights. But there is a bit of a remove: unlike both the French and subsequent English imitations, which are, following the Nights, conventionally set in Baghdad, Persia, Abyssinia, and so on, Eovaai is cast as a “pre-Adamical History” whose only explicit connection to the geographical East is that it is an English translation of a Chinese translation from a “[l]anguage spoken in those remote ages” (Haywood 41, 50). The setting of the story, the countries of Ijaveo, Hypotofa, and others, is in fact invented. This, coupled with the apparatus of notes and commentary to the supposed “msss.,” puts Eovaai directly in line with the later practice of Tolkien and Cabell: the case might be made that Haywood’s novel is the “first” invented-world fantasy novel. The specific satiric subtext perhaps undercuts this to some degree. This, as well as some of the passages on political philosophy, arguably connects Eovaai more to Swift’s Gulliver, Manley’s New Atalantis, and fictions back to Sidney’s Arcadia and Lyly’s Euphues than to later fantasy. However, the satire is less thorough than in Swift, and especially as the story wears on, the romance conventions employed by Haywood seem to themselves determine its direction. In a sense, Eovaai may be said to be on the cusp of earlier and later narrative forms: the fantasy is partly determined by political satire, but it also develops as a fiction on its own terms. But these peculiar features of Eovaai are largely absent in the other quasi-Oriental tales that followed. It was a couple decades before the next appeared, though a steady stream then proceeded for most of the rest of the century. Samuel Johnson’s History of Rasselas, Prince of Anyssinia appeared in 1759. Like Eovaai, Rasselas is a somewhat idiosyncratic exemplar of the genre, although for different reasons. A “philosophic tale,” Rasselas bears a stronger resemblance to those works of Voltaire given Eastern settings, such as Zadig (1747) and The Princess of Babylon (1749),7 than to work more specifically imitative of The Arabian Nights. Despite its Romantic Eastern atmosphere, and recourse to narrative conventions that characterize the genre, it is marked by a complete absence of any element of the miraculous: the only tentative move in that direction, the Astronomer’s purported control of the weather, is pointedly revealed as delusion. This absence, reinforced by the tale’s Enlightenment-allied philosophical underpinnings, renders Rasselas rather peripheral in the present context.8 It might aptly be termed an “antifantasy.” Some of the other less well-known works that appeared subsequently are more in keeping with the genre as a whole and simultaneously more pertinent to the development of fantasy. The first of these was John

THE EIGHTEENTH CENTURY

53

Hawkesworth’s Almoran and Hamet (1761). Despite some similarity in prose style to Johnson’s tale, Hawkesworth’s tale is quite distinct. When the apparently supernatural proves illusory, the function is as plot contrivance: the philosophical implications of Johnson’s are absent. The story unfolds with a much more structured reliance on romance narrative conventions than Johnson’s rather wandering tale. Unlike Haywood’s work, the story takes place in a conventional quasi-Oriental setting, Persia. James Ridley’s collection of shorter quasi-Oriental narratives, Tales of the Genii (1764), though pointedly moralistic, liberally employs magical elements akin to those of the Nights, as does Frances Sheridan’s History of Nourjahad (1767), set in Persia. Quasi-Oriental fictions continued to appear for the remainder of the century, but the most well-known is William Beckford’s Vathek, published in Samuel Henley’s English translation in 1786 and in Beckford’s original French in 1787. Something like the fiction attached to Eovaai’s “msss.” attended the publication of Henley’s English, though here it was literal: Beckford’s name was omitted, and the work was presented as a translation from Arabic. The hasty publication of the French original was clearly motivated by Beckford’s wish to establish his authorship and resulted in the omission of a series of “Episodes” that he had intended to publish with Vathek proper. Of the four intended episodes, two were completed (one in two versions), and a fragment of a third survives. Beckford never published them, and they were believed lost; it was not until 1912 that they were finally published as The Episodes of Vathek.9 Taken together, Vathek and the Episodes form the most extensive English10 quasi-Oriental work that had been attempted, and in many ways, it can be seen as the culmination of the form in the eighteenth century. Of its compeers, it has been the most influential on later fantasy writers, and its admirers include George Meredith, H. P. Lovecraft, Clark Ashton Smith (who wrote a conclusion to the fragmentary episode), and C. S. Lewis. It was included in the BAFS in 1971, in the first edition to contain both the tale and episodes. However, it is as a Gothic that Vathek has been most frequently considered, and this has often been reflected in its publication.11 Partly this reflects the general neglect of the eighteenth-century quasi-Oriental tale, as well as the fact that the Gothic has been a critically recognized genre, while fantasy has not. But it also reflects the luxurious decadence of Vathek’s atmosphere and the sensational horror of the story’s conclusion, in which Vathek, Nouronihar, and the narrators of the “Episodes” are plunged into eternal perdition. Its influence within the BAFScentered canon has been strongest on those writers of the popular strand (Lovecraft and Smith) connected to the horror genre.

54

THE EVOLUTION OF MODERN FANTASY

Nevertheless, Vathek and the Episodes stand more solidly in their context as quasi-Oriental fictions than as exemplars of the Gothic.12 That the former genre stands as initiator of the quasi-Oriental strand in the subsequent fantasy tradition, running from Meredith’s The Shaving of Shagpat (1855) to Susan Shwartz’s post-BAFS anthology Arabesques: More Tales of the Arabian Nights (1988) and beyond, should be self-evident. The republication of Vathek in the BAFS arguably placed Beckford’s work in a more appropriate setting than its frequent “Gothic” editions. The currents driving the development of the quasi-Oriental tale are related to those undergirding the reconceptualizing of classical inheritance. Both evince a marked restlessness, and both, if in slightly different ways, appeal to the imagination in ways that run against Restoration and Augustan ideals. But whereas the latter was built on texts well established in educated English culture and was felt in more “highbrow” circles, the former drew its impetus from a text that was new to English readers, was itself a compendium of popular narrative, and was felt more in the domain of popular culture. Thomson’s and Gray’s appeals to inspiration and the imagination are quite explicit. The imagination in the quasi-Oriental tales is largely implicit, though perhaps reflecting a broader restlessness. Robert Mack, editor of Oriental Tales (1992), which includes the aforementioned works of Hawkesworth and Sheridan, writes that “the oriental tale . . . is a place to be free of the restrictions of mundane realism . . . The spectacular effects of its ‘enchanting’ conventions . . . are themselves the symptoms of a larger freedom at work . . . the freedom of explorations of alternative possibilities” (Mack xvii). In 1692, in the preface to his prose fiction Incognita, William Congreve described the shortcomings of “Romance” as against the virtues of “Novels”13 thus: [Of Romance] . . . lofty language, miraculous contingencies, and impossible performances elevate and surprise the reader into a giddy delight, which leaves him flat upon the ground whenever he gives off, and vexes him . . . when he is forced to be . . . convinced that ‘tis all a lie . . . Novels are of a more familiar nature; come near to us, and represent to us intrigues in practice; delight us with accidents and odd events, but not such as are wholly unusual or unprecedented—such which, not being so distant from our belief, bring also the pleasure nearer us. (Salzman 474)

With the quasi-Oriental tales, Congreve’s judgment has been implicitly reversed.

THE EIGHTEENTH CENTURY

55

Revisions in the English Poetic Tradition Paralleling the rethinking of classical tradition, the tradition of English poetry met with a noteworthy reorientation. Particularly significant here was the reevaluation and rehabilitation of Edmund Spenser’s The Faerie Queene. Though Spenser, partly due to his importance to Milton, had maintained the status of “major poet” throughout the later seventeenth and early eighteenth centuries, The Faerie Queene, his major work, was viewed with disapproval. The multileveled and often obscure allegory, the iconography, and the recourse to narrative modes and verse forms derived from Italian poets such as Ariosto and from medieval romance were alien to the French-imbued classicism that informed the poetic theory and practice of poets like Dryden and Pope. FQ was considered a highly inappropriate model for practicing poets. But as the eighteenth century progressed, this changed. In Observations on the Fairy Queen of Spenser (1754), Thomas Warton challenged the negative Augustan evaluations: “It is absurd to think of judging . . . Spenser by precepts which [he] did not attend to . . . we require the same order and design which every modern performance is expected to have, in poems where they were never regarded or intended . . . It was his business to engage the fancy, and to interest the attention by bold and striking images . . . The various and the marvellous were the chief sources of delight . . . [FQ] engages the affections, the feelings of the heart, rather than the cold approbation of the head” (Warton 79). The sensibility evident here echoes that of Thomson and Gray in relation to classical tradition. Evidence of Spenser’s new significance is widespread in post-Augustan eighteenth-century poetry. William Collins opens his “Ode on the Poetical Character” (1746) with praise to Spenser: At once, if not with light regard, I read aright that gifted Bard, (Him whose school above the rest His Loveliest Elfin Queen has blest.) (Lonsdale 144)

Similar allusions to, odes to, and imitations of Spenser litter the poetry of the period, though his effect on narrative is rather thin. Three narrative poems are cast in the nine-line Spenserian stanza: Thomson’s The Castle of Indolence (1748), Shenstone’s “The Schoolmistress” (1737), and Beattie’s The Minstrel (1771). But of these, Shenstone’s and Beattie’s borrow only Spenser’s stanza form. The content of “The Schoolmistress” is mundane, with no recourse to the marvelous; in The Minstrel, the marvelous exists

56

THE EVOLUTION OF MODERN FANTASY

primarily in the developing poetic imagination of its protagonist and does not become a direct element of the narrative itself. Neither echoes the narrative textures of the FQ. Thomson’s Castle of Indolence, however, does follow Spenser in this regard. The deceptively pleasing setting conjured by the wizard Indolence recalls the Bower of Bliss in Book Two of the FQ: A pleasing land of drowsyhed it was, Of dreams that wave before the half-shut eye; And of gay castles in the clouds that pass, Forever flushing round a summer-sky: There eke the soft delights, that witchingly Instill a wanton sweetness through the breast; And the calm pleasures always hovered nigh; But whate’er smacked of noyance or unrest, Was far, far off expelled from this delicious nest. The landskip such, inspiring perfect ease; Where INDOLENCE (for so the wizard hight) Close-hid his castle mid embowering trees, That half shut out the beams of Phoebus bright, And made a kind of checkered day and night. Meanwhile, unceasing at the massy gate, Beneath a spacious palm, the wicked wight Was placed; and to his lute, of cruel fate And labour harsh complained, lamenting man’s estate. (Thomson 255)

This enchanted setting, the narrative matter (in which pilgrims are drawn to the Castle through the machinations of the eponymous wizard until his defeat by the Knight of Arts and Industry), and the characters (including wizard, knight, and demons) have in common with Spenser the same elements Spenser has in common with modern fantasy. At the same time, with its underlying threads of moral and historical allegory, Thomson’s poem also embodies those elements in Spenser that do not generally figure in modern fantasy. Curiously, Thomson’s relationship to Spenser carries some of the Augustan-hued teacher–pupil dynamic, and Thomson even seems to maintain some of the more Augustan judgments of the FQ: “This poem being written in the manner of Spenser, the obsolete words, and a simplicity of diction in some of the lines, which borders on the ludicrous, were necessary to make the imitation more perfect” (Thomson 252). Nevertheless, the establishment of Spenser, a poet associated with fancy and the marvelous, as model and inspiration is significant.

THE EIGHTEENTH CENTURY

57

But the rehabilitation of Spenser is perhaps more important for the interest it spurred in what lay behind him. In one direction, this led to Ariosto and the Renaissance epic, as well as Italian tradition generally. In a note to “The Progress of Poesy,” Gray would link Spenser, Chaucer, and Milton to the Italian “school [which] expired soon after the Restoration,” replaced by a new school “on the French model, which has subsisted ever since” (Lonsdale 49). While Gray does not explicitly pass judgment on the French school, it is notable that France is not visited by Poesy in the ode, and Pope is not mentioned. The desire to realign poetry with the Italian school during the post-Augustan eighteenth century was an important factor in the reconfiguration of the earlier English poetic tradition, and Spenser’s association with earlier Italian poetry was considered significant. But also behind Spenser, embodied in the vocabulary of chivalric romance he shared and partly derived from Ariosto, was the heritage of the medieval world. Though Spenser borrowed less overtly from medieval English literature than from Ariosto, he was clearly familiar with it. His choice of Arthur as the central figure for the FQ, however little he based him on actual medieval sources such as Malory, combined with his romancebased narrative structures and marvelous content to connect him to the world of medieval poetry and Arthurian legend. However, by the eighteenth century, the world of actual medieval chivalric romance had receded to unknown territory: few Middle English romances had made the transition from manuscript to print, and those that had were, for the most part, long out of print. The medieval form of the Arthurian story had been eclipsed by Spenser and the realignment of Arthur, during the Tudor period, with the quasi-historical emperor derived from Geoffrey of Monmouth. By the eighteenth century, abetted by the notorious “epics” of Sir Richard Blackmore, Prince Arthur (1695) and King Arthur (1697), the Arthurian milieu had become an apt target for satire and burlesque, as seen in Swift’s “A Famous Prediction of Merlin” (1709) and Fielding’s The Tragedy of Tragedies (1731).14 The last edition of Malory, the chief medieval Arthurian source to reach print, had appeared, heavily bowdlerized, in 1634. Like the Celtic and Germanic traditions, if less dramatically so, medieval Arthurian tradition was, by the mid-eighteenth century, part of a vanished past, and I will wait till the next section to discuss it further. Here, I will restrict myself to the figure of Chaucer, whose poems were the major works in Middle English available at the time. Like Spenser, Chaucer was considered a “major” poet, but as Thomas Warton would note with disapproval, he was “regarded rather as an old, than as a good, poet.” Warton then put his finger on the key stumbling block: “It is true that his . . . unfamiliar language disgusts and deters many readers . . . [but] when translation [for example, Dryden’s “Tales From

58

THE EVOLUTION OF MODERN FANTASY

Chaucer”] . . . becomes substituted as the means of attaining knowledge of any difficult and ancient author, the original not only begins to be neglected and excluded as less easy, but also despised as less ornamental and elegant” (Warton 79). If “obsolete words” and “simplicity of diction” were stumbling blocks to Spenser, then Chaucer, separated from the eighteenth century by major changes in the language, augmented by the introduction of the printing press and increasing regularization, was close to inaccessible, even to antiquarian specialists.15 Warton countered the negative evaluations of Chaucer with sentiments akin to those with which he defended Spenser: “His old manners, his romantic arguments . . . his simplicity and antiquity of expression, transport us into some fairy region, and are all highly pleasing to the imagination” (Warton 79). In the present context, this reconfiguring of the English poetic tradition is significant primarily due to the sensibility attached to it. While FQ is often mentioned in the context of fantasy,16 direct effect on works associated with the BAFS are few and superficial in substance.17 Chaucer’s work on the whole, rarely veering to the marvelous, has little direct connection to the modern genre. The “Italian School” with which Gray connected Spenser and Chaucer, even Ariosto’s Orlando Furioso, has had little direct effect on modern fantasy.18 But the perspectives voiced by Thomas Warton in defense of both English poets evince a sensibility that anticipates fantasy. Thomas’s brother Joseph Warton, in a more general and polemical theoretical statement attached to his Odes on Various Subjects (1746), declares “Invention and Imagination to be the chief faculties of a Poet” and that his collection was “an attempt to bring back Poetry into its right channel” (Lonsdale 137). This implies a rereading of the received tradition, whether classical or English, in sharp contrast to the Augustans. When Richard Mathews declares in his Fantasy: The Liberation of Imagination that the “maps and contours of fantasy are circumscribed only by the imagination itself ” and it “consciously breaks free from mundane reality” (Mathews 1–2), he is echoing, whether consciously or not, the sentiments of the Wartons. The Forgotten Past and Lost Antiquities Here I will discuss where Middle English romance (particularly in its Arthurian branch), narrative traditions deriving from Germanic/Scandinavian antiquity, and those deriving from Celtic antiquity stood in the eighteenth century. What all three shared was inaccessibility, though to varying degrees.

THE EIGHTEENTH CENTURY

59

In the first case, as mentioned before, much had never made its way into print. At the time of the introduction of the printing press, the metrical romance had been out of vogue for nearly a century, and what did reach print were primarily adaptations of French prose romances. These included the more fantastic Carolingian romances, such as Four Sonnes of Aymon (adapted and printed by William Caxton, c.1489) and Huon of Bordeaux (adapted by John Bourchier and printed by Caxton’s successor, Wynkyn de Worde, c.1534), and, most famously, Malory’s Le Morte d’Arthur (edited and printed by Caxton, 1485), integrating both French and English sources. The vogue for these waned as the sixteenth century wore on, and by the eighteenth century, they were available only as rare antiquarian volumes. The continental popularity of the Iberian romances Amadis of Gaul and Palmerin of England, both translated into English in the late sixteenth century, spread to England and prolonged enthusiasm for romance founded in medieval conventions for a while, but the last publication of Amadis was in 1702, in a version compressed to a fraction of the (massive) length of the original. By 1750, access to medieval romance was difficult, if not impossible, to the average reader and largely restricted to antiquarian scholars researching old manuscripts, who, encountering a linguistic barrier that had barely begun to be surmounted, would have had at best an imperfect grasp of what they were reading.19 This meant that the medieval Arthurian story we now take for granted was imperfectly known by experts and by the general reader not at all. Though some of the contours of that story were suggested, if in terse, undeveloped form, by the six Arthurian ballads included in Thomas Percy’s Reliques of Ancient English Poetry (1765), editions of actual medieval Arthurian texts would not begin to appear until the Romantic period. Nevertheless, the figure of Arthur, in light of the shifting sensibility of the time, began to shed his association with the ridiculous and assume some of his former grandeur. But the Arthur that Thomas Warton depicts in his “The Grave of King Arthur” (1777) owes more to the quasi-historical emperor than to the courtly king of Malory; his sonnet “On King Arthur’s Round Table at Winchester” (1777) concludes, “Yet Spenser’s page, that chants in verse sublime / Those chiefs, shall live, unconscious of decay” (Warton 78). That the Arthur, and general medieval aesthetic, of the eighteenth century came heavily colored by Spenser reflects the fact, again, that actual medieval texts were still primarily the provenance of antiquarian scholars. While medieval romance waited until the turn of the century for editions, the related form of the traditional ballad—often embodying similar plot structures and subjects, but simpler, more compressed, and usually committed to writing at a later date—began to find its way into print. The antiquarian interest in such material was primarily historical and linguistic

60

THE EVOLUTION OF MODERN FANTASY

rather than aesthetic, and very often scholars were critical of the material that preoccupied them. Percy’s aforementioned Reliques is the most famous of the various collections of ballads published during this period, and his editorial commentary evinces the Enlightenment-colored judgments many scholars had of their subject. Percy’s stated rationale for his choice of selections was that “such specimens of ancient poetry have been selected, as either show the gradation of our language, exhibit the progress of popular opinions, display the peculiar manners and customs of former ages, or throw light on our earlier classical poets” (Percy 8). As far as actual intrinsic poetic merit, Percy notes the “rudeness” of the “artless productions” in his collection, distinguishes them from poetry “of a higher class,” and writes, “In a polished age, like the present . . . many of these reliques of antiquity will require great allowances to be made for them” (8–9). Since his “object was to please both the judicious antiquary, and the reader of taste” (11), Percy freely adapted, edited, and rewrote his materials: the “taste” of cultured readers (including, presumably, Percy himself) in fact demanded editorial tampering in order to make the “specimens” even marginally successful as poetry. Despite the more romantic interest in such material from “former ages,” these types of Enlightenment-bound judgments20 would collude with the lack of access to texts to prevent any widespread, sophisticated absorption of medieval romance conventions into present poetry for some time. Nevertheless, the seed was planted: Thomas Warton, though he anticipates the subject eliciting “disgust” from critics and seeks to “plead their pardon” for discussing it, writes in praise of the chivalric romance: “Such are their terrible Graces of magic and enchantment, so magnificently marvellous are their fictions and fablings, that they contribute, in a wonderful degree, to rouse and invigorate all the powers of imagination; to store the fancy with those sublime and alarming images which poetry best delights to display” (Warton 80). The vernacular English traditions, and the figure of Arthur, of course, carried a patriotic dimension as well—not, as in Spenser’s time, by validating a dynasty’s claim to the throne, but by potentially recentering the poetic impulse on a uniquely indigenous heritage. It is not surprising that the revival of interest in Arthur was accompanied by an interest in even older traditions: those that we would now broadly distinguish as “Germanic” and “Celtic” but that in the eighteenth century were rather indiscriminately designated “Gothic.” After all, Arthur, particularly in the chronicle tradition that had displaced the medieval courtly romance tradition by the mid-seventeenth century, was British, and hence, as we would say, “Celtic.” In such a context, such a literature would become a sort of Holy Grail.

THE EIGHTEENTH CENTURY

61

While the antiquarian belief that such a body of literature might still survive was correct, other than the general notion that it would exist in old manuscripts, there was no focused sense of how to go about finding it, what it would be like, or what exactly to make of it once it was found. The language issue would prove an even greater stumbling block here than with Middle English. If specimens of medieval English literature were sparse in 1750, specimens of “Gothic” literature were virtually nonexistent. This can be explained fairly simply. Of the ancient cultures centered in lower Britain, what had become England and Wales, the “British” had been overrun, absorbed, or marginalized by successive waves of AngloSaxon migration soon after the withdrawal of Rome.21 The primary center of post-Anglo-Saxon “British” culture became Wales—the descendant of British language and tradition, Welsh. Ironically, the actual history of the descendants of “Arthur’s Britons” in relation to England is one of cultural and political subjugation and marginalization. The Welsh language was (and is) little known outside of Wales; the conventional wisdom of the educated characterized the Welsh people as rude and uncivilized. There was an enormous gulf between the world of educated English society and traditional Wales, past or present. The Anglo-Saxons themselves had been overrun, absorbed, or marginalized in the wake of the Norman invasion of 1066, and between the time of the invasion and, say, the time of Chaucer, the English language had changed beyond recognition. Much of the learning of the pre-Norman age had been destroyed by the Normans; the Saxons themselves were seen as savage and ignorant by their conquerors. Of the other adjacent Germanic and Celtic cultures, most pertinently those of Iceland and Ireland, where the greatest repositories of relevant “antiquities” resided, the linguistic difficulties and cultural prejudices were even more pronounced, and they were geographically remote from England. Bits and pieces of this material began to appear in English during the same period as Percy’s Reliques: Evan Evans published a small number of texts and translations in Some Specimens of the Poetry of the Antient Welsh Bards (1764); Percy’s Northern Antiquities of 1770 contained translations from Snorri Sturluson’s Prose Edda; Gray published English translations of Latin translations of two short Icelandic poems and one Welsh poem in 1768. These fragments and excerpts were brief, perhaps more tantalizing than fulfilling the interest that led to their publication. Nevertheless, the seed of interest was planted, and the romantic idea of such material began to attach to poetry, though most often this would manifest in specious references to druids, Stonehenge, and bards. In this regard, Thomas Gray’s second Pindaric Ode, “The Bard” (1757), is interesting. Unlike “The Progress of Poesy,” with which it was published, “The Bard” does not chart the migration of the poetic muse from its origins

62

THE EVOLUTION OF MODERN FANTASY

in ancient Greece but founds that muse in the Welsh bardic tradition. Founded on a tradition concerning the suppression of the bards during Edward I’s subjugation of Wales in 1283, the poem is framed as a prophecy spoken alternately by the last bard and the ghosts of his slaughtered peers, foretelling the ruin of the Plantagenet line and its displacement by the “Welch” House of Tudor. Gray writes of Elizabeth’s reign: What strings symphonious tremble in the air, What strains of vocal transport round her play! Hear from the grave, great Taliessin, hear; They breathe a soul to animate thy clay. (Lonsdale 58)

“The Bard” turns the face of the Elizabethan poetic muse not to Greece but to Britain’s descendant, Wales: it is a reawakening of “great Taliessin,” a sixth-century Welsh bard, not of a Greek lyre. The new enthusiasm for the idea of indigenous narrative traditions as a basis for poetry also turned poets to the oral traditions of the nonliterate, rural classes, assumed to be connected to ancient narrative traditions. For example, the current fairy lore of unlettered tradition would relate to the “ancient” fairy lore employed, and validated, by Spenser and Shakespeare. This conviction is evident in William Collins’s “An Ode on the Popular Superstitions of the Highlands of Scotland, Considered as a Subject of Poetry” (1749–50): Ev’n yet preserv’d how often may’st thou hear, Where to the Boreal Mountains run, Taught by the Father to his list’ning Son Strange lays whose pow’r had charm’d a Spenser’s Ear. (Lonsdale 169)

All this together still yields little in the way of narrative, again primarily due to lack of texts to serve as models. However, the desire for “Gothic” antiquities, and the idea of a link between contemporary oral and ancient narrative traditions, merged during the 1760s in a sensational literary phenomenon of crucial significance here. In 1760, there appeared anonymously Fragments of Ancient Poetry collected in the Highlands of Scotland, and translated from the Gaelic or Erse language, including 15 pieces. Fingal: An Ancient Epic Poem followed in 1761, as well as another “epic” in 1763, Temora. In these latter, the “translator” was identified as James MacPherson, and to “both works, MacPherson appended essays on the methods of his research and assured his readers of the poems’ importance as social history” (deGatno 5). Collected as The Works of Ossian in 1765, the poems elicited great enthusiasm: the “author,”

THE EIGHTEENTH CENTURY

63

the third-century Gaelic poet Ossian, was widely praised as the “Homer of Scotland” and sometimes ranked higher than his Greek counterpart; his fame reached the continent, where Voltaire, Goethe, and Schiller praised him, and the poems were translated into nearly a dozen languages. Ossian was exactly what was needed to establish a uniquely indigenous literary heritage. The enthusiasm, however, was undercut by doubts from some quarters—most famously Samuel Johnson—regarding the poems’ authenticity, compounded by MacPherson’s inability to produce the originals he had “translated.” Now consensus supports MacPherson’s critics, more or less. The cycle of poems was not pure invention: many of the characters are attested to in both Irish and Scottish Highlands Gaelic traditions, both in ancient texts dating to nearly a millennia earlier and in oral narratives collected throughout the nineteenth century, though the name forms may vary slightly (Finn MacCumhal as Fingal, Oisin as Ossian, etc.). Behind MacPherson’s poems can often be discerned the hazy outlines of genuinely traditional stories. MacPherson did work with some genuinely traditional material. But as Paul J. deGatno writes, “To say he embroidered and distorted would only begin to acknowledge his irresponsibility in the matter” (deGatno 19). The poems as they stand are no more “translations” of thirdcentury Gaelic poems than Book of the Three Dragons is a translation of The Mabinogion. But it is precisely the “inauthenticity” of Ossian—an inauthentic text weaving in transformed material from authentic tradition—which makes him important in the present context. In order to make his hoax credible, his matter, style, and presentation had to accord with eighteenth-century notions of what such poetry would be like. For matter, actual Gaelic tradition furnished him with “Homeric” warrior heroes, and the sometimes martial, sometimes sentimental narrative frames accorded with such characters. The lofty, diffuse, elegiac style also suited eighteenth-century preconceptions. The appended essays spoke to the presumed context of the poems’ composition. In all three of these areas, MacPherson was doing very close to what many—both Morrises, Cabell, Tolkien—of the pregenre literary fantasy writers were doing. Tolkien built his initial fictional indigenous English mythology out of chosen bits of Germanic and Celtic (and other) tradition and a great deal of his own invention. The content and style of that “mythology” was deliberately intended to evoke an impression of the authentically traditional. The essays—whether independent pieces on language or the appendices affixed to LOTR—further underscore the intended effect of “authenticity.”

64

THE EVOLUTION OF MODERN FANTASY

The major difference is that Tolkien did not actually attempt to deceive his readers: the air of authenticity is clearly part of the fiction. One suspects that Tolkien had more scruples than MacPherson, but it must also be remembered that in the 1760s there was no solid comparative basis for evaluating MacPherson’s claims, which, combined with a broad predisposition to believe the claims, rendered the fraud less immediately obvious. Even if he had been an unscrupulous man, Tolkien would not have been able to pull a similar hoax off. But beyond MacPherson’s extratextual claims, there is not a great deal of difference between what he was doing and what the pregenre literary writers were doing. In a context where ancient Gaelic narrative traditions were largely inaccessible, MacPherson in effect made a tradition up. While, with the qualified exception of Tolkien and to some degree Kenneth Morris, there was not an issue of actual inaccessible tradition to contend with come the twentieth century, the “making up” part is virtually identical. The later 1760s saw another hoax, similar in nature if less spectacular, that deserves mention here. The “Rowley Poems” of Thomas Chatterton appeared piecemeal in 1768 and following, purporting to be transcriptions of work by poets such as Thomas Rowley, William Canynge, and others from fifteenth-century manuscripts that Chatterton had in his possession. One of the most lengthy poems, “The Battle of Hastings,” was attributed to Turgot, a tenth-century Saxon monk, whose original Rowley had translated. The poetry was flanked by more historical matter, such as an account of the opening of the old bridge at Bristol, which Chatterton contributed to the Bristol Journal a week after the opening of the new bridge in 1768; Chatterton’s “researches” also yielded flattering pedigrees for his benefactors. The material, both poetic and “historical,” was made up out of whole cloth: the authors were fictional, and they did not even have an Ossianic degree of tradition behind them. In terms of content, the poems are not of particular interest here: they are largely naturalistic and “historical,” and the “marvelous” elements of romance are strikingly absent. However, Chatterton’s invention of an ersatz Middle English corresponding to the supposed language of Rowley and his compeers— something that MacPherson, as “translator,” did not have to do—is worthy of note. That many, even some “experts,”22 believed the poems to be authentic attests, as with Ossian, to the lack of a viable comparative framework for evaluating the claims. Chatterton, of course, was not the first to deliberately cultivate an archaic English—Spenser had done so nearly two centuries earlier. But in attempting to pass his creations off as genuinely archaic in themselves, Chatterton’s experiment with language became far more involved and thorough than any before him, and it stands as a clear

THE EIGHTEENTH CENTURY

65

precursor to the artificial, archaic prose style of William Morris and E. R. Eddison, as well as the elaborate invented languages of Tolkien.23 Syncretism and William Blake’s Mythology The influx of material from the East during the eighteenth century was not restricted to The Arabian Nights, nor were scholars content to rely on French translations. The Qur’an appeared in a translation from the Arabic by George Sale in 1734; later in the century, the studies of Sanskrit scholars yielded Charles Wilkins’s translation of the Bhagvat Geeta (1785), William Jones’s of Kalidasa’s Sacontala (1789), and others. This, combined with the work of antiquarians with Celtic and Germanic/Scandinavian languages and traditions, laid the groundwork for a more relative perspective on the inherited classical and biblical traditions and the beginnings of syncretistic comparative studies. Given the newness of the ideas and the limited expertise in processing limited source material, it is not surprising that some of the major theories that arose were nearly fantasies themselves. Jacob Bryant’s voluminous and popular Antient Mythology (1774–76) sought to establish the roots of mythology in a pre-Babel “Amonian” culture; earlier, in the 1740s, William Stukeley had linked this early protoculture to the druids. But while Northrup Frye aptly characterized Bryant’s work as a “mausoleum of misinformation and bad etymology” (Frye 173), his work stands, despite its lack of credibility, as a precursor of Max Muller and James Frazer in the nineteenth century and Mircea Eliade and Joseph Campbell in the twentieth. Like the studies in earlier English and “Gothic” antiquities, these new theories found counterparts in imaginative work, and I will close this section with a brief discussion of William Blake. The notion of a single world mythology and religion underscores the mythology embodied in Blake’s Prophetic Books; Blake would declare that all ancient mythologies “are the same thing, as Jacob Bryant and all antiquaries have proved” (Blake 578). His mythology embodies elements drawn from the Bible; from classical mythology; from the bits and pieces at his disposal from Germanic/Scandinavian and Celtic mythology; from the Bhagavad Gita;24 and from Milton, Spenser, Ossian, and more, all woven into his own highly unique construct. While he would cite antiquarians in assertions of the unity of mythology and religion, he saw the idea primarily in terms of its metaphysical implications rather than as a reductive, deterministic, supposedly objective framework. His remarks concerning his (lost) painting, “A Vision of the Last Judgment,” suggest something of

66

THE EVOLUTION OF MODERN FANTASY

his view: “I have represented it as I saw it; to different People it appears differently as everything else does” (Blake 605). Beyond this, Blake’s work finds itself at the heart of the developments this chapter has been concerned with. Even in his earliest work, the Poetical Sketches (1769–78), Blake draws on Spenserian and traditional ballad forms, Germanic traditional subject matter, and develops the Ossianic voice that would inform the Prophetic Books. In the increasing syncretism of the latter, these would be sublimated to Blake’s own unique vision. To pursue Blake at length would take us far beyond the focus of this study, and the present context does suggest a constraining view of his work. Of later fantasy writers, only George MacDonald and David Lindsay— both somewhat peripheral to the template derived from Morris, Eddison, and Tolkien—venture into similar metaphysical territory. The Prophetic Books taken as a whole, though narrative, tend to emphasize the dramatic interactions of characters and develop representational descriptive narrative very little. At the same time, the structure of the mythology, while drawing syncretistically on a wide variety of sources, is very much Blake’s own invention. It is peopled with original mythic figures, including Urizen, Orc, Enitharmon, and many others, who move in invented settings, such as the City of Golgonooza, Ulro, and Beulah. The Prophetic Books embody a clear precursor to the “invented mythologies” of Dunsany and Tolkien, as well as the most extensive, sophisticated mythology prior to the latter’s multiple castings of the narrative matter of the “Silmarillion.” Conclusion As the foregoing suggests, we can see the basic characteristics that inform the literary writers whose work comprised the bulk of the BAFS emerging throughout the course of the eighteenth century, albeit in somewhat piecemeal fashion. Both the classical and English poetic traditions were rethought, with an emphasis on invention and imagination that contrasted with Restoration and Augustan ideals. The revisions in the English poetic tradition can be seen, particularly, in the rehabilitation of Spenser’s The Faerie Queene and in imitations of that work and the widespread adoption of the Spenserian stanza. It can also be seen in the displacement of what Gray called the “French School” by the “Italian School.”25 That the revisions in the classical poetic tradition were little followed up on until the Romantic period probably reflects a restlessness with established tradition, which found an antidote both further afield, in the exotic East, and closer to home, in the

THE EIGHTEENTH CENTURY

67

beginnings of the retrieval of forgotten medieval and ancient Germanic/ Scandinavian and Celtic material. This material was often seen as competing favorably with the classics.26 The craze for The Arabian Nights led to the crafting of original, quasi-Oriental tales that stand as direct precursors to today’s “Oriental Fantasy” subgenre. The paucity of authentic Celtic material available during the period would lead MacPherson to invent a body of ersatz traditional narrative poems,27 which can be seen as the original “Celtic Fantasy.” Working in a context of growing comparative theory, Blake would fashion a syncretistic invented mythology, anticipating those of Dunsany and Tolkien. MacPherson’s Ossian, Blake’s Prophetic Books, and Haywood’s Eovaai stand as the most significant individual works. But perhaps as important as these individual works in themselves (none of which can be said to have had any notable direct influence in this context) is the cultivation of a sensibility anticipating that of twentieth-century fantasy. “Tradition” became something to be engaged with the imagination; the qualities of marvel and enchantment began to be characterized, in themselves, as quintessentially “poetic,” and invention a key quality of the poet. In turn, the imagination looked to the remote—the medieval, Celtic, and Germanic/ Scandinavian past; the “Orient”—for models and inspiration. And the trickles of creative work were accompanied by the halting work of antiquarians: “specimens” comprising the first modern, mediated texts, which would serve as sources, models, and inspiration to later fantasy writers. Works Cited Blake, William. Complete Writings. New York: Oxford UP, 1966. deGatno, Paul J. James MacPherson. Boston: Twayne Pubs., 1989. Dryden, John. The Poetical Works of John Dryden. New York: George Routledge and Sons, n.d. Frye, Northrup. Fearful Symmetry. Princeton: Princeton UP, 1969. Haywood, Eliza. The Adventures of Eovaai. Orchard Park, NY: Broadview Literary Texts, 1999. Lonsdale, Roger, ed. Thomas Gray and William Collins: Poetical Works. New York: Oxford UP, 1977. Mack, Robert L., ed. Oriental Tales. New York: Oxford UP, 1992. Matthews, Richard. Fantasy: The Liberation of Imagination. New York: Routledge, 1997. Parnell, Thomas. Poetical Works. Freeport, NY: Books For Libraries Press, 1972. Percy, Thomas. Reliques of Ancient English Poetry. New York: Dover Books, 1966. Pope, Alexander. The Best of Pope. New York: Ronald Press Co., 1929. Salzman, Paul, ed. An Anthology of Seventeenth Century Fiction. New York: Oxford UP, 1991.

68

THE EVOLUTION OF MODERN FANTASY

Thomson, James. The Complete Poetical Works of James Thomson. New York: Oxford UP, 1908. Warton, Thomas. Observations on the Fairy Queen of Spenser. Excerpted in Woods, George, ed. English Poetry and Prose of the Romantic Movement. New York: Scott, Foresman and Co., 1929.

Other Secondary Work Consulted Cummings, R. M., ed. Spenser: The Critical Heritage. Totowa, NJ: Barnes and Noble, 1971. Feldman, Burton, and Robert D. Richardson, eds. Rise of Modern Mythology. Bloomington: Indiana UP, 1972. Irwin, Robert. The Arabian Nights: A Companion. New York: Penguin Books, 1994. MacPherson, Jay. The Spirit of Solitude. New Haven: Yale UP, 1982. Merriman, James. Flower of Kings. Lawrence: U of Kansas P, 1973. The editorial apparatus in Mack’s Oriental Tales and Lonsdale’s edition of the poems of Gray and Collins, both listed above, were also consulted.

Some Primary Source Editions For most of the poetry, the editions used are listed under Works Cited; in general, poems mentioned can be found in any complete edition of their poet’s work. For Ossian, there has been only one modern edition: MacPherson, James. The Poems of Ossian and Related Works. Edited by Howard Gaskill. Edinburgh: Edinburgh UP, 1995. For the prose, Johnson’s Rasselas is available in countless editions. Mack’s Oriental Tales, listed under Works Cited, includes Hawkesworth’s Almoran and Hamet, Sheridan’s History of Nourjahad, and Reeve’s The History of Charoba, Queen of Egypt. Beckford’s Vathek is also available in many editions, but only one also includes the complete “Episodes,” with both versions of the first episode: Beckford, William. Vathek with The Episodes of Vathek. Edited by Kenneth W. Graham. Orchard Park, NY: Broadview Literary Texts, 2001.

3

Romantic Transformations

Introduction

D

uring the Romantic period—stretching from the few years adjacent to 1800 to the later 1830s—the literary currents of the eighteenth century discussed in Chapter 2 developed considerably and became somewhat less stratified. The number of actual narratives pertinent to the development of fantasy increased dramatically: during the first half of the period, these were almost exclusively poetic, but during the latter half, there were some notable developments in prose fiction. The quasi-Oriental tale, the major prose development of the eighteenth century, migrated almost exclusively to verse; shorter narratives using traditional ballad verse forms, as well as longer narratives drawing on the vocabulary of the metrical romance, proliferated during the first half of the period. During the latter part of the period, the classical and Elizabethan-cum-medieval aesthetics that had developed in reaction to the Augustans merged, particularly strongly in the work of Keats. At the very end of the period, much of this was drawn together in the first work—unless one excepts the Eastern-hued Eovaai of Eliza Haywood—unambiguously embodying the core elements of the BAFS template: Sara Coleridge’s Phantasmion. Though there is less of a distance here from the standard canonical narrative than was the case with the eighteenth century, there are noteworthy differences in emphasis and some departures. The prose fiction, though in some cases attached to names not entirely off the grid (Peacock, Hogg), has been almost entirely neglected. The major narrative poems of Keats and Coleridge are significant both in the canonical narrative and here, while Wordsworth and Byron are of negligible significance. Poets widely read in their day but subsequently written out of the canonical narrative— Southey, Scott—are of notable significance here. Advances in the work of antiquarians, particularly in the field of language, signaled a shift in literary context: the trickle of texts and translations

70

THE EVOLUTION OF MODERN FANTASY

that had begun to pick up in the 1760s became a broad stream, and by the late 1830s, a sufficiently extensive body of traditional material was available to facilitate, even for a lay reader, a level of expertise far beyond that of Thomas Warton, Gray, or Percy. Sir Walter Scott published his Minstrelsy of the Scottish Border in 1802–3,1 which included 75 traditional ballads, including the first printed versions of “Tamlane” and “Thomas Rhymer.” A flood of such collections followed,2 culminating in Robert Chambers’s The Scottish Ballads (1829), an attempt to create a “complete” collection, drawing on the author’s predecessors and other sources.3 With Joseph Ritson’s4 Ancient English Metrical Romances (1802) and Scott’s edition of the Middle English Sir Tristrem (1804), full “specimens” of verse romance, including medieval Arthurian pieces, appeared in print for the first time, and as with the ballad, further collections appeared over the following decades, including Frederic Madden’s 11 romance Syr Gawayne (1839), which included the first publication of Sir Gawain and the Green Knight. Malory’s Le Morte d’Arthur returned to print, an 1816 edition based on the expurgated 1634 edition, followed by the full Caxton text, edited by Southey, in 1817.5 Celtic and Germanic/Scandinavian material remained somewhat nascent: translation of full, major works would await the Victorian period. The proliferation of such printed texts provided poets with a far broader representation of these “antique” literary forms than had been the case in the eighteenth century: the possibility of immersion in primary “specimens,” necessary for a new literary vocabulary to emerge, became progressively less difficult. It should also be emphasized that, though this material was “old” with regard to date of origin,6 printed editions of ballads or romances geared to modern readers, often appended with scholarly notes and essays, were a phenomenon as contemporary as Jane Austen’s novels, and it was as functionally “new” material that it met the writers that made use of it. In the following pages, I will discuss, first, the narrative poetry of the earlier Romantic period, itself often appended with antiquarian apparatus echoing eighteenth-century habits, and, second, that of the later Romantic period, where the antiquarian concerns decrease and aesthetic concerns begin to dominate. Then the discussion will turn to the fiction that began to appear toward 1820 and, last, after an account of some of the significant fiction by German romantic writers that began to appear in English translation during the period, to Sara Coleridge’s Phantasmion.

ROMANTIC TRANSFORMATIONS

71

Earlier Romantic Narrative Poetry The publication of traditional ballads and metrical romances was swiftly followed by original, literary works following the verse forms and, to varying degrees, the subject matter of the two modes. The vogue for the ballad form was largely restricted to the first decade of the century. The most famous of the literary ballads, also a major work in the canon of English Romantic poetry, is also probably the most idiosyncratic. Samuel Taylor Coleridge’s “The Rime of the Ancient Mariner” first appeared anonymously in Coleridge and Wordsworth’s Lyrical Ballads (1798), the collection often considered to usher in the Romantic period, and it was revised for the second edition of 1800. The verse form of “The Rime,” deceptively simple and direct, follows the traditional ballad. Few traditional ballads come anywhere close to the length of Coleridge’s poem, however,7 and the narrative matter is visionary: the fantastic elements do not echo their counterparts in the traditional ballads, localization of event is specious, and symbolism is deliberate, if cryptic, and literary. Coleridge’s preoccupation with German philosophy, and German poets such as Schiller (whose work includes ballads), are more reflected in the content of “Rime” than any traditional ballads.8 Yet it is the combination of traditional verse form with the vividly original fantastic elements—the albatross, the specter Life-in-Death—that perhaps has made Coleridge’s “imitation” the most abidingly provocative of such work from the period. There is an abundance of literary ballads from the opening decade of the nineteenth century that follow the traditional ballad more closely. Scott’s Minstrelsy was appended by nearly two dozen modern imitations, some by Scott9 and some by his friend and contemporary James Hogg.10 The ballads of Scott and Hogg, rooted in Border traditions, carried particularly patriotic resonances: the concern was with specifically Scottish tradition. Other poets roved more widely: the literary ballads of Robert Southey range from Spain to Germany to medieval legend for their subjects.11 It should be noted that many traditional ballads are not what we would term fantasy but are focused on “realistic” dramatic episodes founded in history. Supernatural or magical elements are more often than not spare, serving to reinforce a nonmagical dramatic focus; very often they are entirely absent. Nevertheless, the medieval-tinged atmosphere characterizing most of the ballads appearing in the collections of the time, regardless of the presence or absence of supernatural incident, certainly lent itself to supernatural association (and a Gothic-imbued aesthetic), and fantastic elements are probably more frequent in the literary imitations than in the traditional ballads themselves. It is, perhaps, no accident that the two

72

THE EVOLUTION OF MODERN FANTASY

most enduring of the literary imitations, “Rime” and Keats’s later “La Belle Dame Sans Merci,” are very much fantasies. It is notable that such tales as those of Thomas Rhymer and Tamlane (often Tam Lin), appearing for the first time in print in Scott’s Minstrelsy, have frequently been reworked, embellished, and extended in prose fantasies, from Mrs. Craik’s Alice Learmont (1852) to Henry Newbolt’s Aladore (1914) to Pamela Dean’s Tam Lin (1991). The ballad form itself has frequently been used for inset tales, often heard aloud by the characters, in fantasies from George MacDonald’s ballad of Sir Aglovaile in Phantastes (1858), to the tales of Beren and Luthien and of Nimrodel in LOTR (1954– 55), to some variations on the legend of White Jenna in Jane Yolen’s Sister Light Sister Dark (1988). Yet the spare, dramatic style that characterizes the traditional ballad, as well as the usually compressed narrative scope, imposed certain inherent limitations. This is no doubt related to the decline in literary ballad writing after the first years of the nineteenth century. The metrical romance was one source for ideas concerning less restrictive verse forms and more elaborate, expansive narrative constructs. An early breakthrough in this regard once again came from Coleridge. The incomplete “Christabel” was written in 1797 and 1800, originally intended for the second volume of Lyrical Ballads, but not published until 1816. While the Scottish setting, the ambiguous, unearthly figure of Geraldine, and the enmity between Leoline and Roland de Vaux would evoke ballad associations, the verse form adopted here facilitated a dense, lush texture quite distinct from the ballad, evoking an atmosphere that would seem in part derived from the Gothic novels of Anne Radcliffe. It is neither surprising nor inappropriate that the poem has often been treated in terms of the Gothic. At the same time, the degree to which the atmosphere contains the narrative, rather than being made a vehicle for melodrama, distinguishes “Christabel” from Radcliffe’s work: the poem’s medievalism can be seen to anticipate that of Keats and, later, the pre-Raphaelite medievalism of William Morris, as well as George MacDonald’s Phantastes. The effect of “Christabel” was felt sooner, however, and in a work more directly linked to the metrical romances. Walter Scott heard “Christabel” read aloud in 1801 and, enormously affected by it, sought to apply Coleridge’s approach to his own work: “[T]he singularly irregular structure of the stanzas, and the liberty which it allowed the author to adapt the sound to the sense, seemed to be exactly suited to such an extravaganza as I meditated” (Scott 6). This was The Lay of the Last Minstrel (1805). The poem’s frame depicts the aging minstrel Gilpin Horner at the castle of Anne, Duchess of Buccleuch, sometime after the rebellion of 1685; the “Lay” itself, set in the mid-sixteenth century, is an oral narration by

ROMANTIC TRANSFORMATIONS

73

the minstrel. The “Lay” is indubitably fantastic, with its sinister dwarf, the cursed book of the wizard Michael Scott, and the spell-casting of the “Ladye” of Branksome Tower. As with “Christabel,” there is a Gothic cast to the story, though Scott is clearly more interested in evoking the traditional milieu of the minstrels and, with the “Lay,” creating a tale that generates the illusion of actual tradition. The fantasy is underwritten by antiquarian concerns, and the poem’s preface echoes the more Enlightenment-bound judgments of eighteenthcentury scholars like Percy. Scott’s stated first concern was “to illustrate the customs and manners which anciently prevailed on the Borders,” and the “plan” of the metrical romance, though lacking “the dignity of a regular poem,” was deemed the best vehicle for his intention. The “machinery . . . adopted from popular belief [presumably the supernatural elements] . . . would have seemed puerile in a Poem which did not partake of the rudeness of the old Ballad, or Metrical Romance” (Scott 1). It is difficult to determine to what degree these judgments reflected Scott’s genuine critical view and to what degree he was anticipating, and attempting to stave off, criticism of certain aspects of the poem, justifying “puerile Machinery” such as the dwarf or Michael Scott’s book as reflecting “popular belief.”12 Regardless, such “machinery” disappears from most of the half dozen or so narrative poems that followed also using the “plan” of the metrical romance.13 Most of these, including the most well-known, The Lady of the Lake (1810), are not pertinent here. An exception is The Bridal of Triermain (1813), one of the surprisingly few Arthurian works of the Romantic period. The Bridal incorporates three strands: the tale of two lovers, Arthur and Lucy; the tale of Roland de Vaux, Lord of Triermain, and his love for Gwyneth, asleep for five centuries since the time of King Arthur, told by the Arthur of the frame; and “Lyulph’s Tale,” an invented Arthurian tale that reveals that Gwyneth is the daughter of King Arthur by Guendolen, daughter of a genie. Scott’s antiquarian penchant is evident here: the House of Triermain is historical, and while the Arthurian tale is invented, the traditional material embodied in it is annotated. At the same time, the scholarly apparatus is far slighter than that appended to his other extended narrative poems. Though the poem is not wholly successful, it is certainly the most interesting of Scott’s longer verse narratives in fantasy terms. The inset Arthurian tale itself is engaging, compounding originality and a facility with the conventions of medieval Arthurian romance, and it stands as an early precursor of the Victorian work of Tennyson, Morris, and Swinburne and, in turn, of Clemence Housman and T. H. White. Scott’s poems, peripheral at best in the standard canonical narrative, are significant here for the way they began to define the peculiar relation between traditional narrative modes and the modern imagination, which

74

THE EVOLUTION OF MODERN FANTASY

as I have suggested is at the heart of the literary strand of the BAFS canon. Despite his pedantic tendency to validate his work via scrupulously detailed scholarly annotation, and his deliberate use of conventions derived from ballad and romance, Scott was also very much attuned to what the contemporary readership found attractive in the Romantic past: Merriman notes that the “romantic thrills” that attracted his readers “often owed more to Gothic horror fiction, German and English, than to medieval sources” (119). Even the folkloric Minstrelsy, as noted before, was freely revised and rewritten, which no doubt contributed to the fact that it sold better than the more scientifically sound collections of Ritson. Scott’s work was quintessentially “modern” and in fact contingent on—as is later the case with fantasy—distance from the contexts in which the narrative modes that inspired him flourished. The imaginative sensibility that emanated from Scott’s work had an enormous effect on the literary treatment of the Middle Ages from the end of the Romantic period through the remainder of the century, not least on the types of medievalism that underscore the fantasies of Morris and MacDonald. While drawing on the same patriotic allure inherent in the traditional subject matter of most of Scott’s verse narratives, James Hogg moved beyond the “plan” of the metrical romance in his longer narrative poems. Hogg’s relationship to the traditional material was rather different than the genteel Scott’s: where the latter’s knowledge was the result of deliberate cultivation, Hogg, a self-educated man of the rural classes,14 grew up with the oral traditions of the border country all around him and had, in fact, been the source of many of the traditional ballads collected in Minstrelsy. Where Scott’s interests were generally confined to the provenance of the genteel and aristocratic, Hogg was equally interested in the lore of the unlettered classes. To some degree, Hogg may be said to have followed Scott’s leads. Like the Lay, the long poem that established Hogg’s initial reputation, The Queen’s Wake (1813), stands as a memorial to the Scottish bardic tradition. But Hogg’s work is encyclopedic: within a historical frame,15 a dozen bards tell a dozen tales. Some are historic, but most venture into romance and the fantastic. While the sentimental quality attached to the “last” minstrel and a deliberately cultivated Gothic atmosphere are key features of Scott’s poem, Hogg’s poem, though nostalgic, is not sentimental, nor does he develop a Radcliffean atmosphere or idealize his characters. The supernatural is presented unapologetically: one cannot imagine Hogg referring to such “machinery” as “puerile” and excusing it on the grounds that it reflects “popular belief.” While less self-consciously scholastic than Scott,16 Hogg’s work is, in tone and texture, closer to the actual Scottish traditional

ROMANTIC TRANSFORMATIONS

75

sources, not layered over with a picturesque romantic glaze—which may partly explain why he achieved nothing like Scott’s commercial success. At the same time, Hogg was often more experimental than Scott. His follow up to QW, The Pilgrims of the Sun (1814), is a visionary narrative following the humble Mary Lee and her guide, Cela, on a voyage to the heavens, founded on Scottish tradition. But the voice and verse forms evoke, at different points, not only Scots oral tradition but Milton, Pope, Scott, and the Bible. The later Queen Hynde (1824) is cast as an epic, drawing on prehistoric material in part associated with Ossian but marked by a sophisticated, ironic narrative voice. In addition to these, there are a fair number of shorter narrative poems of fantasy interest scattered through Hogg’s work. While Hogg cannot be said to have had anything like the influence that Scott had,17 his work is in many respects more interesting. His relationship to the traditional material that provided him with much of his subject matter, and was a major determinant of his style, is unique among the writers with which this study is concerned.18 None of the other contemporaries of Scott and Hogg built their work so closely on indigenous tradition. Together with Ossian, which in many respects Scott and Hogg stand as a logical extension of, their poetry represents a filling in of the territory speciously suggested in the lyrics of Gray, Collins, and the Wartons. Looking forward, the two, with their transformation of Scottish legend, indigenous fairy tradition (Hogg), and Arthurian legend (Scott), represent the first substantial development in modern narrative of areas of subject matter now widely considered the domain of fantasy. The earlier Romantic period saw a number of poetic fantasies at the other end of the spectrum. Though the quasi-Oriental fantasy in prose largely evaporated after Vathek, the milieu began to appear in narrative poetry. One of the first, and most abidingly famous of these, is once again by Coleridge. “Kubla Khan,” written in late 1798—though, like “Christabel,” not published until 1816—is brief, and more visionary and descriptive than narrative, but the compressed lush imagery distilled from the same “exotic” cloth that bred the earlier prose tales is significant here. More expansive than Coleridge’s fragment, and probably the first English attempt at genuine quasi-Oriental verse narrative, is Walter Savage Landor’s Gebir (1798, revised 1803), its author’s only extended narrative poem. A bit of the rough syncretism that characterized Parnell’s “Fairy Tale” is evident here: the “Eastern” focus is wedded to Landor’s classicism. The opening of the revised version, “I sing the fates of Gebir!,” recalls Virgil rather than The Arabian Nights, and references to Satyrs, Bacchus, and Nymphs in the deleted opening lines of the original compound the

76

THE EVOLUTION OF MODERN FANTASY

classical association (Landor 3). The tone of the poem as a whole is heavier and more heroic than any of the other quasi-Oriental tales from the adjacent periods, verse or prose. Though the thorough melding of the classical and the quasi-Oriental has rarely been followed,19 Robert Southey did make a superficial move in this direction: the 12 books of Thalaba the Destroyer (1801) and the 24 of The Curse of Kehama (1810) implicitly suggest Virgil and Homer, respectively. But despite his claims to the contrary,20 both are more appropriately seen as quasi-Oriental narratives. Thalaba takes as its point of departure “the Domdaniel . . . a seminary for evil magicians, under the roots of the sea,” mentioned “in the continuation of the Arabian Tales” (Southey 23), while Kehama was inspired by early English accounts of Hindu mythology. Far more expansive and ambitious than Landor’s poem, they are essentially quasi-Oriental “fantasy novels” in verse. Both were widely read in their day, and enthusiasts included Keats and Shelley. On the other hand, they are now primarily remembered due to Byron’s mockery in the opening lines of Don Juan and his pronouncement that Southey would “be read when Homer and Virgil are forgotten, but— not till then” (Byron 109). However, while some of Byron’s criticisms are valid, his attacks were to a large degree politically motivated, and it can be said that the poems are not as bad as legend has them. Southey was a reasonably adept versifier and had some power of invention. However, these powers were best displayed in the less ambitious ballads and metrical tales mentioned previously: Southey had neither the depth of thought nor the sustained imaginative power to render his “epics” (as he called them) successful. Thalaba, for example, has numerous episodes with considerable dramatic potential that are time and again resolved simply by sudden miraculous intrusion: the poem becomes very repetitive. Read as burlesque, Thalaba and Kehama can be engaging, but needless to say, this was not Southey’s intent. The last major poetic adaptation of the quasi-Oriental fantasy by a Romantic poet was Thomas Moore’s Lalla Rookh: An Oriental Romance (1817). Unlike Southey’s attempt to suggest classical form with his book divisions, Moore followed The Arabian Nights. Within a prose frame, narrating the journey of Lalla Rookh from Delhi to her marriage in “Cashmere,” four fantastic verse tales are interspersed, told to relieve the monotony of the journey. The poem is given fairly extensive scholarly annotation, though this historicizes neither more nor less than Henley’s annotations to Vathek. Moore’s work is considerably longer than Landor’s, of higher quality than Southey’s “epics,” and is probably the most exceptional of the Romantic attempts at the quasi-Oriental tale in verse.

ROMANTIC TRANSFORMATIONS

77

The continuity between the literary threads discussed in Chapter 2 and this body of first-generation Romantic poetry can be briefly summarized. The scholastic preoccupations of the eighteenth century can be seen in the annotations that accompany the works of Scott and Hogg, as well as of Moore and (to a lesser degree) Southey, and in the insistence that the basis and often style of those works is founded in authentic tradition or history. The work of Scott and Hogg, the imitations of the “Ancient Ballad,” both to some degree represent a fulfillment of the eighteenth-century ideal of a poetry based on indigenous, rather than classical, models and subjects. At the other end of the spectrum, the quasi-Oriental poems of Southey and Moore develop in verse the marvelous milieu of their eighteenth-century prose predecessors. These threads also look forward to the twentieth century. The quasidocumentary apparatus that we find, to varying degrees, in Cabell, Eddison, and Tolkien is a fictional extrapolation on the structure and content of the type of documentation that accompanied these Romantic poems. The conscious development of an Anglo-Germanic rather than a Latinate linguistic base by both William Morris and Tolkien, and Tolkien’s construction of a syncretistic “English” mythological and historical structure drawing on processed fragments of British and Northern tradition, follow naturally from Scott and Hogg. Southey and Moore stand between Haywood and Beckford on the one hand and Dunsany and Bramah on the other. Later Romantic Narrative Poetry: Neo-Elizabethan Processing The second-generation Romantic poetry relevant here—most particularly the work of Keats—departs from much of the first-generation work in key respects. The antiquarian concerns embodied in meticulous annotation and commentary largely evaporate, and the more emphatically aesthetic preoccupations that become dominant are more akin to Coleridge than to Scott or Southey. Traditional subject matter becomes more specious, significant not so much in itself as in its ability to serve specifically poetic ends.21 The canvas becomes broader and more inclusive: while the quasiOriental narrative as an organic entity largely ceases with Lalla Rookh, the imagery attached to it often becomes integrated into a more syncretistic poetic vision. Hearkening back to some of the tentative developments earlier in the eighteenth century, we find a turn to subject matter drawn from classical antiquity and a developing neo-Elizabethanism, the latter becoming a lens through which the former, as well as the medievalism of the previous generation, is interpreted.

78

THE EVOLUTION OF MODERN FANTASY

The work of Byron is of, at best, peripheral interest here. While Don Juan (1819–24) turns to the East, it is the less glamorous East of contemporary accounts rather than the fabulous east of The Arabian Nights; while the poem may be read as something of a Cervantes-esque parody of some of the excesses that marked, particularly, the long poems of Southey, it is not a fantasy itself. Child Harold’s Pilgrimage (1812–18) uses the Spenserian stanza form and a mildly archaistic diction, but the content is not fantastic. On the whole, Byron’s work stands apart from the major developments discussed here. The work of Percy Bysshe Shelley has rather more to offer in the present context, though it, too, is peripheral beyond a certain point. Alastor, or the Spirit of Solitude (1815) is of interest here. Akin to Beattie’s The Minstrel in certain key respects, the poem can be seen as a secularized embodiment of the dream vision, its straightforward linear plot following the “education” and development of poet rather than Christian, as in Bunyan.22 Shelley’s poem moves more overtly into the fantastic that hovers on the periphery of The Minstrel and is more marked by the romance and mythic elements that characterize earlier dream-vision narratives. Here, the hazy, impressionistic setting through which the protagonist wanders is shot through with quasi-Oriental imagery and referents, merged with the classical and the biblical: The awful ruins of the days of old: Athens, and Tyre, and Balbec, and the waste Where stood Jerusalem, the fallen towers Of Babylon, the eternal pyramids, Memphis and Thebes, and whatsoe’er of strange Sculptured on alabaster obelisk, Or jasper tomb, or mutilated sphinx, Dark Ethiopia on her desert hills Conceals . . . (Shelley 114)

This aesthetic, progressing on to “Arabie / And Persia” (115) and “the ethereal cliffs / Of Caucasus” (121), implicitly suggests an organic relation between the fabulous representation of the East and the poetic imagination, an extension of quasi-Oriental imagery and atmosphere beyond the bounds of the quasi-Oriental tale itself. Though Alastor certainly had no broad impact on the development of fantasy,23 the more consolidated, syncretistic aesthetic embodied in its imagery is notable here. Shelley’s engaging of indigenous traditional forms or subject matter24 came in works concerned primarily with the articulation of his developing political ideology. He turned to explicitly classical subject matter in work

ROMANTIC TRANSFORMATIONS

79

less overtly political: a minor work, “The Witch of Atlas” (1820), is founded on Greek mythology, though the main plot line is of Shelley’s invention; perhaps his greatest work, Prometheus Unbound (1820), a sequel to Aeschylus’s Prometheus Bound, is dramatic. The Greece of Shelley, as the passage from Alastor suggests, is turned to the Romantic East rather than the sober, staid, instructional classics of Pope. This is also true of the Greece of Shelley’s younger contemporary John Keats. In the early sonnet, “On First Looking into Chapman’s Homer” (1817),25 Homer’s works are depicted as a “demesne” in the imaginative “realms of gold,” and the reader’s “travels” into that work are travels of discovery: the invocation of a dehistoricized Cortez suggests the penetration and opening up of an exotic territory, unknown and remote from the present world, to which its “pure serene” implicitly stands in contrast. The “realms of gold” are as dehistoricized as Cortez; there is not even the kind of appeal to classical practice that Shelley makes for Prometheus Unbound.26 This is not the Homer or Greece of antiquarians or classical scholars, however much it may have depended on their work: as with Shelley, at its center is a peculiarly modern, imaginative perspective on antiquity, far closer to the “imaginary worlds” of fantasy than the classical milieu associated with the Enlightenment. This ahistorical, visionary depiction of Greek antiquity lies at the heart of four of Keats’s works pertinent here: Endymion: A Poetical Romance (1819), “Lamia” (1820), and the two fragments “Hyperion” (1820) and “The Fall of Hyperion,” published posthumously. The first of these, while rooted in Greek mythology, nevertheless embodies the modern poetic concern with invention and the imagination: “I must make 4000 lines of one bare circumstance, and fill them with poetry” (Woods 1288). Further, in casting his “Poetical Romance” (the designation recalls Scott), Keats does not follow any Greek models with regard to form, and in fact his poetic technique and treatment of his subject matter connect far more strongly to the poems on Greek mythological subjects that proliferated during the Elizabethan period and the earlier seventeenth century.27 The impetus behind the poem was a stanza from Spenser’s The Faerie Queene (see Woods 1288). Endymion is not altogether successful, and Keats himself was dissatisfied with it: it is broken and difficult to follow on a narrative level, and in its symbolic emphasis, it is more along the lines of Shelley’s Alastor than his later “Greek” poems. At a sprawling 4,000 lines, it is considerably less concise than Shelley’s poem. “Lamia,” included in 1820’s Lamia, Isabella, The Eve of St. Agnes, and Other Poems, is also founded on a Greek myth, but it is considerably sparer, with a sharper narrative focus. It also follows his source—not a classical source, but Thomas Browne’s Anatomy of Melancholy—more closely,

80

THE EVOLUTION OF MODERN FANTASY

though it is considerably amplified and adapted to convey Keats’s own thematic concerns. Where Browne’s brief redaction of the story suggests little more than that Lamia has deceived Lycius, for Keats, Lamia is the imagination, illusion becomes “charms,” and, in the final confrontation between the philosopher Apollonius and Lamia, it is the philosopher who is demonized: “Do not all charms fly / At the mere touch of cold philosophy?” (Keats 190). It is interesting here to note that Keats depicts as absolutely irreconcilable the two forces that Dryden implicitly conflates in “To Oedipus,” discussed at the beginning of Chapter 2. Keats’s death at 26 left only two fragments of his projected treatment of the Greek myth of Hyperion the sun-god: “Hyperion,” whose three books (the third incomplete) were published in the 1820 collection, and the beginnings of a substantial reshaping of the story, “The Fall of Hyperion: A Dream,” half the length of the first attempt, published posthumously. In the first fragment, Keats was seeking to press his revisioning of the myth into an epic form modeled not on Homer but on Milton. In a move resembling Blake’s progression from the abortive Four Zoas to Milton and Jerusalem, Keats’s second fragment, in which he turned from Milton to Dante as a model, seeks a greater immediacy, introducing the poet as a character and casting the myth as a vision. Keats did not restrict himself to Greek subjects, and some poems connect to the areas already discussed here. “La Belle Dame Sans Merci,” originally written in 1819 but published in an inferior revised version in the 1820 collection, recounts in ballad form the seduction and abandonment of an unnamed “Knight at arms” by a “faery’s child.” The poem strikes a memorable balance between the compression implicit in the traditional form and Keats’s characteristically smooth and polished versification. The early “Calidore: A Fragment” (1817) likewise evokes the world of medieval romance. “The Eve of St. Agnes,” founded on a medieval Roman Catholic superstition, develops a Gothic-imbued medievalism akin to that of Coleridge’s “Christabel,” and “Isabella” is founded on a story from Bocaccio. Yet these works occupy the same imaginative space as those drawing on Greek mythology: as the latter do not draw formally on ancient Greek models, so the former do not (with the qualified exception of “La Belle Dame”) draw on medieval models. The crucial glue for Keats lies in the English tradition, with its alignment toward Italian rather than French models—a tradition that, some three quarters of a century earlier, Thomson, Gray, and Collins sought to revive in their work. Both the classical and medieval strands in Keats’s work are processed through a neo-Elizabethan/ seventeenth-century lens: the Homer of Keats’s sonnet is that of Chapman’s translation; the impetus behind Endymion was Spenser; the source for “Lamia” was Thomas Browne; the formal models for the Hyperion

ROMANTIC TRANSFORMATIONS

81

fragments were Milton and Dante; “The Eve of St. Agnes” is cast in the Spenserian stanza form; Keats’s sole completed play, Otho the Great (published posthumously), is modeled on the five-act Elizabethan and Jacobean dramas. Keats’s work is marked by a more thoroughgoing syncretism than that of his older, first- and second-generation Romantic contemporaries. Lacking the antiquarian impetus of Scott, Keats’s weave of Greek and medieval subjects, given a neo-Elizabethan processing and sporadically flavored with the glamour of the romanticized East,28 becomes in effect a self-contained imaginative world, a consolidated and original aesthetic milieu, from which it is a short step to the invented worlds of later fantasy. The effect of Keats can be seen in the work of the third-generation Romantics Thomas Lovell Beddoes and George Darley. However, this is apparent mainly in lyric and dramatic work. Of their few narratives, Beddoes’s “The Romance of the Lily” (1823) invokes the world of classical antiquity via Shakespeare (Venus and Adon are invoked) and turns to Libya in the main narrative; “Pygmalion: The Cyprian Statue” (1825) retells a classical myth directly.29 Darley’s Nepenthe (1835), which moves into thematic territory akin to Shelley’s Alastor and Keats’s Endymion, is noteworthy here.30 More significant, Keats had a profound effect on those Victorian poets who will be discussed in Chapter 4 (Tennyson, Swinburne, Morris), whose work immediately precedes the almost complete conversion of the fantastic romance to the prose medium. Certainly Keats, directly and indirectly through the Victorian poets, was a factor in the aesthetic that marks the work of Dunsany, Kenneth Morris, and Eddison; in the case of Eddison (who numbered Keats among his favorite poets), the synthesis of Greek, medieval, Italian Renaissance, Celtic, and Northern elements, processed through a neo-Elizabethan/Jacobean prose style, is striking. The relevance of this body of Romantic narrative poetry, early and late, to the history of fantasy should be clear enough. On the most superficial level, Coleridge’s “Rime,” Scott’s Last Minstrel, Southey’s Thalaba, Keats’s “Lamia,” and so on are narrative fantasies as much as anything in the BAFS canon. More important, we see the major strands discussed in the previous chapter—the aestheticization of romance and mythological matter, the “recovery” of tradition by means of imaginative extrapolation, the deliberate imaginative cultivation of the “exotic”—all advancing and evolving. With Scott at one end of the spectrum, and Keats at the other, we can see the essential aesthetic foundation that facilitated later prose fantasy (as it proceeds from William Morris) achieving an increasing degree of definition.

82

THE EVOLUTION OF MODERN FANTASY

Romantic Prose Fiction The related Gothic novel transformed in the early years of the nineteenth century, producing some work, such as Maturin’s labyrinthine Melmoth the Wanderer (1820), of peripheral interest here. But on the whole, the offspring of the eighteenth-century form moved in the direction of horror and supernatural fiction. Scott’s more antiquarian historical romances, beginning with Waverly (1814), often drew, as his narrative poems had, on a Gothic aesthetic, and those set during the Middle Ages, such as Ivanhoe (1820), played a major shaping role in the pre-Raphaelite medievalism of William Morris. In addition, the often swashbuckling action focus of Scott’s romances fed into the popular adventure fiction of the nineteenth century, which in turn fed into the more action-based forms of twentieth-century popular fantasy. But like most of his longer narrative poems, Scott’s prose fiction in general eschews magical or supernatural elements and, on its own merits, stands outside the focus of this study. Prose fiction closer to the strands of literary history developed here is not prolific, but the 1820s saw two novels of note, and a fair spread of short fiction appeared during that and the following decade. Though these works are idiosyncratic and do not resolve into any consolidated pattern, they do stand as interesting developments of the various threads coalescing throughout the period. While three of James Hogg’s four novels do not engage any overtly fantastic elements,31 his second novel, The Three Perils of Man (1822), sits well in the present context. Set in “the days of chivalry and romance” (Hogg 5) during the later fifteenth century, and centered on the Scottish siege of the Castle of Roxburgh, held by the English, the historical thread of the novel evokes the rugged, dramatic world of the traditional ballads, which Hogg had imbibed since his childhood. But this historical emphasis is deliberately undercut: the enigmatic figure of the legendary wizard Michael Scott is transplanted unapologetically from the twelfth century, and the central sequence of the novel follows the quest by a fellowship of seven, led by Charlie Scott, to the Castle of Aikwood, where they encounter the phantasmagoric machinations of the wizard while seeking his prophetic assessment of the siege at Roxburgh. This plot strand, countering the historical drama surrounding the Castle of Roxburgh, subsumes the narrative into a broader fantasy context. Hogg’s style is direct and vigorous: his invocation of “the days of chivalry and romance” is ironic, carrying no idealization a la Scott, and his depiction of the warring aristocrats is often darkly comic; the imaginative energy driving the magical sections is vivid, not built on a cultivated Gothic aesthetic. The figure of the wizard and the

ROMANTIC TRANSFORMATIONS

83

attendant magical scenes, the fellowship and the quest, the context of war, and the opening chapter’s suggestions of destiny all anticipate much in later fantasy. Certainly Hogg’s most ambitious work of fiction, and arguably a greater accomplishment than the better known Justified Sinner, TPM stands as a remarkable imaginative work that deserves to be more widely recognized.32 The second of the two novels, far shorter than Hogg’s work, is Thomas Love Peacock’s The Misfortunes of Elphin (1829). Peacock’s earlier fiction33 had often used Gothic machinery to frame what were satiric conversation novels; Maid Marian (1822) turned to the medieval legends of Robin Hood for its subject, though like the traditional ballads themselves, it is devoid of fantastic elements. Misfortunes is an Arthurian piece, in which Peacock drew not on the courtly metrical romances or Malory but on the allusive “Triads” and other early Welsh poetry, translations and imitations of which are liberally scattered throughout the text.34 The core narrative concerns Taliesin and his foster father, Elphin, displaced king of Caredigion, which has disappeared beneath the sea, and the former’s machinations to free the latter from imprisonment by Maelgon. Arthur, the king of Welsh tradition with Welsh-named champions,35 only appears in the closing section of the story. While the narrative is diffuse and not entirely successful, Peacock’s ironic treatment of his subject—a polished, literary irony quite distinct from Hogg’s—facilitated by the droll pedantry of his narrator, the comic exaggeration of his characters’ conversations, and an artificial prose that often echoes the triple rhythms of the “Triads” produces many brilliant scenes. As the first modern Arthurian—and Celtic—prose fantasy, and as an original and quite readable work in its own right, Misfortunes deserves more attention than it has gotten in studies of fantasy.36 Shorter fiction from this period includes work by a number of writers. Within the voluminous pages of Hogg’s short fiction are a fair number of stories, like “The Hunt of Eildon” (1818) and “The Brownie of the Black Haggs” (1829), that employ fantasy motifs such as magical transformations and encounters with fairies, brownies, and witches. That the darker, demonic strands of Scottish tradition often pervade these stories has frequently led to their being classed as “Gothic,” though as noted before, Hogg’s work is generally free of calculatedly Gothic trappings. Benjamin Disraeli’s two reworkings of classical myths, “The Infernal Marriage” (1832), based on the myth of Proserpine, and “Ixion in Heaven” (1832–33), are noteworthy here.37 Unlike the lush, aestheticized treatments of classical material by Keats, however, Disraeli’s stories are unambiguously satirical and ironic, bordering on burlesque, and it is perhaps noteworthy that he uses Roman rather than Greek nomenclature. In this, Disraeli’s imagination may be seen to have more in common with Dryden and Swift

84

THE EVOLUTION OF MODERN FANTASY

than with the Romantics, though one might note some affinity to Peacock. These stories, together with Peacock’s novel, can be seen to anticipate the more ironic treatments of mythic and romance subjects in the work of Richard Garnett, James Stephens, and James Branch Cabell.38 Between the late 1820s and 1840, John Sterling, a writer little regarded in his own day and largely forgotten after the mid-nineteenth century,39 published a fair quantity of fantasy stories of various types. In those stories closest to the focus of the present study, Sterling drew on mythic and legendary settings and sometimes characters, though the stories themselves tend to be largely his own inventions. The earlier of these, including “Zamor” (1828) and “Cydon” (1829), tend toward Greek subjects, while the later move away from classical reference points: the setting of “The Palace of Morgana” (1837) is Fairyland (while the name “Morgana” suggests an Arthurian connection, this is unelaborated); “Land and Sea” (1838) is given a specifically English setting, and the plot content echoes many of the traditional ballads then in circulation; “A Chronicle of England” (1840) narrates an invented history of “Faery” (England) up to the arrival of humans. Sterling’s work at times suggests Keats and Shelley, at others German writers such as Goethe, Hoffmann, and Fouqué. In its philosophic cohesiveness, his work in some measure anticipates that of both George MacDonald and Kenneth Morris. But apart from the inclusion of “Chronicle” in Stableford’s Dedalus Anthology of British Fantasy, and a brief entry in Clute and Grant’s Encyclopedia, Sterling has remained oddly neglected. His fantasy work certainly deserves a modern edition. Sterling’s fondness of the German Romantics points to the next section, but here should be mentioned a number of loosely allegorical tales based on German folklore and legend written by Edward Bulwer-Lytton. Some, including “The Nymph of Lurlei Burg” (1832)—also included in Stableford’s anthology—were published in periodicals; others formed part of The Pilgrims of the Rhine (1834), a highly eccentric combination of sentimental love story, fairy tale, travelogue, folklore, and literary rumination, worthy of mention here in its own right.40 But while this material—much of it of very high quality—deserves far more attention in studies of fantasy than it has gotten, it is ultimately less significant in terms of influence than the Romantic narrative poetry discussed throughout the better part of the chapter. It simply did not, regardless of its prose medium, have as much shaping effect on the aesthetic that fed into the work of Morris and MacDonald as the work of Scott and Keats did. Hogg’s work is really sui generis, and his great work, TPM, in effect disappeared after its first edition (see note 32). Misfortunes has had its admirers throughout the years, but while Peacock’s trenchant irony anticipates Stephens and Cabell, it is doubtful that the likeness is the result of

ROMANTIC TRANSFORMATIONS

85

influence; though it may be the first prose Arthurian fantasy, it cannot be said to have any discernable effect on later Arthurian work. The stories of Disraeli, Sterling, and Bulwer-Lytton are slighter, and what influence they may have had is limited and speculative.41 German Romanticism and Phantasmion As noted before, there are links to the German Romantics, and preRomantics, in the work of Coleridge, Scott, Sterling, and Bulwer-Lytton, and the Romantic period saw a fair number of translations from German of prose work closer to the BAFS template, and with broader influence, than the prose work just discussed. The most broadly influential phenomenon was, of course, the publication of Jacob and Wilhelm Grimm’s Kinder- und Hausmärchen (1812/1815), appearing in Edgar Taylor’s English German Popular Tales (1823), which set off the folktale-collecting craze of the nineteenth century. However, with regard to fantasy, the effects of this were felt primarily in the Victorian period, in the context of writing for children, so I will discuss the Grimms in Chapter 4. Between the French contes de fees of the late seventeenth and early eighteenth centuries and the huge proliferation of literary fairy tales for children in English during the Victorian period, the German Romantic period stands as the third major pocket of European fairy-tale activity. This stretched back into the work of pre-Romantics such as Christoph Martin Wieland, whose original “The Philosopher’s Stone” appeared with adaptations of French tales in Dschinnistan (1786–89), and Johann Karl August Musäus, whose Volksmärchen der Deutschen (1782–86) contained essentially original tales founded on German legends. Goethe in his “Fairy Tale,” which closes Conversations of German Emigrants (1795), and Novalis in both “Klingsohr’s Tale” and “Hyacinth and Rosebud,” embedded in the fragmentary Heinrich von Ofterdingen (1800) and Novices of Zais (1802), respectively, move in a more symbolic and metaphysical direction, anticipating MacDonald’s Phantastes and “Golden Key.” E. T. A. Hoffmann wove fairy-tale patterns and motifs, and a penchant for Romantic myth making, into a more Gothic fabric in stories such as “The Golden Pot” and “The Mines of Falun,” though the fantastic in Hoffmann is usually marked by a certain ambiguity. Though Hoffmann had a strong effect on MacDonald,42 and “The History of Krakatuk,” embedded in Nutcracker (1815), may be seen behind some of the more humorously satirical Victorian fairy tales, his influence is more notable in areas outside the present focus.

86

THE EVOLUTION OF MODERN FANTASY

Less well-known than these, but perhaps more significant, are many of the tales of Ludwig Tieck and Baron de la Motte Fouqué. The former, in short tales such as “Fair-Haired Eckbert” (1797) and “The Runenberg” (1802), drew on the chivalric romance for settings, and some of his stories adapted old Germanic legend43 and can be seen to anticipate the romances of William Morris. While the effect of Tieck would seem greatest in the early short tales of Morris, which share with the aforementioned stories a sense of psychological dislocation, elements from Tieck turn up in the late romances as well: for example, the motif of the orphaned/stolen/runaway girl brought up by an enchantress or goddess figure, at the heart of “Eckbert,” turns up in The Wood beyond the World, The Well at the World’s End, and The Water of the Wondrous Isles. Fouqué is now remembered primarily for his fairy tale about the humanization of an elemental water spirit, “Undine” (1814), which prompted an opera by Hoffmann and later elicited from MacDonald the judgment that it was “of all the fairy tales I know . . . the most beautiful” (MacDonald 313). But Fouqué’s “chivalric romances,” most notably the novel-length Magic Ring (1813), were widely read in their day, had an impact on the literary medievalism of the nineteenth century, and can be seen as prototypes of later fantasy. MR, with its expansive canvas, its focus on heroism and good versus evil, and its magical thread, suggests much of the terrain William Morris would later develop, and the medievalism of Phantastes owes much to MR. Nevertheless, Fouqué’s rather simplistic Christian crusader ethos and unreflecting piety, and the consequent lack of psychological depth (so evident in Tieck’s tales), can be seen to undercut his virtues: deft plotting, and skill in evoking atmosphere. Much of this work appeared in English translation piecemeal throughout the Romantic period. Notably, MR appeared translated by Robert Pearse Gillies (1825), and Thomas Carlyle’s four-volume German Romance (1827) included tales by Tieck, Fouqué, Hoffmann, Musäus, and Goethe. The debt of Sara Coleridge’s Phantasmion (1837) to the literary fairy tales of the German Romantics is at least as great as its debt to the English verse and prose discussed here. Despite Mike Ashley’s assertion that Phantasmion was “modeled on Spenser’s The Faerie Queene” (Clute and Grant 210), the present writer sees little connecting the work to Spenser that is not specious, and its neo-Elizabethan strain seems more indebted to Keats. But in his introduction to the 1874 edition, John Duke Coleridge appropriately notes, “Its supernatural beings have no English originals; perhaps indeed they have a German rather than an English character. The legends of Number Nip, and the exquisite fancy of Undine are their nearest prototypes” (Coleridge vi). In addition, the more expansive canvas seems more likely to have been suggested by MR than by Spenser.

ROMANTIC TRANSFORMATIONS

87

Phantasmion is the first work of English prose to fit without qualification Lin Carter’s definition of fantasy, the consolidating point of reference for the BAFS and effective “glue” for the subsequent fantasy genre.44 Coleridge’s romance follows the growth and coming of age of its eponymous hero; his travels in the imaginary kingdoms of Palmland, Rockland, and Almaterra; and his struggles through political and magical intrigue to ensure throne and marriage. The quest, the themes of love and of war and diplomacy, the landscape peopled with supernatural agents both beneficent and hostile, and the magical tutelary figure of Potentilla are all elements common in the traditional literatures appearing in print during the decades preceding Phantasmion and also in subsequent fantasy. Though Coleridge’s more elaborate, expansive use of them suggests a stronger link to romance than to the traditional fairy tale, a la Grimm, they are also common to the fairy tale. Phantasmion’s place in the development of modern fantasy, however, has been somewhat ambiguous. In both style and content, the romance directly prefigures much later fantasy. It is the first work to sever even nominal ties to actual geography:45 the milieu of Phantasmion is an unambiguous invention of Coleridge’s. Even within Carter’s somewhat arbitrary contours, it is not The Wood beyond the World, or even Phantastes, but Phantasmion that is the first invented-world fantasy. On the other hand, the book has languished in almost complete obscurity since its original publication. The original edition, published without its author’s name and limited to 250 copies, drew little attention, and the 1874 edition does not seem to have attracted much more.46 Most critics of fantasy seem to have been unaware of it: one would assume that Carter would have pushed for its inclusion in the BAFS if he had known of it; at any rate, he never mentions it. Where it has been mentioned, it has often been misrepresented.47 David Hartwell excerpted the opening three chapters in Masterpieces of Fantasy and Enchantment (1988), and an expensive library edition appeared in 1994, but the work has been otherwise unavailable since 1874. That second edition makes it tempting to suppose that Morris and MacDonald read it, but I have come across no evidence to that effect. Though Phantasmion does stand as prototype for much later fantasy, it is impossible to ascribe any direct influence to it at all. Despite this, the work does stand at an important juncture in the development of the BAFS template, and it remains, if a lesser work than the best of Morris or MacDonald,48 a well-conceived and readable story. It serves as a suitable closing point for the present chapter.

88

THE EVOLUTION OF MODERN FANTASY

Works Cited Byron, Lord. Byron’s Poetical Works. New York: Thomas Y. Crowell and Co., n.d. (c.1900). Clute, John, and John Grant, eds. The Encyclopedia of Fantasy. New York: St. Martin’s Press, 1997. Coleridge, Sara. Phantasmion. Boston: Roberts Brothers, 1874. [Disraeli, Benjamin] Earl of Beaconsfield. Alroy, Ixion. London: Longmans Green and Co., 1919. Gardner, Martin. Coleridge: The Annotated Ancient Mariner. New York: Bramhall House, 1965. Hogg, James. The Three Perils of Man. Edinburgh: Edinburgh UP, 2012. Keats, John. Keats’s Poetical Works. New York: Oxford UP, 1920. Landor, Walter Savage. Poems. London: Centaur Press, 1964. MacDonald, George. A Dish of Orts. Whitethorn, CA: Johannesen, 1996. Merriman, James Douglas. The Flower of Kings. Lawrence: UP of Kansas, 1973. Scott, Walter. Poetical Works. New York: Thomas Y. Crowell and Co., 1894. Shelley, Percy Bysshe. The Poetical Works of Percy Bysshe Shelley. Volume 1. New York: James Miller, n.d. (c.1860). Southey, Robert. Southey’s Poems. New York: Oxford UP, 1909. Woods, George, ed. English Poetry and Prose of the Romantic Movement. New York: Scott Foresman and Co., 1929.

Other Secondary Work Consulted Gifford, Douglas. Introduction and Notes in Hogg, James. The Three Perils of Man. Totowa, NJ: Rowman and Littlefield, 1973.

Some Primary Source Editions The editorial apparatus in the following primary source editions was consulted. Fouqué, Baron de la Motte. The Magic Ring. Kansas City: Valancourt Books, 2010. Peacock, Thomas Love. The Misfortunes of Elphin. New York: Wildside Press, 2004. Ryder, Frank G., and Robert M. Browning, eds. German Literary Fairy Tales. New York: Continuum, 1983. Stableford, Brian, ed. The Dedalus Book of British Fantasy. Sawtry Cambs: Dedalus, 1991. The Stableford anthology includes short fiction by Sterling, Disraeli, and BulwerLytton; the Ryder/Browning anthology includes tales by Tieck, Novalis, Goethe, Hoffmann, and others.

ROMANTIC TRANSFORMATIONS

89

Other titles republished by Edinburgh UP in the Collected Works of James Hogg series, including material pertinent here, include The Mountain Bard (2007), Winter Evening Tales (2002), The Queen’s Wake (2004), Midsummer Night Dreams and Related Poems (2008), Queen Hynde (1998), and The Shepherd’s Calendar (2002). These are the best editions of Hogg’s work. Poems by Coleridge, Keats, Shelley, Scott, and Southey can be found in most collected editions of those poets’ works; editions used here are noted in Works Cited.

4

From Verse to Prose The Victorian Period

Introduction

A

s outlined in Chapter 3, the most significant developments during the Romantic period, leading ultimately to William Morris and the twentieth-century literary writers whose work would compose the core BAFS canon, took place in the poetic medium. Prose works, while not of negligible quality, were few; only the last, Phantasmion, closely approximates the BAFS template, and none, including Phantasmion, can be said to have exerted any noteworthy influence. The Victorian period begins in the shadow of the Romantic period, and after a mild lull during the later 1830s and 1840s, poetic fantasy began to proliferate once more, beginning a stream that would flow until close to the end of the century. But while these poems certainly followed from the work discussed in Chapter 3, there was a notable shift in direction. The narrative constructs were often far more expansive than even the “epics” of Southey. Where Scott invented “traditional” material, or Keats made “4000 lines of one bare circumstance,” the poets of the Victorian period tended to interpret stories—of King Arthur, of Sigurd the Volsung—already substantially developed in their traditional redactions. One detects in this last a certain decrease in the push to invention, a certain stasis, which, while producing many popular poems, perhaps explains the sharp dwindling of such material at the century’s end. The century ends with the prose romances of William Morris, the conventional beginning of the BAFS canon. Prior to Morris, the Victorian “adult” prose fantasy pertinent here followed from that of the Romantic period in that it is fairly thin in quantity and does not consolidate into any marked pattern.

92

THE EVOLUTION OF MODERN FANTASY

The major development in Victorian prose fantasy, however, came in the context of work—loosely categorized as fairy tales—published for children. For reasons that may be in part understandable, “children’s fantasy” was, during the period of the BAFS, considered as a separate phenomenon from “Adult Fantasy.” At the same time, as noted in Chapter 1, “children’s” and “Adult” refer more to audience than form and content, and many writers of children’s fantasy did not (and do not) consider that a projected readership including children excludes adults or that children’s fantasy cannot challenge adult readers. That the three most widely distributed collections of Victorian fairy tales since the BAFS period1 were not published as “children’s books” underlines a certain arbitrariness in such distinctions. Given the pervasive engagement of traditional narrative structures and the frequent settings in invented, magical countries, to marginalize what is a substantial body of work, with a notable degree of cohesion and which strongly influenced the shape of subsequent fantasy, just because it was marketed for children, does not make sense. With regard to the antiquarian researches that were responsible for making the traditional material pertinent to this study available to modern readers for the first time, there were significant developments in the literary context that should be outlined here.2 The texts of the specifically English-language forms of the ballad and metrical romance can be said to have reached a certain critical mass by the beginning of the Victorian period. While further romances and variants of ballads would continue to appear, these were no longer breaking new ground on a broad level but serving to augment and refine the understanding of a body of material already there. In the field of the ballad, Francis James Child’s encyclopedic English and Scottish Popular Ballads (1882–98), comprising 305 ballads in all their known variants, still stands as definitive, though to the creative writer it did not add substantially to what was available fifty years earlier. With the medieval metrical romances, we find a corollary situation: new, more meticulously accurate editions3 of most of the pieces published by Ritson, Scott, Madden, and others appeared throughout the balance of the century, but significant new romances were fewer. Some of the more voluminous prose romances, such as Merlin, or the Early History of King Arthur (1865–98) and the Carolingian Huon of Burdeaux (1882–87), made their first appearances in modern editions, but the vocabulary of prose romance was already to a substantial degree available through Malory or Amadis of Gaul; whatever new Arthurian material turned up, Malory was firmly enshrined as the definitive medieval version of the legends. More significant developments came from outside the Middle and early Modern English matrix. Scholarship in the pre-Norman Germanic

FROM VERSE TO PROSE

93

ancestor of English advanced, and beginning with J. M. Kemble’s translation in 1837,4 Beowulf was cast into modern English more than half a dozen times by the end of the century, including once by William Morris. But apart from Beowulf, only fragments survive of the epic tradition to which it belonged; the pre-Christian mythology is attested to only in fragmented allusions that only comparative work can illuminate. The Nibelungenlied appeared in English translation by W. N. Lettsom (1850), but, as with Anglo-Saxon, pre-Christian and heroic traditions from the Germanic continent were sparse. By far, the most voluminous, and also far less filtered,5 surviving repository of related material was Scandinavian, mainly Icelandic. As noted, epitomes and translations of extracts of eddaic and skaldic verse had begun to appear as far back as the 1760s, but it was not until the Victorian period that a wider array of full translations began to appear, sufficient to convey a degree of definition. The narrative portions of Snorri Sturluson’s Prose Edda appeared in translation by George Dasent (1842); the Poetic Edda by Benjamin Thorpe in his seminal Northern Mythology (1851). These provide the most detailed and developed sources of pre-Christian Northern and Germanic tradition; the latter’s second half contains a sequence of poems on the Volsung legends. Many from the prolific library of Icelandic sagas found their way into English during the Victorian period, but probably the most significant of these here is Volsunga Saga, translated by William Morris and Eirikr Magnusson (1870), the most complete version of the Volsung legend extant.6 This material made itself felt in Victorian work, and it was, of course, of great importance to later fantasy writers, including Tolkien. There were also some significant developments with things “Celtic.” As noted, ancient Welsh poetry, like eddaic and skaldic poetry, had begun to appear in translation as early as the 1760s, and a broad spread of the surviving corpus was available in English a couple of decades prior to Peacock’s Misfortunes of Elphin. However, most of this poetry is allusive rather than narrative and very difficult even for specialists. It was the publication of Lady Charlotte Guest’s translation of The Mabinogion (1838–49) that made available to English readers for the first time nearly all that remains of ancient Welsh narrative and introduced “Celtic” myth and legend.7 Less notably than the Germanic/Scandinavian material, the effect of The Mabinogion was nevertheless felt in some Victorian work, and it became inspiration and source for many subsequent fantasy writers. The corpus of surviving Gaelic traditional narrative is far greater than the Welsh, and much of it considerably older, but perhaps owing to the long history of Irish–English animosity and the stereotyping of both Irish and Scots Highland cultures as barbaric and inferior, this material did not make broad inroads into the English-speaking matrix until the turn of the

94

THE EVOLUTION OF MODERN FANTASY

century and is really an Edwardian phenomenon. Nevertheless, we do see beginnings here: texts and translations began to appear in periodicals as early as the 1850s; P. W. Joyce’s Old Celtic Romances (1879) contained a fair array of tales, including several of the Fenian warriors; much from ancient Gaelic tradition survived in oral form, particularly in western Ireland and the west Highlands; and contemporary tellings of Fenian (and other) stories appeared in collections such as John F. Campbell’s Popular Tales of the West Highlands (1860–62) and Patrick Kennedy’s Legendary Fictions of the Irish Celts (1866). The oral tales were as important in the modern textual, and English, dissemination of Gaelic tradition as the actual ancient texts. However, the impact of this material on Victorian work was slight and, again, awaited the turn of the century. In the wake of British imperial expansion, romances and epics and mythological narratives from all over the globe flooded back to the Englishspeaking world: The Arabian Nights were reedited and retranslated several times throughout the period and were joined by translations from Arabic, Persian, and Turkish, and the Bhagavad Gita was joined by a full translation of its massive parent work, Mahabharata, by K. M. Ganguli (1883–96). In a more dramatic variation on the process that brought the early English, Germanic, and Celtic texts to light, cuneiform tablets and papyri from the ancient Near East—containing, among other things, various bodies of myths and legends reflecting the pre-Islamic traditions of the region— began to be uncovered by archaeologists, and the first, halting translations of Mesopotamian, Egyptian, and Assyrian texts appeared as linguists began the slow process of decoding the long-unspoken languages. Following the aforementioned collections of Campbell and Kennedy, scholars similarly found that ancient tradition was passed on in contemporary oral culture in many regions less touched by encroaching “civilization”: in Europe, the poems comprising the Finnish Kalevala (1835/49) were collected, edited, and adapted to form a unified work by Elias Lönnrot; many of the Russian bylini, oral heroic narrative poems, were also collected.8 From outside the European matrix, similar material from subSaharan Africa, Polynesia, Siberia, and indigenous North America made its way to English readers. A full account of this body of material, which would fill a small library, is unnecessary here. Suffice to say, in terms of work that played direct “sources” to specific modern works, the weight of emphasis falls on the material from Britain and Northern Europe. But more significant than the piecemeal debts is the collective phenomenon itself: the virtual ocean of traditional narrative in its varied forms facilitated comparative studies of much greater sophistication and credibility than was possible a century earlier.9 The necessary syncretism underlying such scholarly work is paralleled

FROM VERSE TO PROSE

95

in the imaginative sphere, not least in the “invented” fantasy worlds that began to proliferate with William Morris’s prose romances. Following from these collections derived from oral traditions, we come, lastly, to the folktale. The seminal collection of the Brothers Grimm, as mentioned in Chapter 3, was published in German in 1812–15, was widely translated and popular throughout Europe,10 and set off a craze for “collecting” that spread to the other nations of Europe and thence to the rest of the world, continuing through the end of the century and beyond. The folktale, like the ballad, tends to brevity and directness in comparison to romance proper—though in terms of structure and content, it has more in common with romance than the ballad, which is often tied to a single dramatic episode.11 Of particular significance here, and tied to the significance of the folktale to the literary fairy tales of the Victorian period, is the mode of publication and target audience for many of such collections. The appearance of the Grimms’ collection coincided with the growing idea among the educated bourgeoisie that such stories were particularly suited to the young, and successive editions, as well as comparisons to manuscripts, show that the Grimms themselves did a good bit of tailoring to suit the projected audience.12 Like the “imitations” of the traditional ballad, and the “poetical romances” and “epics” discussed here, the Victorian literary fairy tales drew amply from tradition for style and substance. They also followed what became the conventional placement of the traditional tales they echoed in the marketplace: they were directed to children. Finally, the proliferation of adaptations for children of the other traditional narrative modes connected to fantasy should be noted here. Retellings for children of the Arthurian legends,13 of classical mythology,14 of The Arabian Nights,15 and so on joined the collections of folk/fairy tales. Indeed, material from all these sources was indiscriminately bunched together in some of the more comprehensive collections, and literary tales were placed side by side with traditional tales with little fanfare.16 They must all be considered part of the same phenomena. This should be emphasized here: it is in the adaptations for children, filtered through a collectivizing lens, that Tolkien and his older contemporaries of the twentieth century would have first encountered the worlds of myth, legend, and romance that inspired their adult work. The balance of this chapter will be broken into four sections: first, I will discuss the relevant narrative poetry of the period; second, I will turn to the literary fairy tale for children. Then I will discuss prose fantasy for adults in two sections, the second of which will be devoted specifically to William Morris.

96

THE EVOLUTION OF MODERN FANTASY

Victorian Verse Fantasy As noted previously, while the poetic fantasies of the Victorian period follow clearly from the Romantic work discussed in Chapter 3, they are marked by two major departures. Length increased notably, and some of the narrative constructs became positively baroque. For subjects, poets began to turn to stories already developed in their traditional redactions: one might say the emphasis turned from “invention,” per se, to “adaptation.”17 Among the complex of possible factors contributing to this are the difficulties posed to many Victorian readers by the actual traditional texts. On the most superficial level, as the English language stretched back beyond the sixteenth century into the more difficult dialects of Middle English and to Anglo-Saxon, it became increasingly difficult for more and more readers. For many, Beowulf was accessible only by way of translation. As one moved into Icelandic, Welsh, Persian, and Sanskrit, direct access was limited to specialists, and most relied on the indirect access of translations. But there were more tangled issues than simply language: what was found once something was translated often conflicted with the strictures of Victorian propriety, particularly its evasive cast in matters of sexual decorum. On the most immediate level, a reader turning, for example, to the Morris/Magnusson translation of Volsunga Saga would find murder on the first page, followed by incest, murder of children, pagan magic, suicide, adultery, more murder of children, and so on. In other words, such traditional material was scarcely proper, if not positively shocking. On a more disinterested level, the texts and scholarly translations themselves were most often the product of “scientific” interest rather than conviction of literary merit, a direct descendant of the Enlightenment attitudes of the eighteenth century. The texts were significant for what evidence might be gleaned pertinent to language, to history, or to “understanding” the “primitive” beliefs of the cultures they derived from, whether of the distant indigenous past or of the remote geographical region (usually) being colonized. The deeply engrained assumption that the pinnacle of pagan, pre-Christian, or even pre-Reformation culture and literary practice was embodied in the works of ancient Greece and Rome resulted in a certain myopia: the accomplishments of Homer and Virgil represented a universal high-water mark, their forms distilling with a certain finality what was “highest” or “best” without the guiding light of (in effect, Protestant) Christian civilization. By contrast, Beowulf, The Mabinogion, The Poetic Edda, Mahabharata, and so on might have “charm” in their parts, but they were themselves, at best, imperfect literature. On the whole, such works were not studied as literature: they might be paraphrased, with “extracts”—the coarse, shocking elements eliminated—in the context of a

FROM VERSE TO PROSE

97

study intended for a popular readership;18 they might be pruned and sanitized for the audience presumed to correspond to the level of the culture they came from (children); or, to return to the present focus, they might provide “rough” material that, elevated by a “civilized” perspective and craft, could be transformed into poetic art. Embodying both the move toward more expansive structures and the move toward adaptation rather than invention, Tennyson’s Idylls of the King is the most famous, in its day and now, of the poems relevant here. Unlike the Romantic period Arthurian poems of Scott and Wordsworth,19 Tennyson followed sources for all but one of the constituent “Idylls”;20 the overall construct does not depart materially from what is found in Malory. The Idylls were written and published at intervals between 1859 and 1885, and the final, complete work appeared in 1886. But Tennyson’s preoccupation with the Arthurian legends extended back to his heavily Keatsian early period of the 1830s,21 manifesting in four poems. The first, originally published in 1832 but extensively revised for his 1842 collection of poems, was “The Lady of Shalott.”22 “Galahad” and “Sir Launcelot and Queen Guinevere: A Fragment,” descriptive rather than narrative, also appeared in the 1842 volume. All three of these poems are cast in rhymed stanzas that lack the high, formal tone of the blank verse of the later Idylls, but they do prefigure Tennyson’s later thematic concerns. The fourth, “Morte d’Arthur,” also included in the 1842 collection, is narrative and follows Malory closely. The most significant of the early Arthurian pieces, it was transplanted almost intact (one line omitted, the frame discarded and replaced by narrative matter leading up to its opening) to the later Idylls. But it was in the 1850s—after he had become, on the death of Wordsworth, poet laureate—that Tennyson began to actively compose, shape, and assemble the poems that would collectively become Idylls of the King. Though frequently hailed in its day as the great “epic” of King Arthur, the national epic of England that had never been written, Idylls is not formally an epic, nor was it intended as one, despite the 12-part division (which was arrived at rather than planned). The cyclic framework of the whole echoes his chief source, Malory, quite plainly, while the kind of architectural design underscoring Milton or Virgil is absent: it is to romance forms that Idylls is indebted for the most part. At the same time, the Idylls, though collectively of considerable length, are sparer and less encyclopedic than Malory’s work. While Malory was less susceptible than, say, Volsunga Saga or Beowulf to accusations of coarseness or crudity, his language (even with spelling modernized) was difficult and his values archaic, and Le Morte D’Arthur is of uneven quality.23 What Tennyson provided, abetted by his status as laureate and ties to the throne, was a thoroughly Victorianized casting of

98

THE EVOLUTION OF MODERN FANTASY

the legends of King Arthur: in the Idylls, the Arthurian cycle served as a transparency through which the ideals of the age could be elevated and the failings critiqued. Arthur, inspired by Prince Albert, becomes a symbol of the ideal virtuous gentleman. The ideal structure evoked is fundamentally conservative, hierarchical, and headed by a divinely sanctioned king. The moral order is likewise conservative: the downfall of Camelot is the result of petty divisiveness and the inability to sustain a collective allegiance to the king, itself largely brought on by Guinevere’s adultery, a breach of chastity and a failure to conform to the expectations attached to her role as wife. Combined with his impeccably skilled and exceptionally readable verse, as well as the natural attraction of his subject, Tennyson’s conservative ideology and conventional morality explains to a large degree why, both in its gradually accumulating installments and in its final form, the Idylls became a bestseller and in fact vied with Malory as the “definitive” King Arthur. It also explains why, while the poem remains quite readable, the image of the groveling queen begging forgiveness of Arthur in the one idyll invented by Tennyson, and the uneasy evasion of sexual love as a subject even when the material would seem to demand it, are apt to strike readers of our time as prudish and psychologically shallow. Despite the phenomenal popularity of the Idylls, the limitations and problems inherent in Tennyson’s interpretation did not wait until the next century to be noticed. The pre-Raphaelite poet Algernon Charles Swinburne objected to Tennyson’s depiction of the Arthurian world in general and particularly decried his alternating evasion of and condemnation of sexual love. He resented the unflattering portrait of Tristram in “The Last Tournament” and Tennyson’s lack of sympathy with what he saw as the tragedy of the lovers. By contrast, Swinburne’s major Arthurian poem, Tristram of Lyonesse (1882), depicts the lovers sympathetically, and their sexuality is treated with forthright approval. The tragedy is heightened by use of the older of the two traditional endings, the story of the sails, rather than the ending Tennyson used, found in Malory, where Tristram is killed by King Mark: the lovers are not punished for sin but are the tragic victims of fate. Some found Tristram obscene, but it was nevertheless quite popular. But while Swinburne’s treatment of the Tristram legend specifically did become more famous than Tennyson’s, his vision of the Arthurian world in general did not supplant Tennyson’s. Partly this may stem from ideological issues, but it also in good part reflects Swinburne’s failings as a poet. These are glaringly evident in Tristram: the pages-long purple passages of description, the sentimentality, the lack of philosophical and psychological depth. That Swinburne failed to offset Tennyson beyond the Tristram legend itself—which is not, after all, a major part of the Idylls as a whole—and

FROM VERSE TO PROSE

99

remains far less widely read today may simply reflect that Tennyson was a much more accomplished poet.24 Both poets’ Arthurian work is important here as forming an integral part of the soil out of which twentieth-century Arthurian fantasy grew.25 Tennyson’s Idylls, particularly, was inextricably bound up in the popular and literary image of the legends during the time in which the major writers of the BAFS canon would have first encountered them. Not that there were any notable attempts to “follow” Tennyson here: Swinburne became the first in a long line to, in effect, write against Tennyson.26 As the lone treatments of ancient Gaelic tradition during the period, the poetic narratives of the Irish poet Samuel Ferguson are significant here. Coming at a time when translations of old Gaelic tales were largely restricted to scholarly journals (to which Ferguson himself contributed), it was through Ferguson’s poetic adaptations that a broader English-reading public encountered the Fenian and Ulster stories.27 However, while his work was on the whole well received critically, it achieved nothing like the popularity of Tennyson or Swinburne. Probably the chief problem was the Irish subject matter: the beginnings of the Irish Renaissance coincided with Ferguson’s death (1886), and while Yeats would declare him “the greatest poet Ireland has produced, because the most central and the most Celtic” (Graves xxxv), the preceding generation of Anglo-Irish readers had been ambivalent to the Irish subject matter, and the many readers in England whose sympathies on the “Irish Question” were decidedly colonial would scarcely be drawn to Cuchulainn, Fergus Mac Roy, and Deirdre as they were to Arthur, Guenevere, and Lancelot, even in a “regular poem.” Nevertheless, this portion of Ferguson’s poetic output, along with Ossian, anticipates “Celtic Fantasy.” Poetry in America found itself attempting to negotiate circumstances that were not particularly conducive to the production of work along the lines of Tennyson, Swinburne, and Ferguson. During a century when the nations of Europe were looking to their respective pasts for “Romantic” traditions that would become the bedrock of patriotic national identities and the core of national literatures, there was a powerful drive in the United States to construct a uniquely American heritage that was not European (more particularly, not British) derived. Given the fact that the beginnings of American tradition were little more distant than Cromwell, the extreme mobility implied by westward movement, dominant religious ideologies that often did not approve of literary endeavor unless it served to articulate Christian doctrine, and the deistic rationalism of the Founding Fathers, it is not surprising that American poets found themselves facing some difficulties. Tennyson had Arthur, Scott had the traditions of the

100

THE EVOLUTION OF MODERN FANTASY

Scots-English borderland, Ferguson had the Red Branch warriors—what did an American poet have? One tack was to poeticize American historical events that had assumed a legendary status, as in Henry Wadsworth Longfellow’s “Paul Revere’s Ride,” which opens the first volume of Tales of a Wayside Inn (1863), or to compose elevated redactions of American domestic legends, such as those included in John Greenleaf Whittier’s Legends of New England (1831) and Longfellow’s Evangeline (1847) and The Courtship of Miles Standish (1856). The results of this curious hybrid of factual history and Romantic elevation in fact owes much to Scott’s most thoroughly historicized narrative poems, and the aforementioned poems of Longfellow were very popular.28 However, given the factual historical basis and resultant dearth of supernatural or magical elements, these poems, like the Rokeby and Marmion of Scott, are really outside the focus of this study. Tales of a Wayside Inn (1863–73) did number among its contents a variety of legendary tales, of European derivation, which drew on the fantastic, and the collection deserves brief mention. Here Longfellow wrestled with the “American” in American literature in a different way: the work’s constituent 22 tales, echoing Chaucer’s Canterbury Tales, are told by a highly international cast of characters, over a period of nights, at the Wayside Inn in Sudbury, Massachusetts. While tales with American subjects are a distinct minority and among the shorter pieces, the rustic domestic setting of the telling passes the non-American content through an American lens, and pride of place is given to the tales on American subjects told by the Landlord: “Paul Revere’s Ride” opens the collection, and “The Rime of Sir Christopher” closes it. Most pertinent in the present context, however, is Longfellow’s earlier The Song of Hiawatha (1855). Rather than “elevating” historical subjects, or turning the rustic parlor of a New England village inn into a crossroads of international storytelling, here Longfellow turned to American Indian oral traditions—no doubt recognizing that, if there was narrative matter indigenous to the North American continent roughly corollary to the Arthurian, Greek, or Norse myths and legends, it would emerge from the traditions of the American Indians. Perhaps spurred by his contemporary James Russell Lowell’s brief “A Chippewa Legend” (1843), Longfellow turned to Henry Schoolcraft’s collection of (mainly) Chippewa legends, Algic Researches (1837),29 for narrative matter. There are some issues attendant on Longfellow’s poetic use of American Indian traditions that should be mentioned. These are directly related to broader cultural and historical factors, a detailed discussion of which would be out of place here. Suffice to say, a history of “contact” circling around land appropriation and attempted cultural eradication rather

FROM VERSE TO PROSE

101

than genuine engagement did not facilitate understanding on the part of the rapidly expanding dominant Anglo-American culture. A mechanical understanding of Native languages and an assumption that verbal texture and sophisticated conception of narrative were qualities uniquely relegated to “civilized” literary endeavor led to a blindness to those qualities and elicited judgments even more acute than those attending Beowulf or Volsunga Saga. The rather bald prose of Algic Researches no doubt reflects these issues. Hiawatha was affected by all this in some odd ways. For example, though Longfellow’s subject and the better part of the substance of his poem is founded on the stories connected to the Chippewa culture hero Manabozho30 in Schoolcraft, Longfellow somehow received the mistaken impression that the Iroquois Hiawatha was the same character. In fact, the mythic Manabozho has no relation at all to Hiawatha, the historical/ legendary founder of the Five Nations Confederacy; the Chippewa are entirely distinct from the Iroquois as a cultural/linguistic entity. (Imagine, for example, a poetic version of the tales of the Norse gods in which Odin is renamed Charlemagne.) But to much of Longfellow’s readership—and Hiawatha proved quite popular—he was transforming “puerile” (to use Scott’s term) material into literary art. Hiawatha is as thoroughly a fantasy as Thalaba the Destroyer and no more “Chippewa” than Lalla Rookh is “Oriental”; in fact, the exotic attraction of the subject must be seen as emanating from the same imaginative space as the Oriental. Longfellow’s verse form resulted from his reading of the 1851 German translation of Lönnrot’s The Kalevala, and narrative matter derived from the Finnish work is patched into the mainly Chippewa content. “Hiawatha,” his Manabozho, embodies little of the trickster of Chippewa tradition and is instead made over into the peculiarly European “Noble Savage,” admirable as a simple Child of Nature but belonging to a time now gone. Hiawatha sold well, if not as well as some of Longfellow’s other poems mentioned here, and all told, of the poets of the period, he probably sold second only to Tennyson. While the degree of influence attributable to Hiawatha in the present context is limited,31 it was nevertheless a widely read part of the fabric of the Victorian era poetry informing the sensibility underlying the literary canon of the BAFS. Like Tennyson’s and Ferguson’s works, Hiawatha, if in a more convoluted fashion, fed on a patriotic feeling, elevating “national” traditions into a refined work of art; like Tennyson’s Idylls and Swinburne’s Tristram, it is large, like a monument. It is in the context of Victorian narrative poetry that we first meet William Morris. His first collection, The Defense of Guenevere and Other Poems (1858), included six Arthurian pieces. However, it is primarily the subject

102

THE EVOLUTION OF MODERN FANTASY

matter that makes these poems relevant in the present context: excepting “Near Avalon,” a brief lyric, they are dramatic in focus, centering on single scenes. Narrative passages, which occur only in the title poem and “King Arthur’s Tomb,” are brief and explicate only immediate context: the broader contours of the Arthurian story emerge only by allusion in the characters’ speeches. The fantastic is largely absent. While Tennyson’s first collection of “Idylls” did not appear until the following year, which would preclude seeing Morris’s poems as a direct response to Tennyson,32 Morris was nevertheless at odds with the ideological substance that drove Tennyson’s interpretation. In the title poem, Guenevere’s “defense” just as she is about to be burned at the stake for adultery, it is obvious Morris had little interest in the question of guilt: Guenevere denies the charges yet voices recollections that are self-incriminating. What emerges, rather than an issue of marital propriety and morality, is the very human picture of the queen trapped in an arranged marriage, unable to experience fulfillment in love or to assert her own identity. While these poems are a bit removed from the present context in terms of form and content, it is not unlikely that Morris’s engagement of character inspired some later, novelistic interpretations of the Arthurian legends. Morris’s 1858 collection provoked some hostile reviews and didn’t sell well, and this probably combined with the commercial success of Tennyson’s first gathering of “Idylls” the next year to prompt him to abandon what had been plans for a more extensive treatment of the Arthurian legends. In his next published work, nearly a decade later, Morris turned to Greek myth in The Life and Death of Jason (1867). The approach was very nearly the inverse of that of the Arthurian poems: emphatically narrative, the dramatic immediacy exchanged for a distanced perspective, the compressed brevity exchanged for a length of more than 10,000 lines. While the choice of classical subject matter suggests continuity with Keats, Morris eschews the neo-Elizabethanism of his Romantic period mentor in favor of a neomedievalism: though Jason is ultimately a Greek legend, Morris drew as much from medieval sources as classical sources.33 The latter tendency continued in the work Jason had been at first intended for, The Earthly Paradise (1868–70). Explicitly echoing Chaucer, EP, like Hogg’s Queen’s Wake and Longfellow’s Tales of a Wayside Inn, is a collection of tales, two for each month of the year, told by characters within a frame. Many of the included tales, and a few others written for the collection but dropped, were based on classical myth. However, the frame of EP places the telling of its tales in a clearly medieval milieu—moving from a vague Europe that suggests a Northern rather than a Mediterranean point of reference34 to the invented geography of the lands encountered by the “Wanderers” in their voyage to find the Earthly Paradise. In execution,

FROM VERSE TO PROSE

103

the “classical” stories are characterized by medieval romance conventions much more than by classical epic ones, and like Jason, they are more often than not indebted to medieval versions for their details. Where Keats Elizabethanized his classical subjects, Morris medievalized them. This last point is underscored by the fact that the classical stories are told side-by-side with medieval stories. “The Man Born to Be King” is a medieval story existing in numerous variants; Morris syncretistically weaves together details from a redaction in Gesta Romanorum, a French romance he would later translate as “The Tale of King Coustans,” and Grimms’ “The Devil with the Three Golden Hairs.” “Ogier the Dane” is founded on a Carolingian romance, trimmed of historical matter and the magical elements foregrounded. Most noteworthy, and pointing to his next and last major poem, are several stories deriving from Germanic and Norse sources: “The Lovers of Gudrun” is based on the core section of Laxdaela Saga; “The Fostering of Aslaug” is based on The Saga of Ragnar Lodbrok; and while it is to a significant degree Morris’s invention, “The Land East of the Sun and West of the Moon” utilizes motifs based on the Swan Maiden tales. These pieces are arguably the finest in the collection, perhaps suggesting that Morris’s imagination was more spurred by Northern and medieval than by classical subjects. While the latter portion of EP was being completed, Morris’s interest in Icelandic was blooming. He met Eirikr Magnusson, under whose tutelage his fledgling knowledge of the language expanded rapidly, leading to their collaborative translations, The Story of Grettir the Strong (1869) and Volsunga Saga (1870). Visits to Iceland in 1871 and 1873 further deepened his interest in its literature, ultimately leading to the composition of The Story of Sigurd and the Fall of the Niblungs (1876). For his source, Morris mainly followed Volsunga Saga, though he incorporated elements from the eddaic poems and used the (less shocking) conclusion of the Nibelungenlied. While the poem was not a commercial failure, it did not repeat the remarkable success of Jason and EP: its narrative matter was no doubt less in line with Victorian sensibilities, and it was bound to stand in the shadow of Richard Wagner’s operatic adaptation of the same legends, The Ring cycle (1853–74). Some reviewers did, however, regard it as a greater achievement than the earlier two poems. In one of those reversals not altogether rare in literary history, Morris’s early poems, poorly received on their appearance, found themselves redeemed with the onset of modernism and are now highly regarded. In contrast, the long narrative poems that established Morris’s critical and popular reputation into the early twentieth century are now virtually unread. Yet these poems are essential for understanding Morris’s development, and they stand as direct antecedents of the prose romances. The

104

THE EVOLUTION OF MODERN FANTASY

settings, while still at least theoretically on the map (excepting the Prologue of EP), are prototypes of what we find in the prose romances; we find the same dramatic distance and archaistic idiom; we see Morris’s syncretistic capacity to bridge romance and saga (and to some degree epic) narrative techniques into one. Three of the late romances were first assayed as verse tales; longer speeches in House of the Wolfings are in verse: certainly Morris saw continuity with the earlier poetic narratives. The prose romances, important as they are to fantasy, do not simply appear out of thin air in the contexts of either the author’s development or literary history at large. These verse fantasies of Morris, like narrative poetry in general, have not been considered significant in the history of modern fantasy. Yet they were common reading during the youth of the writers of the BAFS canon. In “On Stories,” C. S. Lewis lists the “Prologue” to EP and Jason in company with The Worm Ouroboros and Vathek (Lewis 88), and there are unmistakable echoes of Jason in The Voyage of the Dawn Treader. When Tolkien expressed the wish to write something “on the lines of Morris’s romances,” those romances would have included the present poems, and the frame of the Book of Lost Tales (Eriol hearing tales after a voyage to magical islands over the western sea) recalls that of EP. But after these poems of Morris, Tennyson, Swinburne, Longfellow, and Ferguson,35 the type of romance narrative associated with the BAFS migrates almost exclusively to prose. There is not a clear-cut reason for this, but some speculation is possible. In the move from invention to adaptation, a certain static quality emerges. Ideologically, one can see in this material a retreat from the revolutionary intentions of many of the Romantics: the desire to combat the forces of Enlightenment rationalism and growing utilitarianism and the notion that the poetic imagination could form the foundation of a rejuvenated humanity. Instead, we begin to find a tendency to nostalgic lament: the poet becomes the “idle singer of an empty day” (1Morris 1:1). This ideological retreat bleeds clearly into the thematic substance of many of these poems. When, at the closing of “The Passing of the King,” Tennyson’s Arthur declares to Bedivere, “The old order changeth, yielding place to the new” (Tennyson 251), the transition is from an order characterized by dedication to the ideals of the Round Table to an “order” marked by infighting, betrayal, self-motivation, and chaos. The gradual erosion of Camelot and what it symbolizes is the consolidating substance of the Idylls as a whole, a dark apocalyptic current lending a collective power to the poems. Yet Tennyson’s diagnosis of the fall of Camelot reveals simply a breach of conventional Victorian morality and propriety: it is the queen’s sin of adultery that precipitates the fall into chaos and therefore the queen who stands primarily responsible for Arthur’s failure. A few lines

FROM VERSE TO PROSE

105

later, Arthur asks Bedivere, “[W]hat comfort is in me?” (251), a shrug of defeat suggesting that the glory of Camelot is now part of a past that has no bearing on the present. But even in Swinburne’s attempt to counter Tennyson’s rigid conventional moralism and prudish evasion of the erotic, a certain stasis is evident: the younger poet’s lack of psychological acumen and philosophical depth leaves his great theme awash in sentimentality and overelaborate description. The Morris of Jason and EP is more diffuse and as fatalistic as Tennyson, and it is no surprise that both poems were written during the period of Morris’s greatest social and political ambivalence. EP opens, Of Heaven or Hell I have no power to sing I cannot ease the burden of your fears, Or make quick-coming death a little thing, Or bring again the pleasure of past years, Nor for my words shall ye forget your tears, Or hope again for aught that I can say, The idle singer of an empty day. (1Morris 1:1)

The quest of the Wanderers for unending life is unfulfilled, and the narrator likens their failure to his: “[T]hese poor tale-tellers, who strove / In their wild way the heart of Death to move, / E’en as we singers, and failed, e’en as we” (1Morris 4:437). The most our “idle singer” can accomplish is to bring back a “fragrance of old days” to “folk aweary” (1Morris 4:442). The predominant mood is one of rather powerless nostalgia. The thematic substance running through these poems does not really encourage further development. And then there is the bulk: how many book-length or multivolume36 elegies to lost ages can be produced before there are enough? The decline in the production of such poetic work at the close of the nineteenth century was a harbinger of its rejection in the twentieth, particularly in the wake of World War One and the attendant rise of modernism. When the sensibility that bred these poems faded, they too faded: though the Idylls did not disappear, they underwent a critical demotion; Morris’s long poems were erased from the canon of English poetry; and it was Walt Whitman and Emily Dickinson, not Longfellow, who would come to be seen as the major forces shaping an American poetry. Arthurian and classical subjects treated according to Victorian notions of epic/ heroic poetry found a ready contemporary audience, but those notions reflected a peculiarly Victorian aesthetic—an aesthetic that came to seem irrelevant after the disaster of World War One. The long verse fantasies of the Victorian period are less vital than “Lamia” or “Christabel” or The

106

THE EVOLUTION OF MODERN FANTASY

Queen’s Wake; they are also less vital than the works they were perceived to be “elevating”: Malory, Volsunga Saga, or many of the genuinely Chippewa tales of Manabozho. But despite the failings of much of this work, and despite the fact that it represents a rather static endpoint for poetic fantasy narrative in English, it was a little too glibly dismissed by the modernist knee-jerk. While the present writer does not expect EP or Hiawatha to suddenly appear at the top of the canon of English poetry, they do deserve to be read. And again, to collectively omit Victorian narrative poetry in the present context is to ignore an important part of the fabric out of which twentieth-century fantasy grew. Victorian Fairy Tales for Children The dimensions of the “Victorian Fairy Tale” phenomenon are sufficiently large that any attempt at a comprehensive or inclusive overview would extend this section far out of proportion to its place in the study. In the following pages, then, I will attempt to sketch some of the major lines of development, with attention to some of the more significant authors and works. The fairy tale had had little presence in the English-speaking literary world during the period that saw its development first in France and then in Germany.37 The utilitarian ideals that pervaded both the intellectual and the predominantly Calvinist religious spheres saw it as an affront to Enlightenment reason as well as dangerously amoral—inappropriate fare for both child and adult. While certain structural elements of the fairy tale may be discerned in such novels as Richardson’s Pamela, and fairy elements appear in much of the poetry discussed in the preceding chapters, the primary English “Faerie” heritage was largely restricted to Elizabethan works such as A Midsummer Night’s Dream and The Faerie Queene. The circulation of fairy tales themselves was predominantly in the medium of chapbooks read mainly by the lower classes. But this view underwent notable revision during the opening decades of the nineteenth century. A reaction against what was viewed as the stultifying nature of an education founded solely on Gradgrindian “Facts” fostered the notion that the imaginative apparatus of the fairy tale could be brought to serve constructive ends: to instill sound moral principles and encourage behavior traits in keeping with the prevalent notions of reasoned responsibility. Children were, due to lack of experience and sophistication, assumed to be naturally attracted to the marvelous, and that attraction might be manipulated so as to help form responsible, moral adults.38

FROM VERSE TO PROSE

107

The initial catalyst behind the fairy-tale publishing phenomenon came with the appearance of Edgar Taylor’s translation of Grimm stories, German Popular Tales (1823). The Grimm tales, combined with selections from the contes de fees of Perrault, d’Aulnoy, and others, and some influx from The Arabian Nights and the few actual English tales, came to determine the initial template for the fairy tale until the middle of the century. One strand in the development of the fairy tale came directly out of this work: adaptations and retellings of “standard” tales. Many of these are of greater sociological than literary interest39 and amount to little more than reductive simplifications with a forced moral imposed. Some were more creative engagements, assuming an interest in their own right.40 From here, we move to tales that are “invented” but still modeled closely on traditional and “classic” tales, from which motifs and plot material are freely appropriated. One of the most sophisticated examples of this is also probably the first major Victorian “literary” tale.41 John Ruskin’s The King of the Golden River was written in 1841, though not published until 1850. It is a good example of the way traditional elements were appropriated in the course of creating an invented tale: the core dramatic focus, centered on the two greedy “Black Brothers,” Schwarz and Hans, and their generous (and therefore abused) younger brother, Gluck, embodies a common folktale pattern but is in its details and specific development an original invention. The successive quests of the two older brothers, both unsuccessful due to their selfish and arrogant stinginess, and that of Gluck, successful due to his generosity and self-sacrifice, for the eponymous “Golden River” also embodies a common folktale motif, though it is clear that Ruskin drew particular details from the Grimms’ “Golden Goose” and “Water of Life.” The clear, though unobtrusive critique42 of the values of Victorian utilitarianism evinces the vision of Ruskin and provides a uniquely literary cohesion to the tale on a philosophical and ethical level. This kind of mesh of traditional form and originality informed many of the short tales throughout the remainder of the century. The more noteworthy among these include the Irish writer Fanny Browne’s Granny’s Wonderful Chair (1857), a series of fairy tales told by a magical chair in a fairy-tale frame. The tales of Mary de Morgan—like Ruskin’s, often offering some barbed reflections on Victorian ideology—appeared in On a Pincushion (1877), Princess Fiorimonde (1880), and The Windfairies (1900). The American illustrator Howard Pyle wrote (and illustrated) three collections of readable, if workmanlike, fairy tales: Pepper and Salt (1885), The Wonder Clock (1888), and Twilight Land (1894). With the first English edition of the tales of the Danish writer Hans Christian Andersen in 1846, some different approaches to the fairy tale were suggested. Often misleadingly paired with the Grimms, only a handful

108

THE EVOLUTION OF MODERN FANTASY

of Andersen’s more than one hundred tales actually derive from Danish folklore, and it is to the German Romantic writers of kunstmärchen, and particularly Hoffmann, that his work is allied. Unlike the more anarchic and imbalanced Hoffmann, however, Andersen’s simpler narrative structures were more amenable to the Victorian fairy-tale sensibility, and his moral concerns suited that sensibility perfectly.43 Even so, Andersen’s tales did not follow the fairy-tale template derived from Grimm and the contes de fees closely, and some of the most powerful of his tales, such as “The Snow Queen,” move into mythopoeic territory that doesn’t easily reduce to moral/allegorical readings. Andersen’s work, driven by a fertile, wonder-evoking imagination, certainly served to suggest to English language writers ways to push the fairy-tale template beyond its bounds formally. The Scots writer George MacDonald was one of these, though probably Andersen was most significant only for the suggestion; the influence of the German Romantic writers on MacDonald, in particular Novalis, Hoffmann, and Fouqué, is far more readily apparent. Some of the shortest of MacDonald’s fairy tales, including “The Giant’s Heart” (1863) and “Little Daylight” (1871), do not stray appreciably from the pre-Andersen template; conversely, the shortest tales include “The Shadows” (1864), which does seem directly influenced by Andersen. It is possible that Andersen’s “Snow Queen,” not novel length but long enough to stand as a short volume in its own right, suggested the use of a larger narrative canvass.44 Nevertheless, “The Light Princess” (1864) evinces little of Andersen, but a lot of Hoffmann, and seems to have been inspired by “The History of Krakatuk,” which it directly echoes at points. “The Golden Key” (1867) moves into the mythopoetic territory occupied by the kuntsmärchen of Novalis. The presence of Andersen can perhaps be felt in “The History of Photogen and Nycteris”45 (1879), though the presence of Novalis still seems stronger. These tales stand among the greatest accomplishments in the Victorian literary fairy tale.46 Elsewhere, Andersen is simultaneously invoked and subverted in Oscar Wilde’s haunting, poetic tales, collected in The Happy Prince and Other Tales (1888) and The House of Pomegranates (1891). The tales comprising Bram Stoker’s early Under the Sunset (1881) also seem to fall in the stream of Andersen’s influence. The fairy tales of the American writer Frank R. Stockton,47 in substance and tone quite distinct from Andersen’s on the whole, do follow the Danish writer in their invention and originality. There were also ventures into “long form” throughout the period, and the use of the larger canvass enabled a more novelistic development of character. It also enabled greater depth in the depiction of invented settings, beginning to merge into something akin to the “world making” associated with the BAFS.

FROM VERSE TO PROSE

109

One of the first of these was William Makepeace Thackeray’s “Fireside Pantomime,” The Rose and the Ring (1855). The setting alternates between two kingdoms, which (though by implication contiguous with actual geography) are invented, and the story follows the intrigues attending the successions, and ultimate marriage, of Prince Giglio of Paflagonia and Princess Rosalba of Crim Tartary, whose thrones were usurped in childhood. Thackeray’s fairy-tale vocabulary is firmly rooted in the classic French contes de fees, with a bit of Andersen, but moves unambiguously in the direction of irony and burlesque, and it is ultimately rather reminiscent in its narrator persona of Hoffmann’s “Krakatuk,” which Thackeray had in fact translated. Thackeray’s work had its successors. Among them are Tom Hood’s Petsetilla’s Posy (1870), whose narrator persona, invented kingdom of Aphania, and satiric subtext clearly derive from The Rose and the Ring. More well-known than Hood’s work are the three chronicles of the kingdom of Pantouflia by Andrew Lang, who considered The Rose and the Ring his favorite fairy tale (Avery 10). The first two volumes, Prince Prigio (1889) and Prince Ricardo (1893), often combined in a single volume, are as or better known than Thackeray’s work48 but very much in its ironic mode, though the thread of political satire is perhaps more muted. The third of the chronicles, Tales of a Fairy Court (1907), which followed after the turn of the century, is less well-known than its predecessors and does not seem to have ever been combined with them in a single volume. Two of George MacDonald’s three book-length children’s fantasies follow here, though MacDonald’s work moves decidedly away from the tongue-in-cheek of Thackeray. In The Princess and the Goblin (1870– 71/1872) and The Princess and Curdie (1883), the setting (identified as the kingdom of Gwyntystorm in the second book) is invented, and unlike the magical world in “The Golden Key” or those of Phantastes or Lilith, it is a representational milieu, and the adventures, while accorded a penetrating psychological and symbolic depth, are external. The core dramas of each tale—the discovery and foiling of the conspiracy of the goblins to kidnap Princess Irene in the first, the defeat of the conspiracy by corrupt ministers to assume the power of Irene’s father in the second, both overseen by the magical figure of the princess’s great-great-grandmother Irene—build directly on traditional fairy-tale and romance motifs; in varying ways, the Curdie books can be seen behind much later children’s fantasy, not least The Hobbit and The Chronicles of Narnia. These longer pieces, particularly, begin to look forward to twentiethcentury fantasy. The larger canvass facilitated more developed characterization and more intricate story structures. It also facilitated more detailed development of setting: while these magical imaginary countries are rather

110

THE EVOLUTION OF MODERN FANTASY

tentative, they do anticipate not only Oz and Narnia but Dorimare, Erl, and Middle-earth. The bulk of Victorian fairy-tale work, short or long, falls more or less within the terrain initiated by, on the one hand, the Grimms and the contes de fees and, on the other, Hans Christian Andersen. There were a few fantasies that moved outside the territory suggested by these continental tales, however. Oddly, few attempts were made to engage specifically indigenous traditions. Even George MacDonald, who would center some dozen of his Victorian three-deckers in Scotland, set only one fairy tale, “The Carasoyn” (1866/71), in his homeland.49 Two other exceptional works engage Scottish tradition directly. Dinah Mulock Craik’s Alice Learmont (1852) draws on the lore of lowland Scotland, centering on the abduction by fairies of its eponymous heroine, and incorporates the legend of Thomas Rhymer into its dialect-studded pages. Perhaps Craik’s finest fairy tale, it was less successful and remains less well-known than its author’s blander, more conventional The Little Lame Prince (1874) and Adventures of a Brownie (1872). Andrew Lang’s The Gold of Fairnilee (1888) likewise draws on borderlands fairy lore and the theme of fairy abduction, though its more precise, historically based setting recalls more than Alice Learmont the world of Scott’s Minstrelsy and some of the work of James Hogg.50 Like Alice Learmont, it is perhaps its author’s finest fairy tale, though The Chronicles of Pantouflia were more successful and remain Lang’s best known work. In both cases, the specific preoccupation with local folklore, wedded to a darker, more subdued tone, serves to distinguish these tales from the majority of Victorian fairy tales; both qualities perhaps account for their meager commercial success. Finally, there are a few idiosyncratic works that, though peripheral in the present context, should be mentioned here. Charles Kingsley’s The Water Babies (1863) is one. Though the magical part of the chimney sweep Tom’s adventures are initiated by a fairy, little else in the story draws specifically on the fairy-tale vocabulary common to the stories previously discussed. Tom’s journey in fact takes place after death, and the incorporation of information lists, extended authorial asides, and its strand of the realistic all combine to create a chaotic formal amalgam.51 George MacDonald’s third book-length children’s fantasy, At the Back of the North Wind (1868–69/1871), is set in Victorian London and centers on the encounters of Little Diamond, the ill son of a London cabbie, with the magical North Wind. These encounters are, however, marked by ambiguity, and a reader could take them as the product of Diamond’s illness; nor do they divert from the London setting. Unlike the Curdie books, North Wind has been inherently limited in its influence. In his excellent biography of

FROM VERSE TO PROSE

111

MacDonald, William Raeper speculates that “the double story [of Sylvie and Bruno] may have been suggested by Diamond’s adventures” (Raeper 177), though if this is the case, Lewis Carroll thoroughly appropriated the idea to his own ends. The garden at the Moon-Angel’s back, which David travels to in Howard Pyle’s book-length The Garden behind the Moon (1895), seems likely to have been suggested by North Wind, though otherwise Pyle’s work seems more aligned with Andersen and MacDonald’s other fairy-tale work. But the most abidingly popular departures from the conventional fairy-tale template in Victorian children’s fantasy are Lewis Carroll’s Alice’s Adventures in Wonderland (1865) and Through the Looking Glass (1871).52 Carroll’s subversive combination of the journey to invented worlds, dream logic, nonsense, Mother Goose, fairy-tale elements, and mockery of Victorian social norms amounts to books that cannot properly be called “fairy tales.” Nevertheless, the two books clearly targeted the children’s fairy-tale market, have remained popularly attached to the genre since their publication, and for that reason warrant mention here. However, the works are really sui generis: Carroll’s influence can be discerned to some degree in, for example, Jean Ingelow’s Mopsa the Fairy (1869) and Maggie Browne’s Wanted—A King (1890), but neither of those works come near to Carroll’s thoroughgoing extension of the logical into nonsense or his command of dream fluidity. Both, ultimately, have more in common with the more conventional fairy tales discussed here than with the Alice books—no one has ever written another Alice. Though Carroll’s two later fantasies, Sylvie and Bruno (1889) and Sylvie and Bruno Concluded (1893), have some similarities to the Alice books, they were highly unsuccessful on their publication, have been alternately extolled or decried ever since, and are far less accessible and less well-known than their predecessors. The stream of fairy tales for young readers continued unbroken into the twentieth century, and it will be taken up again at the beginning of Chapter 5. The importance of this work to the genre canonized in the BAFS should be clear enough. The play with traditional narrative forms—here the folk fairy tale—which is at the heart of the work on the literary side of the BAFS canon, is also at the heart of the Victorian fairy tale. We see in this work the invented magical country becoming increasingly common. In the frequent use of motifs such as the quest, diplomacy and war, and curses and intimations of destiny, we see something like the BAFS vocabulary emerging. The exclusion, or at least sidestepping, of the Victorian fairy tale in accounts of the history of fantasy from the BAFS period, and to some degree still, is largely attributable to its projected audience.53 But again, this proves rather slippery: collections of Victorian fairy tales, originally

112

THE EVOLUTION OF MODERN FANTASY

published for children, are now found in the adult sections of bookstores. The BAFS included a collection of three of George MacDonald’s tales originally published for children, editor Lin Carter insisting they were not “really” for children.54 Even in their time, we find three of MacDonald’s children’s fairy tales making their first book appearance in Adela Cathcart (1864), a three-decker for adults; Stockton’s The Queen’s Museum appears to be intended for an adult audience. Exactly what “writing for children” meant to the various authors leads into equally murky territory. Many Victorian fairy tales were clearly written because there was a market for them, by writers who were not inherently interested in their adopted literary form and either not really interested in their audience or hampered by a view “of children as a special kind of creature, almost a different race, rather than as normal, if immature, members of a particular family, and of the human family at large” (Tolkien 130). Such material, it might be said, was written more at than to children. In such cases, the results were frequently not terribly well realized fantasy coating a not terribly profound didactic message. These are not, on the whole, the Victorian fairy tales that have maintained a readership among children or adults. Other, more sophisticated writers, who were clearly more enthusiastic about their adopted literary form (and hence apt to have some genuine command of it), did seek to engage adult readers in the form itself. Different writers negotiated this in different ways. Sometimes a satirical subtext and tongue-in-cheek narrative voice was used as a means to engage the adults: this is readily evident in The Rose and the Ring and The Chronicles of Pantouflia. However, some of the patronizing qualities of work written at children are evident here: often in both works, it is the magic and fairy-tale conventions themselves—the specific elements that Victorian adults would have associated with children—that are treated in tongue-in-cheek fashion, while satiric insinuations and asides seem to be winks directed more exclusively to the adults.55 An uncomfortable tension can emerge from this, and certainly the feeling that the child part of the audience was not taken altogether seriously. George MacDonald’s view was more integrated: to MacDonald, the fairy tale was a transparency whose dynamic nature facilitated different meanings according to the perspective and nature of the reader. Writing “for children,” by implication, simply meant that a level of clarity and directness needed to be maintained to ensure accessibility—not that what was written was different than what might be written for adults. That MacDonald’s work for children is clearly intended to be accessible to children does not mean that he saw its proper audience in any way restricted to children.56

FROM VERSE TO PROSE

113

But regardless of these issues, the time period of the Victorian fairy tale crosses with the childhoods of Dunsany, Cabell, Kenneth Morris, Eddison, Lewis, and Tolkien, and it is fair to assume (and in some cases attested to) that some cross section of the aforementioned work formed part of their childhood reading. It is, needless to say, more than possible that a childhood encounter with an invented world in a children’s fairy tale could plant the seed for the creation of an invented world for an “adult fantasy.” In addition, this material would have been linked from an early age, through the medium of adaptations for children, to the other strands of tradition discussed here apart from the fairy tale: Arthurian romance, Celtic legend, the Greek and Norse myths, and so on. It was as a “fairy tale,” in Andrew Lang’s Red Fairy Book (1893), that Tolkien first encountered the tale of Sigurd. Victorian Fantasy for Adults Until William Morris began publishing his prose romances in the late 1880s, Victorian prose fantasy amenable to the BAFS template and not published for children was very thin in quantity, with no more generic cohesion than the prose of the Romantic period.57 George Meredith’s first work of fiction, The Shaving of Shagpat (1855), was a return to the quasi-Oriental tale in prose (albeit interspersed with a large quantity of incidental verse), though it did not follow its eighteenthcentury forbears closely. It is more than twice the length of any of its prose predecessors, unless one considers Vathek and its Episodes or Tales of the Genii as single works. Correspondingly, with its weave of interspersed tales of various genres58 narrated by its characters, and the constant incursion of spoken verse, Shagpat embodies some of the labyrinthine structure of The Arabian Nights in a more overt, developed manner than the eighteenthcentury work. Its prose is more elaborately ornate and more baroque in its textures and idiom than its prose antecedents, and it may be said to appropriate some of the sensibility underlying much Romantic narrative verse fantasy. While, say, the inset “Story of Bhanavar the Beautiful” bears some affinity to the work of William Beckford, particularly in its demonic elements, the mannered irony that characterizes the central narrative is more akin to Thomas Love Peacock than Beckford. Shagpat did elicit some positive responses on its publication, including an enthusiastic review by George Eliot, but it also elicited hostility, and it sold very poorly. Farina (1857), a far less successful fantasy in a Romantic German setting, sold worse, and Meredith left fantasy altogether.59 Shagpat is not unflawed: the main narrative becomes dissipated among a welter of extended magical scenes in its latter pages; the conclusion is less successful

114

THE EVOLUTION OF MODERN FANTASY

than earlier sections. Its high points, however, are remarkably well realized, and at least the opening half of the book is consistently maintained. The textured, ornate, artificial prose idiom never flags. Shagpat’s place in the history of fantasy is somewhat checkered. On the one hand, it cannot be said to have generated any followers beyond a possible influence on the equally baroque indirection and mannered irony of Ernest Bramah’s Kai Lung books. However, it was reprinted in the BAFS, and Lin Carter often cited it, along with Vathek, as one of the logical forerunners of William Morris.60 Certainly, beyond its magical quest structure, the textures and imagery of Shagpat’s descriptive passages anticipate, without supposing direct influence, those of some of the canonical BAFS writers. Compare this to the following passage from The Worm Ouroboros: It was broad light when they had passed the glade and the covert of the wood. Before them between great mountains, glimmered a space of rolling grass fed to deep greenness by many brooks. The shadow of a mountain was over it, and one slant of the rising sun, down a glade of the mountain, touched by the green tent of the Emir, where it stood a little apart from the others of the tribe. Goats and asses of the tribe were pasturing in the quiet, but save them nothing moved among the tents, and it was deep peacefulness. (Meredith 37–38) The turf was like a cushion, a place for elves to dance in. The far bank full half a mile away was wooded to the water with silver birches, dainty as mountain nymphs, their limbs gleaming through the twilight, their reflections quivering in the depths of the mighty river. In the high air day lingered yet, a faint warmth tingeing the great outlines of the mountains, and westward up the river the young moon stooped above the trees. (Eddison 166)

The only other book-length quasi-Oriental fantasy to appear during the Victorian period was Khaled: A Tale of Arabia (1891), by the American writer F. Marion Crawford. Well written and still readable, if slighter in substance than Shagpat, the book was included in the BAFS. There is a small spread of shorter work from the period of some interest here. Dinah Mulock Craik published some short work for adults early in her career, collected in three volumes as Avillion and Other Tales (1853). Among the contents are “Erotion: A Tale of Ancient Greece,” which draws, as its subtitle suggests, on Greek mythological material, and the eponymous novella, “Avillion; or the Happy Isles,” an idiosyncratic visionary narrative in which the protagonist encounters Ulysses and a number of Arthurian characters in what resembles a near-death experience. The first edition of Richard Garnett’s The Twilight of the Gods appeared in 1888, the

FROM VERSE TO PROSE

115

stories of which rove through various bodies of world myth, legend, and history, anticipating, though with a very different imaginative temperament, the short stories of Kenneth Morris.61 The remaining major works here are George MacDonald’s two visionary romances published for adults: Phantastes: A Faerie Romance (1858), the author’s first book-length prose fiction, and Lilith: A Romance (1895), his last major work of fiction. A few short fantasy tales appeared early in MacDonald’s career,62 but far fewer than his output directed to young readers; it is on Phantastes and Lilith, both of which appeared in the BAFS, that MacDonald’s “adult fantasy” reputation rests. Insofar as Lilith is contemporary with Morris’s romances, MacDonald’s status as a precursor rests primarily on Phantastes. On a quite literal level, Phantastes does in fact embody all the key elements that form what I have called the BAFS template. The book’s Fairyland is an imaginary world, in which Anodos the protagonist encounters magical beings and spells. With its castles, rustic homesteads, dark forests, and knights; its journey structure; and its rather pre-Raphaelite medieval atmosphere, the vocabulary of romance and fairy tale is marshaled unambiguously. In terms of MacDonald’s own inspirations and “sources,” this vocabulary to some degree relies on elements from actual medieval, and particularly Arthurian, romance; however, Spenser and much else from the Elizabethan period and the seventeenth century inform MacDonald’s treatment of his medieval vocabulary. The traditional fairy tale is echoed in many of the book’s motifs63 and is explicitly alluded to in places; specifically Scottish patterns underscore the battle of Anodos and the two brothers against the three giants. The inset ballad of Sir Aglovaile is clearly modeled on the border ballads. MacDonald was deep in Romanticism, both English and German, and in varying capacities, Shelley, Coleridge, Novalis, Hoffmann, Fouqué, and many more can be scented behind sections of Phantastes.64 Carter’s note of Phantastes as a possible precursor of Morris’s romances can be taken as conformation that he recognized these factors. Nevertheless, he ultimately exempts it (and Lilith) from consideration in the following terms: “[S]uch novels . . . are set in Dreamland: we are not expected to take their imagined landscapes seriously; over and over again MacDonald reiterates that his heroes are wandering through a waking dream. Thus, if we accept MacDonald as the founder of the heroic fantasy laid in invented worlds, we would also have to admit Alice’s Adventures in Wonderland to the fold . . . And, obviously, that would never do” (1Carter 10). On the one hand, Carter’s articulation of the issues is somewhat simplistic and misleading and show a limited grasp of MacDonald’s practice. Both Alice books are dreams in a quite straightforward sense: Alice falls asleep, has a dream, and wakes up at the end. This is not the case in either of MacDonald’s

116

THE EVOLUTION OF MODERN FANTASY

romances, where the invented worlds would be more aptly described as alternate dimensions outside the continuum of the five senses. Precisely the relation between the primary and the invented worlds is ambiguous, but the latter are not the product of a simple sleeping dream. Nevertheless, there is a valid perception underlying Carter’s contention. Dream or not, the geography of Fairyland is certainly fluid and dreamlike; an attempt to map it would be meaningless. There is no attempt at depicting a representational historical verisimilitude. The landscape functions as an externalized representation of the inner state of Anodos. Though the work draws on romance and the fairy tale for its vocabulary, it belongs more to the tradition of Langland and Dante than that of Malory and Chretien de Troyes. Among the work discussed in this study, its closest cousins are Shelley’s Alastor, Darley’s Nepenthe, and Blake’s Prophetic Books. In fact, Phantastes is more aptly read along symbolist poetic lines than as a cohesive dramatic tale—focus on the latter has led to frequent judgment of the work as disorganized and random rather than as a work of notable cohesion. Like Shagpat, Phantastes was a commercial failure, and though it found some admirers,65 the critical reception was largely marked by bemusement and some hostility. Like Meredith, MacDonald produced one more, minor fantastic work for adults, the Gothic tale The Portent (1860/1864), and then turned exclusively to the conventional Victorian three-decker format for his adult fiction. It was only with Lilith, written and published at the other end of a long, prolific writing career, that MacDonald would return once again to the fantastic mode of Phantastes. Like its predecessor, Lilith’s invented world functions more as an alternate dimension than a representational world of the five senses. The idea of “dimensions” is in fact explicitly mentioned several times;66 the narrator repeatedly stresses that his narrative is a translation into a verbal construction of forms of something that did not happen in those forms. But though the imaginary landscape represents on one level the psyche of the protagonist Mr. Vane, it does assume a more definite form than the earlier Fairyland. The geography is consistent and could be mapped on the basis of the text; the hazy woodland recesses of Phantastes are replaced by a starker, more diagrammatic landscape; the background history of Lilith, Lona, and the Ravens gives the narrative a focused dramatic cohesion.67 Nevertheless, the visionary frame of Lilith connects it, along with Phantastes, to the dream vision. MacDonald’s own contention that the book was written under a sort of divine compulsion further underscores a connection to Blake and Dante. Lilith is a much darker, more disturbing work than Phantastes, with flashes of anger altogether at odds with the more

FROM VERSE TO PROSE

117

genial tone of the latter. It is clear MacDonald put more painstaking effort into the composition of Lilith: while Phantastes was written in about two weeks, the final text of Lilith was preceded by half a dozen different drafts, and the initial draft, dating to 1890, had gone through a dramatic transformation by the time the final was reached. Lilith is, at least, a more accessible book to many than Phantastes.68 Like its predecessor, Lilith did find some admirers, though it was largely met by bemusement and hostility in the press, and it was not popular with much of MacDonald’s sizable audience built on his novels. It is not amiss to remember that, though the two visionary romances stand out in bold relief among MacDonald’s works a century on, in their day they occupied an obscure branch of a popular novelist’s output: in 1900, MacDonald would have been recognized by the bulk of the reading public as the author not of Lilith or Phantastes but of David Elginbrod and Robert Falconer. In the present context, George MacDonald’s two visionary romances must be seen as somewhat peripheral. They do share significant elements with other major works here, sufficient that they should not be ignored, and it is not inappropriate that they were both included in the BAFS. At the same time, they belong to a fundamentally different literary mode than The Well at the World’s End, The Worm Ouroboros, and LOTR; the Curdie books are much closer to the latter three titles than Lilith or Phantastes. Lin Carter’s distinction between the two visionary romances and the core BAFS mode is problematic in its elaboration, but insofar as it indicates a genuine and significant difference, it is in fact quite valid. The only significant work evincing the influence of Lilith and Phantastes tied to the BAFS is David Lindsay’s A Voyage to Arcturus, a work similarly peripheral to the core BAFS template. The Prose Romances of William Morris William Morris’s eight prose romances appeared between 1888 and 1898.69 The first two, The House of the Wolfings (1888) and The Roots of the Mountains (1890), are the “Germanic romances,” usually treated as a semiautonomous subgroup. Both are historically situated, though vaguely so, and focus on ersatz Germanic tribes: the former at the time of Roman incursion into central Europe, the latter at the time of the Hunnish invasions. These features led to some misrepresentations during the BAFS period. For example, in Imaginary Worlds (1973), Lin Carter opined that Wolfings was “just another historical novel, the sort of thing Sir Walter Scott was writing twenty years before Morris was born” (2Carter 24). But Morris’s Germanic romances bear little comparison to Scott’s novels beyond the fact that they

118

THE EVOLUTION OF MODERN FANTASY

are placed in the past. Where Scott’s tales—verse and prose—are invariably tied to specific, documented historical events, including actual historical figures in their casts of characters, Wolfings and Roots are more impressionistically placed in broad movements of history, and if they can be dated to within a century, that is as close as one can get. There are no identifiable historical figures in either book, and the Wolfings and Burgdalers were Morris’s inventions. Scott’s tales take place in precisely identified locations, which an interested reader could visit; the settings of Wolfings and Roots are specious and cannot be pinpointed. To have a goddess appear on the stage of one of Scott’s novels would have been a jarring incongruity, but Thiodolf ’s relationship with Wood-Sun is a crucial strand in the dramatic structure of Wolfings. It is true that Morris is very precise in his detailed descriptions of material culture and social structure, but his “historicity,” in a narrow sense, ends with these things. Morris’s two Germanic romances simply do not embody the same intentions as Scott’s novels. They are in fact much closer to the six romances that followed than to Scott’s work. The archaistic, poetic prose that would become an exceptionally controlled and supple instrument finds its first development here, and the narrative vocabulary—plot strategies derived from romance, saga, and epic; broadly delineated character types; and so on—directly anticipates the romances that followed. Wolfings incorporates the supernatural as actively as any of the later romances; Roots no less than some. The third of Morris’s prose romances was The Glittering Plain (1891). A brief work, it is scarcely half the length of Wolfings. The invented Clevelandby-the-Sea, where the story begins and ends, and its people of the Houses of the Raven and of the Rose, are cut from the same cloth as the communities of Wolfings and Burgdalers: material culture and social structure are founded on Germanic models, and the sense is of an undisclosed coastal region of northwestern Europe in the early Middle Ages. But this is only a sense: while the underlying suggestion is of a region contiguous with Europe, where that region actually touches Europe is not even speciously suggested, as in Wolfings and Roots. This is the major break with the Germanic romances, and the remaining five of Morris’s romances are similarly “off the map.” In contrast to the integrated foci of Wolfings and Roots, the structure of Plain is loose on a dramatic level. Hallblithe, the protagonist, moves smoothly from Cleveland-by-the-Sea to the Isle of Ransom to the Glittering Plain and back, but the interrelation of those core locations is loose. They function as semiautonomous worlds in themselves, worlds that Hallblithe’s adventures lead him through and that touch each other mainly by virtue of those adventures. While the telling is marked by Morris’s characteristic vividness and specificity of detail, the story is marked by a

FROM VERSE TO PROSE

119

dreamlike quality, particularly underscored by the main magical element, the land of youth, which is the Glittering Plain. The Wood beyond the World (1894), also brief, followed Plain in its abandonment of explicit ties to documentable history and geography, though it likewise conveys a sense of region and time contiguous with the historical. But here Morris moved away from Germanic tribal models: the protagonist, Golden Walter, is the son of a wealthy merchant dwelling in a port city, and the main narrative sequence takes place in “parts unknown” following being lost in a storm at sea. The sense at the beginning is of an English location during the mid to late medieval period, considerably later than the period implied in Plain. Once again, structure is loose on a dramatic level: Golden Walter moves smoothly from Langton-on-Holm, the port city, to the enchanted wood presided over by the Lady, to the land of the Bears, to the city of Stark-wall, but there is little sense of a connection between them. Unlike Plain, which assumes a “there-and-back” structure, Wood never returns to Langton, and Stark-wall, where the story concludes, has no perceptible connection to Langton beyond the fact that Golden Walter’s adventures led him from one to the other. Like its predecessor, Wood is marked by a dreamlike quality, again underscored by the magical elements: the enchanted Wood beyond the World, and the supernatural machinations of the Lady. Child Christopher and Goldilind the Fair (1895), once again brief, suggests in its landscape and nomenclature an imaginative extension of medieval England.70 The story follows a more integrated structure than its two predecessors, and the various regions of the imaginary landscape are linked by a clear dramatic framework. The central drama, whose background is furnished by a temporal canvas that follows both Christopher of Oakenrealm and Goldilind of Meadham from infancy to early adulthood, is a twofold succession struggle that does not unfold around a linear quest or journey. Child Christopher, unlike its predecessors, is centered in the world of human society, and magic is restricted to a premonitory dream early in the tale and a few offhand allusions to woodland fairy folk who never appear and never do anything. The longest of Morris’s eight prose romances, The Well at the World’s End (1896), was also the most protracted in its composition time, and it parallels rather than succeeds Wood and Child Christopher. Like Plain and Wood, Well is structured around a quest. The human sphere is developed in far more meticulous detail than in the worlds of Hallblithe and Golden Walter, however, and though the geographic canvas is vast compared to that of Child Christopher, the development of setting in Well is more like that of the latter work. Where Plain and Wood move fairly rapidly between a finite number of key settings, and Child Christopher does not leave a narrowly

120

THE EVOLUTION OF MODERN FANTASY

charted stage, the combination of “there and back” quest structure with precise local detail and geography lends an encyclopedic dimension to Well. There is far more a sense of Ralph, the tale’s protagonist, growing and processing his experience than was the case with Hallblithe or Golden Walter. The various towns, manors, forests, and cities Ralph passes through are inextricably linked to stages in his development, embodying a variety of configurations of human society: the localized, egalitarian kingdom of Upmeads; the aristocratic manor of the Lady of Abundance; the merchant town of Whitwall; the autocratic tyranny of the Kingdom of Utterbol; the tribal society of the Innocent Folk. In contrast to Child Christopher, magic is very much a part of Well, although more muted than in either Plain or Wood. The magical powers of the Lady of Abundance are real and implied to be considerable, but there is no lavish display of them; the Sage of Swevenham casts a minor spell of illusion. Beyond this, there is only the Well at the World’s End itself, which confers long life and strength but no dramatic, immediately perceptible change. The culmination of the quest, Ralph and Ursula’s drinking of the water of the Well, derives its great beauty through the understated limpid simplicity of the language through which Morris conveys it: its “magic” is in the quietly ecstatic mood evoked by Morris’s prose, not any overt show of the supernatural. The final two romances, The Water of the Wondrous Isles (1897) and The Sundering Flood (1898), were posthumous. The former is the second longest of the eight; the latter seems somewhat abbreviated.71 Both include quests that form important parts of the tales, but they are not themselves structured as quest tales. Rather, as in Child Christopher, the stories follow their protagonists—Osberne in Flood, Birdalone in Water—from infancy to early adulthood. The worlds of each are developed with a level of detail on par with Well, and especially with Water, the geography is tightly woven into the dramatic substance. Magical elements figure more prominently in Water than in any other of the eight romances; in Flood, barring the subplot with the magical dwarf-man introduced early in the story and then abandoned, there is no recourse to the magical. These eight romances were published to no great fanfare, and while Wolfings did attract a fair amount of press attention, more favorable than not, Peter Faulkner notes that its followers attracted increasingly less (Faulkner 19). Of course, Morris was well-known, with a considerable prior literary reputation, and this, if nothing else, ensured that they would not be completely ignored. It should be noted that the writing of the romances on the whole seems to have served as a kind of respite and means of relaxation to the main thrust of Morris’s energies.72 Following his estrangement from the Socialist

FROM VERSE TO PROSE

121

League in the late 1880s, Morris focused on his work as a designer, including the designing and printing of books at the Kelmscott Press, where the first editions of the latter five romances were published. Like Blake with his illuminated books, Morris not only was concerned with the literary aspects of the romances but considered their mode of presentation an issue of equal importance. There is little to suggest that Morris was much concerned with making a popular splash with these works; he was certainly aware that they did not adhere to Victorian preconceptions concerning literary form. While Morris’s late prose romances were far from being bestsellers, they were far from unread. Most went into multiple editions and remained in print through the period of World War One. Indeed, Paul Fussell, in The Great War and Modern Memory (1976), discusses The Well at the World’s End as a “common text” of the generation that found itself on the Somme (Fussell 135–37). After World War One, however, the rise of modernism eclipsed this mild popularity, and by the time the BAFS began their reprints in 1969, Morris’s prose romances had all been out of print for forty years or more.73 Now they were the basis on which Lin Carter declared Morris “the man who invented fantasy”: here were the “fantasy novels” that initiated the imaginary-lands-where-magic-works “tradition” that resulted in Tolkien’s LOTR. Published in the company of Lord Dunsany, E. R. Eddison, James Branch Cabell, and so on, Morris was defined by what followed him. Though there is a certain rough validity to Carter’s claim,74 the series’s depiction of Morris’s romances suggested a connection to the past couched almost entirely in terms of vague references to medieval romances, blurring their connection to currents in nineteenth-century literary history: while indubitably idiosyncratic, they were very much an organic outgrowth of forces that had been operative in English-language literature since the age of antiquarianism. The company Morris published his romances in at the Kelmscott Press is relevant here: in keeping with what was suggested by Lin Carter, we find many medieval titles, most famously the Kelmscott Chaucer, but also Middle English verse romances, adaptations by Caxton of French prose romances, and Morris’s translations of French romances and of Beowulf. At the other end of the chronological spectrum, with very little between, there were books of poetry by Keats, Coleridge, Shelley, Tennyson, Swinburne, as well as Morris’s past poetic work. The latter connection is what was blurred by the BAFS, which, in turn, distorts the nature of the relationship to the medieval material. Morris was in fact an inheritor of the eighteenth-century tradition of antiquarianism: he worked first hand with old manuscripts and early

122

THE EVOLUTION OF MODERN FANTASY

printed books, which shaped the idea of bookmaking on which the Kelmscott Press was based. Though not an academic, his expertise in the literature brought to light in the wake of eighteenth-century antiquarian practice was acute. His translations, as previously noted, included medieval French romances, Beowulf, an array of Icelandic sagas (in collaboration with Erikr Magnusson), the Aeneid, the Odyssey, and even the beginning of Firdowsi’s Shah-Namah from a French translation. This literary activity bears a close relation to the prose romances. Simply on the level of language, the idiosyncratic, archaized prose of Morris’s own work finds a clear corollary in the equally idiosyncratic, archaized language of his translations. And Morris’s archaisms are not, as with Scott, elevated evocations of the Shakespearean but rooted in a precise, expert familiarity with the Germanic forms Middle English maintained from Old English.75 His narrative vocabulary was intimately shaped by the work he translated and printed and other like work: saga and romance techniques thread all the romances; the “epic” elements that set Wolfings apart from the others (the contained time frame; the death of Thiodolf at the end) suggest something of Beowulf; fairy-tale motifs (the youngest son makes good in Well; Water’s Habundia as a fairy godmother) proliferate.76 Where the BAFS suggests that Morris just started doing again something that hadn’t been done since Spenser, it is pertinent to note that he brought to his work a sophisticated syncretistic imagination wedded to an encyclopedic knowledge, which makes it uniquely modern and in fact quite different from Volsunga Saga or Huon of Burdeaux. The preceding pages of this study amply demonstrate that Morris was not the first person to do this, though he did perhaps bring to his work a level of expertise and breadth of knowledge beyond that of his predecessors: the period between Spenser and Morris, and especially the hundred years before Morris began publishing, is filled with narrative “fantasies” built on a dialogue with traditional narrative forms and subjects. Some of these would have been found in the pages of the Kelmscott Press editions of Keats, Coleridge, Shelley, and Swinburne. Others that Morris was quite aware of include Ossian, Lay of the Last Minstrel, and Vathek. Morris’s prose romances, written during the closing decade of his life, equally proceed from his own past work. To the extent that this has been explored in discussions of Morris as a fantasy writer, connections made to his earlier work tend to be confined to the body of eight short prose romances published in The Oxford and Cambridge Magazine (1855) during the same period that saw “The Defense of Guenevere.” Three of these tales in particular, “Gertha’s Lovers,” “The Hollow Land,”77 and “Svend and His Brethren,” take place in invented settings with a distinctly medieval cast and embody narrative elements akin to those of his late work. Neither the

FROM VERSE TO PROSE

123

first nor the third contain any magical elements of note (though, again, this is true also of Roots and Child Christopher). However, the more striking difference between these and the late romances is their visceral immediacy, violence, and tragic character. “The Hollow Land,” the one of the three in which the magical plays an important role, is a first-person narration embodying abrupt, dreamlike transitions of setting and recalling aspects of MacDonald’s Phantastes more than Morris’s late romances. The differences between the early short and longer late romances are as striking as their similarities; interestingly, the early tales were not published at Kelmscott. Morris abandoned the psychological immediacy of both his early poetry and his early prose after the 1850s. As noted before, in Jason, EP, and Sigurd, narrative distance is crucial to the overall effect, and though tragedy is frequent, there is little of the close, almost modernist, depiction of psychological motivation that characterizes the earlier work. The supernatural and magical play a far more frequent and foregrounded role than in the early work. In both ways, these poems can be seen to form a natural step between the early and late work; they were published in the company of the prose romances at Kelmscott. While modern perspectives tend to see a dramatic divide between poetry and prose, the sense is that Morris did not: long speeches (about one fifth of the text) in Wolfings are in verse; several of the late romances were first assayed in verse, suggesting that the border between poetic and prose narrative was fluid to Morris, his subject matter not inherently attached to one or the other. More specific echoes of the long poems in the prose romances can be discerned: the land of youth, the “Glittering Plain,” echoes, with a much revised perspective, the “Earthly Paradise”; the focus of the two Germanic romances follows naturally from Sigurd; the seagoing passages of Plain and Water echo those of Jason. The prose romances of “the man who invented fantasy,” then, would be more aptly said to advance, rather than break with, much Romantic writing and his own earlier work. The choice of prose may have been suggested by the sagas, by Fouqué’s Magic Ring, by Victorian literary fairy tales, or, most likely, some cross section of them all, but it does not seem to have stemmed from a consciousness of “changing direction.” In terms of the BAFS template, qualifications are in order: while magic occurs to varying degrees in most of the romances, its near absence in Roots and Child Christopher suggests that Morris did not see it as a defining element in the kind of narrative he was attempting; the invented settings appear in an almost offhand way and are by implication contiguous with late ancient or medieval Europe rather than posited as autonomous or independent worlds. These crucial BAFS elements are there; the question is to what extent Morris saw them as definitive.

124

THE EVOLUTION OF MODERN FANTASY

These qualifications and issues of context and continuity by no means diminish Morris’s importance in the history of fantasy. Regardless of whether or not he saw the BAFS characteristics as defining, those characteristics are there. Morris’s influence can be detected in the work of E. R. Eddison, Fletcher Pratt, and Tolkien. To question his status as “the man who invented fantasy” is to question something he did not claim; the suggestion behind this study is that modern fantasy in the BAFS mode developed incrementally and that it was not created in a blinding flash by anyone. Though Morris’s romances ran in the face of conventional Victorian expectations for prose fiction,78 they were tied to Victorian (and earlier) literary phenomena nonetheless. Works Cited Avery, Gillian. Introduction, in Lang, Andrew. The Gold of Fairnilee and Other Stories. London: Victor Gallancz, 1967. 1Carter, Lin. “About The Well at the World’s End and William Morris,” in Morris, William. The Well at the World’s End. Volume 1. New York: Ballantine Books, 1970. 2Carter, Lin. Imaginary Worlds. New York: Ballantine Books, 1973. Clute, John, and John Grant, eds. The Encyclopedia of Fantasy. New York: St. Martin’s Press, 1997. Coe, R. H. Preface, in Ruskin, John. The King of the Golden River. Boston: Ginn and Co., 1916. Cook, Cornelia. “The Victorian Scheherazade: Elizabeth Gaskell and George Meredith,” in Caracciolo, Peter L., ed. The Arabian Nights in English Literature. New York: St. Martin’s Press, 1988. Eddison, E. R. The Worm Ouroboros. New York: Alfred and Charles Boni, 1926. Faulkner, Peter, ed. William Morris: The Critical Heritage. Boston: Routledge and Kegan Paul, 1984. Fussell, Paul. The Great War and Modern Memory. New York: Oxford UP, 1975. Graves, Alfred Percivale. Introduction, in Ferguson, Samuel. Poems of Sir Samuel Ferguson. Dublin: Talbot Press, 1916. Lewis, C. S. On Stories and Other Essays on Literature. Edited by Walter Hooper. New York: Harcourt Brace Jovanovich, 1982. Meredith, George. The Shaving of Shagpat. New York: Ballantine Books, 1970. 1Morris, William. The Earthly Paradise. 4 volumes. New York: Longmans, Green, and Co., 1896. 2Morris, William. Child Christopher and Goldilind the Fair. Van Nuys, CA: Newcastle Pub. Co., 1977. Raeper, William. George MacDonald. Batavia, IL: Lion Publishing Corp., 1987. Tennyson, Alfred Lord. Idylls of the King and a Selection of Poems. New York: Signet Classics, 1961. Thompson, E. P. William Morris. New York: Pantheon Books, 1977.

FROM VERSE TO PROSE

125

Tolkien, J. R. R. The Monsters and the Critics and Other Essays. Boston: Houghton Mifflin and Co., 1983. Zipes, Jack, ed. Victorian Fairy Tales: The Revolt of the Fairies and Elves. New York: Methuen, 1987.

Other Secondary Work Consulted Attebury, Brian. The Fantasy Tradition in American Literature. Bloomington: Indiana UP, 1980. Kirchhoff, Frederick. William Morris. Boston: Twayne Pubs., 1979. Noyes, Alfred. William Morris. London: Macmillan and Co., 1926. Prickett, Stephen. Victorian Fantasy. Waco, TX: Baylor UP, 2005. Reis, Richard. George MacDonald’s Fiction. Eureka, CA: Sunrise Books, 1989. Robb, David S. George MacDonald. Edinburgh: Scottish Academic Press, 1987. Silver, Carole. The Romance of William Morris. Athens, OH: Ohio UP, 1982. Zipes, Jack. Fairy Tales and the Art of Subversion. New York: Methuen, 1988. Zipes, Jack. The Brothers Grimm. New York: Routledge, 1988.

Some Primary Source Editions All William Morris’s work can be found in The Collected Works, edited by May Morris. The long narrative poems have not had new, independent editions for nearly a century. The prose romances were all published in either the BAFS or Newcastle’s Forgotten Fantasy Library; there have been sporadic paperback editions of some of them in the interim, of which the most noteworthy are the Dover Books facsimiles of the Kelmscott editions of The Glittering Plain and The Wood beyond the World. All George MacDonald’s work has been reprinted by Johannesen. Lilith, Phantastes, The Princess and the Goblin, and The Princess and Curdie have been reprinted in many editions over the last several decades, as have many of the shorter fairy tales. The best edition collecting all the short fairy tales is the following: MacDonald, George. The Complete Fairy Tales. Edited by U. C. Knoepflmacher. New York: Penguin Books, 1999. The poems discussed of Tennyson, Swinburne, Ferguson, and Longfellow can be found in most of the many editions of their poets’ work. Most of the children’s work discussed here, including Thackeray’s The Rose and the Ring, can be found in one or more of the following contemporary anthologies: Auerbach, Nina, and U. C. Knoepflmacher, eds. Forbidden Journeys: Fairy Tales and Fantasies by Victorian Women Writers. Chicago: U of Chicago P, 1992. Cott, Jonathan, ed. Beyond the Looking Glass: Extraordinary Works of Fairy Tale and Fantasy. New York: Stonehill Pub. Co., 1973.

126

THE EVOLUTION OF MODERN FANTASY

Hearn, Michael Patrick, ed. The Victorian Fairy Tale Book. New York: Pantheon Books, 1988. West, Mark I., ed. Before Oz: Juvenile Fantasy Stories from Nineteenth Century America. Hamden, CT: Archon Books, 1989. Zipes, Jack. Victorian Fairy Tales listed in Works Cited. Andrew Lang’s work can be found in The Gold of Fairnilee, listed in the Works Cited under Avery, as well as the following: Lang, Andrew. The Chronicles of Pantouflia. Boston: David R. Godine, 1984.

5

Twentieth Century The Literary Canon

Introduction

T

he period from about 1900 to 1960 saw the publication of the bulk of the material that either made up the BAFS or was cited regularly by Lin Carter as part of the “tradition” that William Morris “invented.” In the present context, this material may be said to constitute an almost exclusive migration to prose of narrative drawing on, and in certain respects emulating, the various premodern modes of romance, saga, epic, and fairy tale, after more than a century dominated by poetry. This material will form the primary focus of the next two chapters: the present chapter will center on the “literary” branch of the BAFS “canon” (Dunsany, Eddison, Tolkien), as it is here that the currents discussed over the previous chapters are directly evident; Chapter 6 will center on the “popular” branch (Smith, Howard, Leiber), the relationship of which to premodern narrative modes is rather different, and which on the whole stands in a closer relationship to modern forms of popular romance. The Victorian vogue for fairy tales for children washed into the early twentieth century, though the flood began to abate after World War One; important work continued to appear, but out of a less prolific ground.1 Some of this work followed the loose templates characterizing the Victorian material, but there were also some new developments away from the conventional fairy tale. Book-length narratives became dominant, and there were fewer short fairy tales. At this point, poetry effectively ceases as a significant force in the development of fantasy, though it does not wholly disappear. During the post–World War One period, the Arthurian legends once more became a popular subject for poetry: the American poet Edward Arlington Robinson

128

THE EVOLUTION OF MODERN FANTASY

published his volume-length Merlin (1917), Lancelot (1920), and Tristram (1927); John Masefield published Midsummer Night (1928), a collection of Arthurian narratives. Though unpublished at the time, Tolkien’s The Fall of Arthur and C. S. Lewis’s “Launcelot” were composed in the early 1930s and must be seen in this company.2 Fellow Inkling Charles Williams’s more abstruse Taliessin through Logres (1937) and The Region of the Summer Stars (1944) are also notable here, though narrative per se is very much sublimated to Williams’s symbolic construction. Both Tolkien and Lewis wrote a number of non-Arthurian fantasy verse narratives during this period as well. However, Lewis’s Dymer (1926), whose matter was invented, is the only of these to have been published at the time; “The Nameless Isle” and “The Queen of Drum,” also based on invented matter, only appeared posthumously. Tolkien completed substantial fragments of verse versions of the tales of Turin and of Beren and Luthien during the 1920s,3 and The Legend of Sigurd and Gudrun (2009) is believed to date to the following decade. In this work, we can see that the poetic interests of both authors were not unrelated to their interests in prose narrative;4 certainly these attempts were in part inspired by the narrative poetry discussed in the preceding chapters. But while this is not an entirely negligible body of work, in the present context it is somewhat peripheral: the Arthurian work stood very much in the shadow of Tennyson and Swinburne, and the prose novels of T. H. White shaped twentieth-century conceptions of the legends far more; apart from Dymer, the non-Arthurian material was not published. The period in which it was written is the period when modernism (and especially the influence of T. S. Eliot), with its focus on fragmented perspectives and more compact, shorter forms, for many redefined what poetry was. In some cases, this material may be seen as a reaction against modernism and an attempt to realign poetry with Romantic and Victorian models;5 it failed to achieve its goal in this capacity. By the 1960s, the underlying assumption had grown to be that modern fictional narrative was done in prose: this is arguably the major factor in the relative neglect of eighteenth- and nineteenth-century narrative poetry in histories of fantasy. In terms of the broader literary context, academic and popular scholarship and research pertaining to the material that antiquarians had begun to publish more than a century earlier continued apace (and continues unabated to the present day). Texts have been updated, translation practices have become more exacting, and the range of traditional literatures accessible to the nonspecialist in fairly reliable forms has considerably broadened. At the same time, the movement from retrieval and collection to refinement and analysis that had begun during the Victorian period became

TWENTIETH CENTURY

129

more pronounced. As far as “new” material with a broad significance to fantasy, the proliferation of English translations of ancient Gaelic narrative during the opening part of the twentieth century essentially completes the core of pertinent traditional material.6 Apart from this, most “new” discoveries—for instance, the Winchester manuscript containing an alternate to Caxton’s Malory—served more to refine, augment, or complicate perspectives on something already substantially on the table. The movement to “refinement and analysis” mode is underlined by the erstwhile catchall antiquarian having, by this point, largely subdivided into discrete disciplines: philology, folklore, anthropology, and so on. Nevertheless, this material continued to be significant in a very direct way to most of the writers of the literary branch of the BAFS canon; most had a notable degree of antiquarian expertise; some were in fact academic specialists. On a more general level, the sheer bulk of material available, both within the core branches of tradition and elsewhere, provided a solid groundwork on which a creative vocabulary founded on the tropes and conventions of romance, saga, epic, and so on might be constructed. In the following pages, after a brief discussion of children’s fantasy, I will turn to a detailed consideration of the literary branch of the BAFS canon. Children’s Fantasy A fair amount of work from the period followed the loose Victorian template for the fairy tale, with roots in Grimm and the contes de fees, as well as an influx of Andersen.7 However, much of the best-known work moves away from the standard fairy-tale vocabulary. Among the most popular, and certainly the most voluminous, children’s works from the opening two decades of the twentieth century are L. Frank Baum’s 14 Oz books, beginning with The Wonderful Wizard of Oz (1900) and ending with Glinda of Oz (1920). Following Baum’s intention to modernize and Americanize the fairy tale, the “stereotyped genie, dwarf, and fairy are eliminated” (Baum Introduction), and since Anglo-American culture lacked corollary traditional magical beings, Baum peopled Oz with his own whimsical creations (Scoodles and Wogglebugs) and magical objects inspired by American popular culture (Lunchbox Trees). While the story structures are framed around the common cross-worlds journey by a present-day child,8 and narrative motion reflects the fairy tale, the feel of Oz is quite distinct from Ruskin, MacDonald, and Lang. Likewise, J. M. Barrie’s Peter Pan (1911) employs the cross-worlds journey, with Wendy magically transported to Neverland by Peter Pan, and follows a fairy-tale structure. Yet Neverland itself, populated by the Lost

130

THE EVOLUTION OF MODERN FANTASY

Boys, Indians, pirates, and Captain Hook, draws more on Victorian boy’s adventure fiction than the fairy tale in its makeup, regardless of Tinker Bell and fairy dust. Kenneth Graham’s The Wind in the Willows (1908) invokes the beast fable, a traditional narrative form if not the fairy tale per se. But apart from the appearance of Pan in the beautiful, breathless midpoint chapter, “The Piper at the Gates of Dawn,” Graham’s book stays fairly close to the mundane world of Edwardian England, with motorcars, railway stations, and washerwomen, and the story unfolds in a leisurely, whimsical fashion.9 But probably the most notable catalyst of change in the formal development of children’s fantasy came with the more well-known of E. Nesbit’s magic books: the trilogy comprising Five Children and It (1902), The Phoenix and the Carpet (1904), and The Story of the Amulet (1906), as well as The Enchanted Castle (1907). These books center on the impingement of the magical on realistically drawn “modern” children (not essentially different from the children of Nesbit’s nonmagical books, such as those in The Wouldbegoods) in otherwise mundane settings. While the children may travel, in individual scenes, via magic to enchanted gardens or ancient Babylon, the core tensions reside in the playing out of the magic in the “real” world. The more formal tone of the fairy tale is replaced by an appropriately familiar, sometimes flippant voice. These elements did not originate with Nesbit, but it was in her work that they came together with a particularly deft and memorable force, and her work exerted a strong influence on twentieth-century children’s fantasy. Rudyard Kipling’s Puck of Pook’s Hill (1906) and Rewards and Fairies (1910) are indebted to Nesbit; half a century later, one can discern Nesbit’s presence in the Pevensie children of the Narnia series; a century later, the Nesbit voice is evident in J. K. Rowling’s Harry Potter series. At the same time, while Nesbit influenced work like the Narnia series and Wyke-Smith’s The Marvelous Land of Snergs (1927), the works themselves are on the borderline of the BAFS template. Further away from the focus of this study, her influence can also be seen in more whimsical fantasies such as Hugh Lofting’s Dr. Dolittle series, beginning with The Story of Dr. Dolittle (1920), and P. L. Travers’s Mary Poppins series, initiated in Mary Poppins (1934). It was during the 1920s and earlier 1930s that J. R. R. Tolkien wrote the bulk of his work for children. His most famous children’s fantasy is, naturally, The Hobbit, completed by the early 1930s and published in 1937. The invented world of The Hobbit follows Thackeray, MacDonald, and Lang but owes more to the world of Germanic and Scandinavian tradition than to these predecessors. It was certainly the most detailed and precisely developed secondary world of any children’s fantasy up to its

TWENTIETH CENTURY

131

publication—owing, of course, to Tolkien’s conjoining it to the universe of his already developed “mythology.” But in its voice (which Tolkien came to disapprove of) and in its key invention, hobbits (for which Tolkien was partly indebted to Wyke-Smith’s Snergs), The Hobbit is clearly connected to the children’s fantasy of its period, and earlier, and to the other children’s tales he wrote in the 1920s and 1930s, none of which was published until considerably later.10 These latter include Roverandom (1998), in the fairy-tale mode, and Mr. Bliss (1983), reminiscent of A. A. Milne and Beatrix Potter. The finest of these, and the only to be published in Tolkien’s lifetime, was Farmer Giles of Ham (1949), and the distance from the initial short version of the late 1920s to the final published version provides a barometer for Tolkien’s developing views on writing “for children.” The short version is very much a children’s fairy tale, the narrative voice in keeping with The Hobbit, with frequent chatty asides to “the children.” In the well-known, expanded version of 1949, the asides are removed, the texture more nuanced without losing the comic quality, and the narrator has become an antiquarian “translating” the tale from Latin, explaining the etymology of place-names. A mock scholarly “Forward” is added. While the setting is Britain, the period is vague and compounded of elements from incompatible eras; Farmer Giles and his cohorts, and the little world they occupy, are cut from the same cloth as hobbits and what would become the Shire.11 By the time The Hobbit was published, children’s fantasy in keeping with the BAFS template had ebbed. The next significant work, C. S. Lewis’s The Chronicles of Narnia (1950–56), really stands at the beginning of the transition point to the genre period as the first to evince the influence of Tolkien.12 Nevertheless, the roots of Chronicles lie in Victorian and early twentieth-century work. Lewis’s children and, at points, his narrative voice echo Nesbit; the debt to George MacDonald is clear; the crossworlds journey, though Lewis added a vaguely science-fictional speculative edge, was a standard motif. The influence of Tolkien is seen primarily in the development of Narnia itself, which, though initially rather offhand, gradually assumes some of the sense of historical depth that inheres in Middle-earth.13 After Narnia, it is a short leap to the genre period, with Carol Kendall, Lloyd Alexander, and Ursula Le Guin initiating a new wellspring of children’s fantasy that is still with us.

132

THE EVOLUTION OF MODERN FANTASY

The Literary Branch of the BAFS Canon Again, the designations of “literary” and “popular” are strictly quantitative, not qualitative, and as discussed in the Introduction, they point to differing publication venues, as well as to different relationships with the literary developments outlined here. The major authors representing the literary strand of fantasy between the period of William Morris’s romances and the publication of LOTR, included in the BAFS, were Lord Dunsany, James Branch Cabell, and E. R. Eddison. To these I add the Welsh writer Kenneth Morris, the revival of whose work postdates the BAFS, and the Irish writer James Stephens, oddly neglected despite the fact that he must be twinned with Morris as a major early twentieth-century progenitor of “Celtic Fantasy.” Tolkien, of course, is a major author here. I will also discuss work by Ernest Bramah, Evangeline Walton, T. H. White, and some others. It should be reiterated that there is no evidence among the writers themselves of any sort of genre consciousness—no sense that they saw themselves as collectively representing any definable literary movement. While the BAFS grouped them together retroactively on the basis of what were in fact common elements, the degree to which those common elements can be attributed to any degree of interinfluence is quite limited. In addition, the actual direct influence of William Morris’s prose romances on most of these writers is, at best, questionable. I will explore the question of interinfluence in the following pages and discuss closely the exact nature of the relationship of the varied writers’ work to the terms of the BAFS template. But before turning to the writers themselves, it would be pertinent to reflect on what other common ground they shared, with William Morris and among themselves, apart from producing work adhering to the BAFS template. Like Morris, the major writers of the twentieth-century literary canon all had some degree of expertise in the traditional literatures that proliferated in the wake of eighteenth-century antiquarianism—an expertise that extended well beyond the handbook and popular retellings by Pyle, Spence, and Guerber.14 Stephens and Kenneth Morris knew the works they reimagined in their original languages; Morris’s expertise stretched beyond Wales and Europe to Persia, India, and China. E. R. Eddison published a translation of Egil’s Saga (1930). Though Cabell and Dunsany had no relevant academic affiliations and produced no relevant scholarship, their work evinces a ready familiarity with the vocabulary of heroic romance, epic, saga, and fairy tale. And Tolkien was one of the foremost philologists of his day, with an intimate knowledge of both the traditional literatures pertinent here and their original languages: not only did he inherit the

TWENTIETH CENTURY

133

work done by scholars since the eighteenth century; he was himself in the vanguard of academic research during the second quarter of the twentieth century. Emerging from this expertise, a common characteristic of the work of these writers is that it does not simply adapt stories about warriors, wizards, and magic swords to the modern novel or adventure story forms but in fact directly engages the narrative conventions and structures of the material that served as inspiration—often adopting (most noteworthy in Cabell and Tolkien) a quasi-scholar/editor narrator persona to create the illusion that the work itself is something that has been passed down through tradition in variant and imperfect forms. In this, the twentieth-century writers were following William Morris, but they were also following a great deal else from Morris’s contemporaries and predecessors. The authors of the “epics” and “poetical romances” and “traditionary ballads” of the nineteenth century also worked closely with original materials, not handbook content summaries. My discussion in Chapter 4 suggests that William Morris’s move from verse to prose advanced, rather than changed, the direction already evident in his work. The poetry of Morris, Tennyson, Keats, Scott, and others was something the twentieth-century writers had in common; that at least Tolkien and Lewis produced some relevant poetic narrative highlights this connection. Another common feature of the major twentieth-century literary work resides in the carefully nuanced, artificial, usually archaized prose styles. In part, this echoes the lofty language of the “epics” and “poetical romances.” However, the tendency to mannered archaism in diction and syntax also informed many nineteenth- and early twentieth-century translations of actual traditional material, with the goal, essentially, of creating an effect akin to that of reading Malory or the metrical romances. William Morris’s translations of both poems (Beowulf) and prose works (Volsunga Saga) are classic exempla of this approach to translation. The language of these translations and the language of much nineteenth-century poetic fantasy are of a piece, and both must be seen as shaping factors in the prose styles of our twentieth-century literary fantasists. And given the enormous disparities between the King James rhythms of Dunsany, the Welsh syntax of Kenneth Morris, the high-flown Elizabethanism of Eddison, and the sober AngloTeutonic constructions of William Morris, nineteenth-century poetic and translation practices are more apt to have been a determining factor than any close interinfluence. Though the use of mannered archaistic language is a similarity here, it is a vague and specious similarity, and none of the twentieth-century prose styles can be taken as imitations of Morris’s. Finally, one must consider the form in which all the major canonical writers—born between 1878 (Dunsany) and 1892 (Tolkien)—would have

134

THE EVOLUTION OF MODERN FANTASY

first encountered the material that inspired them: adaptations for children. Experts are made and not born, and it is likely, for example, that Tolkien and Eddison encountered William Morris’s translation of Volsunga Saga and his “epic” The Story of Sigurd before learning Icelandic and that the translation and the poem provided some of the motivation to learn it. Pushing further, Tolkien’s first encounter with the Sigurd legend was Andrew Lang’s adaptation in The Red Fairy Book; it is likely Eddison first encountered it there, too. As noted in Chapter 4, these adaptations were, in their presentation to the reading public, of a piece with folk-based fairy tales and literary fairy tales and often mixed with them indiscriminately in anthologies such as Lang’s. While, naturally, the failings of such a catchall approach would become clear once knowledge had extended to the material in its original forms, the collective would serve to link myth, legend, and fairy tale (traditional or literary) in the crucial formative stage of childhood. So these writers did have certain significant things in common beyond the BAFS template—things that the conventional picture of “Morris the Inventor” with a tradition of followers does not take account of. The trajectory of “Adult Fantasy” here is close to that of children’s fantasy: there is a fairly prolific output from the turn of the century until the 1920s, followed by a decline. Of the major writers, only Eddison and Tolkien produced relevant work after this point. I have broken down the canon into four chunks: (1) Dunsany, Stephens, and Kenneth Morris, who had initial significant associations with the Irish Renaissance; (2) Cabell and some secondary writers who began publishing prior to World War One; (3) Eddison and other writers who began publishing after World War One; and (4) Tolkien. The Celtic Revival While Dunsany, Stephens, and Kenneth Morris were all three associated in some capacity with the Irish Renaissance and, more broadly, the Celtic Revival of the turn of the century, the nature of the association and its effect on their work was quite distinct. Lord Dunsany came into his first popularity primarily through his plays, which were initially produced at Yeats and Lady Gregory’s Abbey Theater in Dublin. Yeats edited Selections from the Writings of Lord Dunsany (1912). At the same time, Dunsany, an Anglo-Irish aristocrat, was conservative in his political views,15 which put him at odds with Yeats, Gregory, and most of the other major writers of the Irish Renaissance. He did not write anything explicitly concerned with Ireland—none of his major fantasies deal with Irish subjects—until The Curse of the Wise Woman (1933).16 So while

TWENTIETH CENTURY

135

there is little directly “Irish” about Dunsany’s fantasy work, the popularity garnered by his plays, first produced in Ireland, fed into a readership for his fiction.17 According to the chronology set up by Lin Carter, Dunsany was “the second major writer in the English language to work in the new genre of the imaginary-world fantasy tale” (1Carter 22), and he “adapt[ed] William Morris’ invention . . . to the short story” (2Carter 29–30). While the latter contention is not strictly true (several of Morris’s early tales fit the tab), Dunsany was the first to produce a substantial quantity of short work in strict, literal keeping with Carter’s definition. The first of Dunsany’s two blocks of work significant to fantasy is the eight collections of short tales, beginning with The Gods of Pegana (1905) and ending with Tales of Three Hemispheres (1919), which provided the bulk of the material included in the three BAFS Dunsany anthologies.18 The contents of the first two volumes, Pegana and Time and the Gods (1906), are all pretty well in keeping with the BAFS template. The first volume comprises more than two dozen short pieces, many more descriptive than narrative, centered on the acts and attributes of the gods of Dunsany’s invented pantheon. The second volume remains in the same cosmos but moves predominantly to the human sphere, often exploring the tensions between human and deity. There is a greater emphasis on narrative, and Dunsany employs motifs drawn from the hero tale, myth, and fable. The atmosphere of these two volumes veers strongly in the direction of the quasi-Oriental tale, compounding names (Snamarthis, Sheenath, etc.) that recall those of the Old Testament and other ancient and medieval Near East traditions. The prose rhythms echo those of the King James Bible. While these stories are, in fact, set in an invented cosmos, Dunsany was clearly not interested in some of the more technical matters that are now built in to writing fantasy. The content of the first volume is generally referred to as a “mythology,” though in fact, despite its gods, the narratives are more akin to poetic fables in prose. One flows smoothly to the next, but the motion is whimsical: there is no central dramatic tension, no sense of historical motion. The stories of the second volume occasionally crisscross each other, and a few (“The Cave of Kai” and “The Sorrow of Search,” for example) compose minisequences, but on the whole they are, beyond their setting, miscellaneous. Only in a later story connected to Pegana, “Idle Days on the Yann” from A Dreamer’s Tales (1910), does even a limited sense of geography emerge. After Time, Dunsany largely left Pegana behind,19 and his work underwent a gradual shift in tone and narrative strategy. The Sword of Welleran and Other Stories (1908) contains a fair proportion of work similar to the Pegana tales, though the quasi-Oriental flavor becomes less pronounced.

136

THE EVOLUTION OF MODERN FANTASY

Perhaps the finest story in the collection, “The Fortress Unvanquishable, Save for Sacnoth,” largely eschews the quasi-Oriental, and wedded to a brilliant, mannered tongue-in-cheek delivery, its content figures much in the later Sword and Sorcery subgenre. But the collection contains fantasies that stray closer to the present world, such as the brilliant reversal of the journey into Fairyland, “The Kith of the Elf-Folk,” in which a magical “Wild Thing” ventures into the contemporary industrial world. And it contains stories such as “The Ghosts,” a Poe-esque tale of obsession in a mundane setting, outside of the focus of the present study. In A Dreamer’s Tales (1910) and Book of Wonder (1912), the emphasis remains decidedly on material in keeping with the BAFS template; the two contain some of Dunsany’s best work. At the same time, the tone is lighter and the irony more overt and sustained.20 In the last three volumes, Fifty-One Tales (1915), The Last Book of Wonder (1916), and Tales from Three Hemispheres (1919), Dunsany moved away from engagement with the forms of myth, heroic legend, and the fairy tale—from the kind of work that characterized the first five. The bulk of Dunsany’s short fantasy, then, belongs to the period preceding World War One. In the 1920s, he returned to fantasy once again in a mode amenable to the BAFS template, producing three book-length works significant here: Don Rodriguez: Chronicles of Shadow Valley (1922), The King of Elfland’s Daughter (1924), and The Charwoman’s Shadow (1926). All three were republished in the BAFS. The first of these is set in an imaginary “Golden Age of Spain,” echoing, with gentle mockery, the conventions of medieval chivalric romance. The pairing of its protagonist, Don Rodriguez, with the Sancho Panza–esque Moreno clearly invokes Don Quixote, as does the wandering, episodic character of the narrative. At book length, Dunsany is able to achieve a level of intimacy between reader and character that was not possible in his shorter work, and despite the far too frequent, long, cloying addresses to “My Reader,” the story achieves a certain poignancy that does not collapse into sentimentality, and it has some genuinely beautiful passages. The second, The King of Elfland’s Daughter, recalls the early tales in ways that Don Rodriguez does not. The narrator does operate on the familiar level with the reader as in the earlier book, but though the novel-length canvas permits a more sustained view of its chief characters—Alveric, the Lord of Erl, and his son, Orion—they remain distant to the reader, with a nominal degree of intimacy developing with Lirazel, the King of Elfland’s daughter. The prose remains consistently distant and formal, achieving much of its effect with repeated phrases. The combination of elevated poetic language (with echoes of Tennyson’s “The Horns of Elfland”), narrative distance, and conventional romance and fairy-tale plot motifs (the

TWENTIETH CENTURY

137

quest for the fairy bride, the Hunt, etc.) and character types (the Witch, the King of Elfland, etc.) evokes an archaistic effect. The brief preface locates the story nominally in England, but this is even more tenuous than the “Spain” of its predecessor. Better integrated and without the cloying asides, King is far more successful on the whole than Don Rodriguez, and many consider it to be Dunsany’s finest work. The last of the three, The Charwoman’s Shadow, is once again set in the imaginary “Golden Age of Spain,” though the thread connecting this work to the first—its Duke of Shadow Valley, a minor if important character, is the son of Don Rodriguez—is nominal. As with Don Rodriguez, the characters are presented with a degree of intimacy, and the narrator speaks familiarly to the reader. In contrast, the cloying addresses to “My Reader” are largely gone, and the story has a much tighter dramatic structure. Elements of traditional romance and fairy tale are liberally sprinkled throughout the work (the quest of the son, the wizard in the wood, the bride under a disfiguring spell), and at the close, Dunsany makes use of the convention of the ersatz authority, common in medieval romance, to vouch for the “historicity” of the wedding of the protagonist’s sister, Mirandola, and the Duke of Shadow Valley. Though lacking some of the haunting quality of King, Charwoman is better constructed dramatically and is at least its equal. After Charwoman, Dunsany’s work moved away from the BAFS template, though two further works, both with contemporary settings, deserve note here. In both The Blessing of Pan (1927), set in a rural English village gradually infiltrated and transformed by the forces of the pagan god Pan, and The Curse of the Wise Woman (1933), set in an Irish village where a “Wise Woman” invokes the forces of Tir-Nan-Og in an attempt to drive out developers, something of the thematic substance of the earlier books, especially that of King, is evident, and the Wise Woman’s evocations of TirNan-Og carry some of the effect of Dunsany’s early short tales. But both books move distinctly away from the romance and fairy-tale conventions of their three predecessors in the direction of more conventional literary fiction.21 The core Dunsany work here fits the BAFS template fairly unequivocally, allowing a mild qualification for the “Spain” and “England” of the book-length stories. But the function of magic and the nature of even the fully invented worlds deserve further discussion. Magic has a pervasive presence in virtually all Dunsany’s relevant work, though there is often a whimsical tone to its presentation that at times verges on outright tongue-in-cheek: the power of the sword’s name in “The Fortress Unvanquishable”; the description of reading and writing as a branch of magic in Charwoman. In places it is associated with a dehumanizing obsession with knowledge (the magicians in both Don Rodriguez

138

THE EVOLUTION OF MODERN FANTASY

and Charwoman), in others with perception at the crossing point of imagination and memory (King). Its significance in context tends more to the metaphorical, and it does not carry the more speculative and systematic dimension that emerged in the pages of Unknown and has become common in the post-BAFS genre. At the same time, it is much more foregrounded and less downbeat than it tends to be in William Morris’s romances. As noted, the “Spain” and “England” of the book-length work are a mild stretch if Carter’s “invented worlds” is taken literally, but even the imaginary Pegana and the later, more whimsical “Edge of the World” and “Third Hemisphere” need some qualification. The latter two are described explicitly as “countries of my dreams” (as is Pegana in the three late tales), and they are clearly represented as emanations of the imagination, not chronicled histories underscored by a cohesive realism. Geography is vague, and references to background places serve an aesthetic purpose. These invented realms contrast dramatically with Morris’s geographically precise, realistic landscapes: they are regions of the imagination, sometimes fabulous, sometimes extensions of the primary world as seen when memory meets imagination. Dunsany does not take them seriously as independent representational realities. The BAFS template, then, simultaneously bonds Dunsany and Morris and highlights what are considerable differences. Other differences are substantial enough that it is difficult to speak of Dunsany “following” Morris in more than chronology. As with Morris, Dunsany’s prose style is artificial and archaized, with a rhythmical poetic texture. But Dunsany’s style is more varied (and more given to prolix self-indulgence) than Morris’s, which sharpens through the course of the late romances but does not vary, and Dunsany’s archaism proceeds from the King James Bible, not the Middle English prose romances published by Caxton. Place-names in Morris tend to be Anglo-Teutonic compounds (Upmeads, Oakenrealm) whose meanings are closely tied to the places; Dunsany’s are generally inventions that carry no linguistic significance. Both did synthesize conventions and motifs drawn from traditional narrative forms, but the specific sources that were important to them were quite different. Where the Icelandic sagas, with their spare, direct style, and Middle English prose romance were of crucial importance to Morris, the former were of no discernable and the latter only of sporadic significance to Dunsany; the King James Bible may be vaguely echoed at points in Morris, but no more. Morris’s sober, direct narrative style and his concern for practical things like precise geography is in stark contrast to Dunsany’s more elusive persona, with its ironies and asides and its tendency to fabulation. Dunsany’s work did achieve a certain popularity in its original context: his books sold, and some went into multiple printings. Nevertheless, they

TWENTIETH CENTURY

139

were not bestsellers, and his audience was not large. There was no attempt to create a commercial genre in his wake (as with Tolkien decades later), nor did his work spawn any consolidated critical appendage. He was not seen as representative of any broad literary movement, and contemporary reviews fail to mention William Morris: Dunsany was, simply, Dunsany. With the Great Depression and World War Two, literary tastes and fashions changed, most of this work went out of print, and when the BAFS began republishing it in 1969, most of it had been out of print for four decades or more. In the genre period, however, Dunsany has probably generated more extravagant praise than any of the other pregenre writers, excepting Tolkien. In the editorial apparatus of the Dunsany volumes of the BAFS, Lin Carter would routinely dub him “the greatest fantasy writer who ever lived” (3Carter ix, and elsewhere). The 1999 Del Rey Impact reprints of King and Charwoman sport a considerable array of contemporary writers voicing similar sentiments. James Stephens, in contrast to Dunsany, was orphaned and grew up in poverty in Dublin. He was a strong voice on Irish social and political issues, as well as a proponent of Irish independence, and most of his fiction, including all the fantasy work relevant here, takes Irish subjects. His three volumes reworking stories from Irish legend, which followed immediately on the heels of independence, can be seen—like the work of MacPherson, Scott, and Hogg—as partly patriotic in their motivation. Oddly, Stephens has not been accorded any particular attention from the standpoint of the post-1960s genre beyond passing reference to his most well-known work, The Crock of Gold (1912). David Pringle’s entry on Stephens in Clute and Grant’s Encyclopedia runs to a bare eight lines and does little save list book titles. However, as I have said, he does stand as a significant early twentieth-century progenitor of “Celtic Fantasy” and the first writer to produce literary adaptations of traditional Red Branch and Fenian tales as prose fiction. Crock is the earliest of Stephens’s relevant work here, with idiosyncrasies that skirt the borders of the BAFS template. The general setting is more or less contemporary with the author’s world, as indicated by references to city councilors and policemen, among other things. At the same time, the specific setting, the wood of Coilla Doraca and its environs, home to two Philosophers, carries more of the timeless feel of a fairy-tale world. The core of the story’s action centers around traditional Irish motifs, such as the theft of a group of leprechauns’ crock of gold and their subsequent abduction of the Philosophers’ two children, as well as the confrontation between the alien god Pan and the indigenous Angus Og. The effect of this is a curious reversal: the specifically modern world operates as part of the fantasy rather than as a foil to it. The texture of the narrative itself

140

THE EVOLUTION OF MODERN FANTASY

combines a strong subcurrent of (often mock) philosophical speculation with the directness of a children’s fairy tale. The three volumes adapting traditional Irish legends are the portion of his work most directly amenable to the BAFS template. The first of these, Irish Fairy Tales (1920), comprises ten short stories, half adapted from the tales of Fionn mac Uail and the Fenian warriors, the rest from various other sources.22 The second of the three is Deirdre (1923), based on the “Exile of the Sons of Usnech,” probably the most focused and wellintegrated of Stephens’s book-length fictions. The third is In the Land of Youth (1924), an intricate, stories-within-stories weave of various related tales into one continuous narrative. In structural terms, this was Stephens’s most ambitious work, but it was ultimately the least successful. The first part, framed by content adapted from “The Tale of Nera,” within which Maeve tells “The Dream of Oengus,” within which Bove tells “The Quarrel of the Two Pig Keepers,” is by itself fairly well balanced. The thread frays in the second part, however, which mainly consists of Maeve telling “The Wooing of Etain.” This part is overly prolonged, the frame never closes, and the book never wholly pulls together. These latter two volumes, both based on Red Branch material, were intended as the opening sections of a projected five-volume treatment of the Irish epic Tain bo Cuailnge, the chief parts of which were never written. Stephens’s narrative mode in these volumes is quirky, and he is, like Dunsany, first and foremost an ironist, though the quality of his irony is quite distinct from Dunsany’s. Stephens’s main focus is dramatic and centers on the dynamics of human interactions and motivations, particularly those concerned with relations between the sexes on an intimate, even domestic level. The narratives are notably antiheroic. At the same time, while the emphasis on motivation may be taken as “novelistic,” Stephens’s way of exploring motivation is not through interiority but through narrative speculation and extrapolation. The characters are not made into “real people” a la T. H. White but remain, as in the originals, larger than life. The reader does not overhear their thoughts but does hear their speech, which, like the narrative voice, is formal and artificial (if not archaic in the manner of Morris or Dunsany) and often builds directly on a Gaelic-founded syntax; the narrative voice is that of bard and redactor, frequently echoing the wording and rarely departing from the plot content of the originals, even on occasion noting conflicting accounts of events where there are multiple versions. With regard to magic, Stephens is quite unapologetic: he does not downplay, or soften, or rationalize magical elements. On the other hand, magic is not a theme, and a traditional tale with little magic (Deirdre, for instance) has little magic in Stephens’s adaptation.

TWENTIETH CENTURY

141

In his original context, Stephens did attract a readership, partly due to his association with the Irish Renaissance, and it is possible that this association has contributed to his neglect as a “fantasy writer.” And though he became, inevitably, “the author of The Crock of Gold,” he also built some reputation as poet and later as a radio personality; his Insurrection in Dublin remains a major eyewitness account of the events of 1919. But while The Crock of Gold has never gone out of print, his three volumes of adaptations of traditional Irish material have been largely ignored—oddly, since the works of Evangeline Walton, Kenneth Morris, and T. H. White have been embraced by the genre. His treatments of Irish legend deserve more attention than they have gotten, not least because they stand as a corrective to the “reduce to bare essentials of plot and character then rebuild as genre action fantasy with a dash of Tolkien” approach that has characterized too many more recent adaptations of Celtic material. Dunsany and Stephens’s contemporary, Kenneth Morris, was, of course, Welsh, but he had an early, influential association with some of the major figures of the Irish Renaissance, including Yeats, AE (George William Russell), and Ella Young, and he saw his Welsh books as an attempt to do for Celtic Wales what had been done for Celtic Ireland. Several of his earliest pieces were published in an Irish Theosophical periodical. As with Stephens, the settings of Morris’s tales, long and short, are not “invented” but mythological/legendary locales. As Stephens drew on Irish tradition, Morris drew frequently on Welsh tradition. Five of his six early stories, published between 1898 and 1902, represent Morris’s first grappling with his own heritage (one of the stories draws on Irish tradition)—a preoccupation that eventually resulted in his two most well-known works. The first of these, The Fates of the Princes of Dyfed (1914), is a reworking of “Pwyll Prince of Dyfed,” the First Branch of The Mabinogion. While the second part of the tale—the marriage of Pwyll and Rhiannon, and the birth, abduction, and coming of age of Pryderi—is notably amplified, including characters and events of Morris’s invention, the basic narrative movement remains true to the original. The second, originally drafted during the same period but not revised and published (in its cut form) until 1930, is Book of the Three Dragons. While centered mainly on the adventures of Manawyddan, the eponymous hero of the Third Branch of The Mabinogion, the story is largely of Morris’s invention. Both works are written in a highly mannered, elaborate, and archaized English, drawing heavily on Morris’s knowledge of Welsh for its syntax; the archaistic quality echoes Lady Guest’s translation of The Mabinogion. Morris’s preoccupation with Theosophy strongly informed his perspective on mythology, which he saw as narrating the “Journey of the Soul,” and both of the Welsh works operate on a distinctly metaphysical level. Characters are even more rooted in

142

THE EVOLUTION OF MODERN FANTASY

stock types than those of Dunsany or William Morris and represent stages of spiritual development. The Welsh geography maintains the shape of Wales but is symbolic rather than representational in its depiction, and the characters in both books frequently stray off the known map. However, Morris’s considerable literary skill is sufficient, and his metaphysics sophisticated enough, that one might say he teases this dimension out of his sources rather than heavy-handedly superimposing it onto them. Morris’s third book-length work, The Chalchiuhite Dragon (1992), was written in the late 1920s and early 1930s. Centered on the Toltec/Aztec legends of Quetzalcoatl, the story is focused on the childhood of the hero, not narrated in any surviving sources, and hence it is largely Morris’s invention. More novelistic than either of the Welsh works, his characters and narrative approach here assume a more notable realism in presentation. The language is more compact and nonarchaistic, while maintaining the formal, rhythmic cadences of his sentences. Morris’s collected short stories number about forty, and apart from the early six, they were published in various Theosophical periodicals between 1915 and 1930.23 Ten of these were collected under the title The Secret Mountain (1926); published by Faber and Gwyer, this was Morris’s only conventional “literary” release during his lifetime. In both subject matter and prose style, the short stories evince a variety not evident in the three longer works. Some are not in keeping with the BAFS template, though most remain amenable to it. Like Book and Chalchiuhite, most of the stories attach themselves to actual mythological traditions but are in fact Morris’s inventions. He ranges widely, from Welsh (“Sion ap Siencyn”) to Norse (“Regent of the North”) to Persian (“The Rose and the Cup”) to Chinese (“Red-Peach-Blossom Inlet”) subjects, and as with the longer works, the language becomes less archaized and more compact as he moves away from his homeland. In form, Morris shows remarkable control in varied areas, with stories modeled on the folktale, the wisdom tale, the dream vision, and others. In his context, Morris was easily the least known of the writers discussed here. Of the two book-length narratives published in his lifetime, Fates was put out by the Theosophical Press in Point Loma, California, where Morris taught for two decades; Book was brutally abridged and published as a book for young readers. With the lion’s share of his short stories restricted to the readership of Theosophical periodicals, the basis of his contemporary reputation was largely restricted to the ten stories in The Secret Mountain. Coupled with Morris’s apparent ambivalence about achieving recognition, none of this invited even the degree of attention drawn by Dunsany and Stephens.

TWENTIETH CENTURY

143

The revival of Morris’s work largely postdates the BAFS. The only selection Lin Carter included in the series, a brief excerpt from Book, was misleadingly presented as an excerpt from The Mabinogion in Dragons, Elves, and Heroes (1969), the first of two BAFS anthologies of “ancient fantasy.” Ursula Le Guin’s mention of Morris in her well-known essay “From Elfland to Poughkeepsie” (1973), as (with Eddison and Tolkien) one of fantasy’s three great stylists, clarified things a bit but yielded only a paragraph more of text. It was only in 1978 that Fates was included in the Newcastle Forgotten Fantasy Series and that the short tale “Red-Peach-Blossom Inlet” appeared in Boyer and Zahorski’s Fantastic Imagination II. Other stories appeared sporadically in anthologies throughout the 1980s, but the bulk of Morris’s work has become available only since the early 1990s, thanks mainly to the efforts of Douglas Anderson: Chalchiuhite appeared in 1992; in 1995, The Dragon Path collected all Morris’s known short fiction; in 2005, the complete text of Book appeared.24 Though Morris’s work has drawn increased attention over the last three decades, it deserves to be far better known than it is. His command of a remarkably cadenced yet highly variable prose stands him easily in the company of Dunsany, Cabell, and Eddison, as do his imaginative power and the exceptional philosophical cohesion that underscores all his work. Before the War: Cabell and Some Others The fourth of the major writers of the literary canon to begin publishing prior to World War One is the Virginian James Branch Cabell. With six volumes, Cabell was broadly represented in the BAFS, and those volumes cover the bulk of his relevant work.25 Some complexities arise as a result of Cabell’s involved, convoluted megaconstruction, the Biography of Manuel, which came to comprehend virtually all his work through the end of the 1920s, running to 18 volumes in its final version (1927–30). Properly speaking, the Biography comprises as much nonfantasy as fantasy, including poetry, literary reflection, ersatz genealogy, as well as fiction with no elements of the fantastic at all. This formal patchwork sometimes touches individual works: The Cream of the Jest juxtaposes a couple chapters in the invented medieval French province of Poictesme, with a primary focus on the domestic twentieth-century world of Felix Kennaston; several fantasy narratives are embedded within literary ruminations in Straws and Prayer Books (1924). Yet Cabell linked all this material, regardless of form, into one massive entity connected by an adeptly executed genealogical webwork, in which several lines of descendants are traced from thirteenth-century Poictesme to twentieth-century

144

THE EVOLUTION OF MODERN FANTASY

Lichfield (Virginia, many readers assume, but this is not stated).26 Conceptually, the fantasy becomes real, and the real becomes fantasy. In a sense, to isolate the fantasy components of the Biography is to decontextualize them. Nevertheless, some general observations can be made about those components. Most occupy the early, “medieval” portion of the genealogical construct.27 The more elaborate, archaized qualities of Cabell’s highly artificial prose style are most evident in these earlier works; all of them are pervaded by an unrelenting irony. The “invented world” here is in fact an invented French province, and characters do travel from Poictesme to France and England without dreaming. The ersatz genealogy, tracing the descendants of Manuel and Jurgen down to twentieth-century Lichfield (an invented town in the manner of Faulkner’s Jackson in Yoknapatawpha County), clearly locates Poictesme as a region of this world, within an identifiable historical period, regardless of Cabell’s tongue-in-cheek. In this, Poictesme is both more removed from (it is an invented region of France) and more connected to (it has a precise historical correlation) the real world than Stephens’s Ireland, Kenneth Morris’s Wales, or Dunsany’s Spain and England. The relation of Poictesme to our historical world led to the quasischolarly apparatus accompanying most of the medieval portions of the Biography, with discussions of the dissemination of the relevant “legends,” their versions, and the source of the “text” being translated. “It is a tale they narrate in Poictesme,” repeated countless times, underscores the idea that what we have is a piece of oft-repeated tradition; Cabell drew on the narrative devices of medieval romance—from plot motifs and structure to character types to verbal texture—to give his mock construct an “authentic” feel. In Figures of Earth, in a sense the cipher-text of the Biography,28 Manuel embarks on the quest to free the Lady Gisele from the magician Miramon Lluagor with the magical sword Flamberge; episodes are frequently centered around repetitions of three; and Manuel’s two verbal tags, “I am Manuel and must follow my own will and desire” and “I will travel to the ends of the world and judge them,” are repeated over and over. The simplistic, literal-minded Manuel—whom we meet shaping and reshaping a clay figure, because “my mother, sir, was always anxious for me to make a figure in the world . . . she wanted me to make myself a splendid and admirable young man in every respect” (Cabell 5)—recalls the similarly simplistic Perceval of Chretien de Troyes’s romance, who spends the early part of his narrative following his mother’s advice in an absurdly literal fashion. Following his mysterious disappearance from Poictesme at the end of Figures, Manuel the “Redeemer” is transformed into a messianic figure who will return, like King Arthur or Finn MacCumhail. On a more superficial level, Cabell meshes legendary and mythological figures from distinct

TWENTIETH CENTURY

145

traditions—Russian, Scandinavian, Celtic, Greco-Roman, and so on—into his own syncretistic construct. Cabell’s debt to the texts collected, edited, and elucidated by scholars since the eighteenth century is obvious—from content to style to his fiction of scholarly presentation. Like both Dunsany and Stephens, Cabell was primarily an ironic writer, tending to undercut the conventions he invokes: Figures operates in part as a functional parody of the heroic romance. As with Stephens, Cabell’s approach is not to develop novelistic characters via interiority, juxtaposing the psychological drama in the head of the protagonist with a reality shorn of romance. If anything, Cabell moves in the opposite direction: we never are made privy to what Manuel is thinking and aren’t even presented with Stephens-style speculations. In this, Cabell in fact grounds his story in old romance and epic techniques: Manuel is what he does (or appears to do) and says: “I am Manuel and must follow after my own will and desire.” The story purports, at points, to correct what “they tell” (and often this is less flattering to Manuel than “legend”), but what “really” happened is usually (1) as tenuous in its veracity as what “they tell” (it is “told” also) and (b) as extravagantly fantastic and often absurd. Magic is a presence in the relevant portions of the Biography, but as with Dunsany, it is as often as not presented in a mocking, tongue-in-cheek light—such as the canceling of the passage of time by moving the decimal point in The Music from behind the Moon. As an ancestor of the fantasy genre, Cabell’s accomplishment is certainly remarkable. With the possible exception of Eddison’s Zimiamvia trilogy, and even restricting consideration to the overtly fantasy segments, the Biography stands as the most extensive and intricately developed fantasy construct prior to the publication of LOTR. Cabell’s command of English prose, his development of a style uniquely suited to his material, and his deft facility with the vocabulary of heroic romance, fairy tale, and epic are exceptional, and the best of his work repays close reading a century later. In his context, he was probably the most well-known of the writers discussed here, primarily due to the notoriety that followed the trying of Jurgen on the grounds of obscenity in 1919, though his earlier work had already garnered admirers from Mark Twain to H. L. Mencken to Sinclair Lewis. But as with Dunsany, he was not perceived as part of any broad literary movement: Cabell was, simply, Cabell. With the shifts of taste that attended the Depression and World War Two, Cabell became unfashionable, and his subsequent work29 was less successful, commercially and artistically, than his earlier work, and gradually the earlier work went out of print. By the time Ballantine began their reprints in 1969, he was simply “the author of Jurgen.”

146

THE EVOLUTION OF MODERN FANTASY

There were some other noteworthy works to appear in the period prior to World War One. Richard Garnett’s The Twilight of the Gods appeared in a new edition in 1903, with 12 new stories not included in the original edition of 1888. Clemence Housman’s brilliant Arthurian novel, The Life of Sir Aglovale de Galis (1904), expanded from the brief passages concerned with Aglovale in Malory. Housman’s approach is more narrowly novelistic than the other writers mentioned here, in some ways anticipating T. H. White, though the language, poetic and often archaic in syntax, evokes the atmosphere of medieval romance effectively. Oddly neglected in discussion of fantasy,30 the 2000 Green Knight Press edition is the only edition from the genre period. William Hope Hodgson’s The Night Land (1912) was included in the BAFS and is often considered in the present company, but the futurist scenario has more in common with science fiction than the core material of the BAFS. The poet Henry Newbolt produced a lone fantasy, Aladore (1914). In the context of the period, the work is unique in that the influence of William Morris is readily apparent. The blurb for the 1975 Newcastle edition, “Here’s a book that William Morris could have written,” is not entirely inappropriate: the cadenced prose, archaized but straightforward, with syntax and diction rooted decisively in the Anglo-Germanic rather than the Latinate, recalls Morris sufficiently that unidentified passages might in fact be mistaken for Morris. Characters are drawn simply in broad strokes, rooted in stock romance types. The landscape, from Sulney to Paladore, is invented but by implication contiguous with medieval Europe: Aithne, the hero Ywain’s half-fairy love, is connected to Ireland and Broceliande; Ywain identifies her with the great mythic “loves” of European tradition (Helen, Isolde, etc.); the “Rhymer” mentioned over and over is undoubtedly the traditional Thomas the Rhymer of Scottish lore; the landscape is populated by hermits, archbishops, and so on. Nevertheless, Newbolt does not take characters or scenes from Morris and was undoubtedly influenced by many of the same sources Morris was: the relation is closer to that of Le Guin to Tolkien than of Terry Brooks to Tolkien. The otherworldly, metaphysical element, encapsulated in the “ideal” city of Aladore, which Ywain and Aithne bounce inexplicably back and forth to, runs counter to Morris’s practical, dramatic focus. The jaded Ywain of the opening, undergoing a kind of midlife crisis that leads him to reject the mundane responsibilities that gall him, is quite distinct from the by and large youthful, innocent protagonists of Morris’s romances. So it is doubtful that Morris would have written this book, though it is also doubtful that Newbolt could have written it without Morris’s example. Finally, descended from the quasi-Oriental tales of the eighteenth century, are the Kai Lung books of Ernest Bramah (Smith). The first Kai Lung

TWENTIETH CENTURY

147

story, “The Story of Yung Chang,” was published in 1896 and collected with eight others in the first Kai Lung volume, The Wallet of Kai Lung (1900). Like most of its followers, Wallet is a framed collection, its inset tales told by itinerant storyteller Kai Lung—often, as with Sheherezade, as a means of dealing with a life-threatening situation. The elaborate and highly artificial prose they are written in, characterized by a sustained, exaggerated, comic indirection, recalls both the baroque language of Burton’s translation of The Arabian Nights and George Meredith’s epigram-studded The Shaving of Shagpat. The setting is China, but despite the presence of mandarins and emperors, familiar allusions to Chinese place-names, and students undergoing rigorous examinations in pursuit of lucrative administrative positions, there is even less “China” here than there is “Spain” in Don Rodriguez. The prose points to Burton and Meredith; the actual narrative content, from the intricate, novella-length “The Transmutation of Ling” to the short pure-minded fool story “The Probation of Sen Heng,” draws liberally on motifs common to popular tales and romances throughout the European world. Hilaire Belloc wrote that Bramah employed the “Chinese idiom” (Bramah 6); it would be more accurate to say he employed an English idiom used to evoke the English idea of the Chinese. The profusion of money-hungry hucksters and disingenuous governing functionaries, the absurdly enforced etiquette, and the oftentimes impenetrably Kafkaesque bureaucratic apparatus suggest that the barbed irony of its author was, at any rate, more specifically directed at England. It was more than two decades before the appearance of the next Kai Lung volume, Kai Lung’s Golden Hours (1922), though three more would follow in the subsequent two decades: Kai Lung Unrolls His Mat (1928), The Moon of Much Gladness (1932),31 and Kai Lung beneath the Mulberry Tree (1940). Six stories published in Punch (1940–41) were collected as Kai Lung: Six (1974). The second and third volumes saw more developed and sustained frames than that of Wallet,32 though the frame of Mulberry Tree is again cursory, and Moon and the Punch stories are simply identified as “told by” Kai Lung. The tales themselves do not deviate from the mode of Wallet, though Bramah’s prose grew sharper over time. The chief departure was with Moon, a novel-length tale engaging detective fiction conventions with no noteworthy magic in it, which has led some to single it out as “not fantasy.” However, it is also true that many of the short tales in the other volumes do not include magic (beyond references to spirits and demons that are only references, also true of Moon). It would seem that, to Bramah, while Kai Lung stories might include magic, it was not a necessary feature. The Kai Lung stories had a small but devoted following during their author’s lifetime, and the first three volumes in particular have had new

148

THE EVOLUTION OF MODERN FANTASY

editions at regular intervals, mainly in England and not attached to the fantasy genre. But two Kai Lung tales did appear in the 1972 BAFS anthology, Discoveries in Fantasy, followed by a full-volume edition of Golden Hours (1972), “The Transmutation of Ling” from Wallet in Great Short Novels of Adult Fantasy Volume Two (1973), and the full Unrolls His Mat (1974). Like Dunsany, Stephens, and Cabell, Bramah was a thoroughgoing ironist; in the vein of the quasi-Oriental tale, he is the major writer from the pregenre twentieth century.33 After the War: Eddison and Some Others Most of the writers who began publishing in the prewar period continued into the 1920s, some with their best work,34 and the decade was in fact a rich one for fantasy. At the same time, there were a diminishing number of newer writers emerging, and Dunsany, Stephens, Morris, and Cabell had completed all their most significant work by 1930. Of the writers who did emerge after the war, E. R. Eddison is (apart from Tolkien) the most important.35 His most well-known work was his first, The Worm Ouroboros (1922), the most substantial fantasy between William Morris’s The Well at the World’s End and Tolkien’s LOTR. The core thematic focus of Worm is heroic action, though his treatment of the great war between Demonland and Witchland, and of the quest of Lord Juss to rescue Goldry Bluzco from an enchanted captivity, recalls the romantic epic rather than adventure fiction. Like Dunsany, Morris, and Cabell, Eddison’s prose is artificial and mannered, and its elaborately archaized Elizabethan/Jacobean constructions result in a style more difficult than those of his predecessors. It is, however, a vigorous language, rarely marked by excessive verbiage, and it has a controlled variability, ranging from the subdued and saga-like to the floridly dramatic. The world of Worm is invented: it is identified early in the story, to the fleeting Lessingham before his disappearance, as “Mercury,” though it seems probable that Eddison had an astrological rather than an astronomical conception in mind. The concluding appendix, “Argument: With Dates,” gives a compact, realistic sense of dramatic context more akin to Tolkien’s “The Tale of the Years” in the second appendix of LOTR (though far less extensive) rather than to the tongue-in-cheek of Cabell’s Lineage of Lichfield. Combined with Eddison’s precisely delineated geography, the effect is quite convincing, only marred by Eddison’s nomenclature, particularly of his nations: Demonland, Witchland, Goblinland, Impland, and so on. While Tolkien characterized Eddison’s nomenclature as “slipshod and often inept,” he continued that “I still think of him as the

TWENTIETH CENTURY

149

greatest and most convincing writer of ‘invented worlds’ that I have read” (1Tolkien). The texture of Eddison’s world and story, while invented, is nevertheless a syncretistic compound of varied elements. As noted, his prose is heavily inflected by Elizabethan and Jacobean idioms, and his fondness for the drama of the period, especially Jacobean tragedy, is readily apparent. His “epic” scope certainly evinces something of Homer, as well as of the Nibelungenlied and Italian Renaissance epic. This last, together with Elizabethan and Jacobean England, clearly supplied Eddison with much of his material culture. In addition to the Jacobean revenge tragedies, the Icelandic family sagas profoundly affected Eddison’s dramatic sense. Last, and less frequently noted, is something of the flavor of the old Irish tales, both in the exuberant, headlong daring of the heroes and in the vivid color of the descriptions. In his introduction to the 1926 edition, James Stephens would note, “His mind has more affinities with Celtic imaginings and method, and his work is Celtic in that it is inspired by beauty and daring rather than by thoughts and moralities” (Eddison xxi–xxii). The Icelandic-saga-inspired strand of Eddison’s imagination became dominant in his next work, Styrbiorn the Strong (1926), a quasi-saga centered on a protagonist built from passing mentions in Eyrbyggja Saga and Heimskringla. For various reasons, Styrbiorn is generally considered to stand apart from Eddison’s other fiction. The setting is not Eddison’s invention but historical in the manner of the nonmythic Icelandic sagas. At average novel length, it is considerably less expansive than Worm or the three Zimiamvia volumes. The language is, compared to those works, simpler and less archaized, if more mannered and ornate than the sagas that inspired it. Though ultimately placed within the world of Norse mythology, the novel itself incorporates little of the overtly magical into its main narrative. The last issue points to some of the fuzziness that emerges in applying genre period criteria to pregenre “fantasy” works. Lin Carter contended that Styrbiorn was not a “work of fantasy” but an “adventure of the Viking Age,” which he compared to H. Rider Haggard’s The Saga of Eric Brighteyes (1Carter 40–41). Nevertheless, Carter did excerpt the novel’s epilogue for the BAFS anthology The Young Magicians, which, set in Valhalla, he dubs a “fantasy scene.” But the implicit contextualization provided by that concluding “fantasy scene,” combined with the (admittedly sparse) supernatural touches, could be seen as moving the entire work into fantasy—which Haggard does not do in Eric Brighteyes (which also includes some minor supernatural touches).36 Certainly the “historical” saga age has elsewhere been a milieu deemed appropriate for fantasy.37 But Styrbiorn once again

150

THE EVOLUTION OF MODERN FANTASY

raises the question: must magic play an active, foregrounded role in a work for it to be deemed fantasy? Eddison’s next work was an actual saga translation: Egil’s Saga (1930). Thereafter, he spent the rest of his life on the work for which, after Worm, he is most remembered. The Zimiamvia sequence comprises two completed works, Mistresses of Mistresses (1935) and A Fish Dinner in Memison (1941), and a third, incomplete work, The Mezentian Gate (1958), published posthumously in a private edition. Individually, the first two are of expansive proportions—slightly shorter than Worm. For the last, Eddison had left behind an extensive “Argument,” a plot synopsis outlining the unfinished sections to enable publication of the fragment if it was not finished; completed according to the outline, it would have been longer than the first two Zimiamvia books combined. Zimiamvia is linked, if with nominal significance, to Worm. The Lords Juss and Brandoch Daha catch a fleeting glimpse of it from atop Koshtra Pivrarcha in Chapter XII of Worm, though it is not mentioned again; Lessingham, the fleeting viewpoint of the earlier book’s opening pages, becomes the central figure of Zimiamvia. In some respects, the world of the trilogy bears a resemblance to that of Worm: it is a syncretistic compound of, for the most part, similar elements (English and Italian Renaissance, Greek, etc.); the prose style is consonant with that of Worm. However, the trilogy is much slower moving, dramatically centered on painstakingly nuanced personal and political intrigue—the kind of thing that is restricted to a handful of the Witchland chapters in Worm—rather than on heroic action. Ultimately, the trilogy is closer to Sidney’s Arcadia than to Homer. Magic is an unambiguous element in Worm, playing an important role in many episodes, and two of the major scenes in the book surround Gorice XII’s conjurings (one successful, one not) in the Iron Tower of Carce. In the trilogy, however, magic is rarely evident and not a determinant of key events, connected primarily to the machinations of the philosopher Vandermast and reflecting on the metaphysical substructure of Zimiamvia itself. As noted, the worlds of both Worm and the trilogy are invented, and the precision of Eddison’s development of dramatic/historical context and geography certainly anticipate Tolkien and the post-1960s genre. However, it is evident that Eddison may not have seen his worlds in exactly the same practical terms presumed by the genre. For example, Greek deities and fauna drawn from medieval bestiaries appear without explanation or qualification, and characters refer familiarly to classic European literature, quoting from Shakespeare, Dunbar, and Webster. Entering with generic expectations, this may seem to constitute an odd blunder on Eddison’s part—a “flaw.” However, it is important to note that these invented worlds

TWENTIETH CENTURY

151

are tied to the this-world character Edward Lessingham. The connection is nominal in Worm, where Lessingham seems to be there only to get the reader to “Mercury.” However, in the trilogy, Lessingham is the major character, both in our world and after his death in Zimiamvia. A Fish Dinner in Memison in fact alternates between this world’s present and Zimiamvia. Ultimately, Zimiamvia emerges as a distillation of Lessingham’s ideals of beauty and perfection, created and given to him by Aphrodite. It is presented not as a speculative “other world” a la Narnia, not as an imaginary province of this world a la Poictesme, and not as a lost epoch of this world a la Middle-earth, but as precisely a world cobbled together from Lessingham’s ideal bits of this world. Eddison’s works did attract critical attention and some praise at the time of their publication, though the increasingly involved aesthetic/ philosophic preoccupations and difficult language precluded any genuine popularity. Worm inevitably remained the most widely read and recognized, reprinted in hardcover in 1952 and in paperback in Crown’s Xanadu Library in the early 1960s. Both Worm and the trilogy were republished by Ballantine as the first attempt to follow up on Tolkien, between 1967 and 1969, prior to the BAFS proper. When reviewers cast about in the mid1950s for modern work to compare LOTR to, Worm was probably the most popular choice. Nevertheless, as with Dunsany and Cabell, Eddison was effectively seen as a maverick rather than as an exponent of any identifiable genre or literary movement. Beyond Eddison (and Tolkien), work from the later 1920s onward begins to break away from some of the salient characteristics of the major canonical literary writers. Hope Mirrlees’s brilliant Lud-in-the-Mist (1927), included in the BAFS, straddles the human realm of Dorimare and Fairyland. The world is invented, but of modest proportions, and bears a certain impressionistic resemblance to Great Britain—both in its geography, physical and political,38 and in its folk culture, unmistakably built on its author’s antiquarian knowledge of sixteenth- and seventeenth-century England—though there is nothing in the book suggesting Dorimare is contiguous with England or Europe. But though the story’s pattern recalls that of Dunsany’s King of Elfland’s Daughter, with the initial piecemeal infiltration of the human sphere by Faerie culminating in the tide of Fairyland/Elfland over Dorimare/Erl, it is far more shaped by the novel in its approach to characterization and plotting than by archaic romance forms. The antiquarian element is evident in an artifact capacity rather than in narrative style. Mirrlees’s language, while styled and highly artificial, is novelistic, as is the intimacy of the narrator: the reader is more apt to be reminded of Jane Austen than

152

THE EVOLUTION OF MODERN FANTASY

William Morris. Neither of Mirrlees’s other two novels, Madeline (1919) and The Counterplot (1925), are fantasy. It would be appropriate to briefly mention David Lindsay’s remarkable A Voyage to Arcturus (1920), one of the pre-BAFS titles that were kept in print in tandem with the series subsequently. Lindsay’s invented world stands as one of the most vivid in twentieth-century literature, and the cohesion and intensity of its author’s metaphysical vision ranks with that of Kenneth Morris and George MacDonald, whose two visionary romances would seem to have been an influence. At the same time, the planet Tormance is constructed primarily in metaphysical terms, a la Lilith, putting Arcturus on the borderline of the BAFS template. The frame of Lindsay’s tale, the voyage through space in a ship propelled by “backrays,” would also technically class Arcturus as science fiction. Lindsay’s subsequent fiction saw its author clothing his visionary imagination in the garb of supernatural fiction in contemporary settings, of which The Haunted Woman (1922; included in Newcastle’s Forgotten Fantasy Series) and Devil’s Tor (1932) are notable. As we move into the 1930s,39 the imaginary provinces and worlds largely cease in literary fantasy outside of Tolkien and Eddison. Probably the two most important literary writers from this period are Evangeline Walton and T. H. White, who both, like James Stephens and Kenneth Morris, produced new interpretations of traditional stories. Evangeline Walton’s publication dates—like those of Tolkien—are deceptive. The only of her relevant work to be published during this period was The Virgin and the Swine (1936), her reworking of “Math Son of Mathonwy,” the Fourth Branch of The Mabinogion. The rest was published in the BAFS and after: following Ballantine’s new edition of Virgin as The Island of the Mighty (1970) came The Children of Llyr (1971), The Song of Rhiannon (1972), and Prince of Annwn (1974), reworking, respectively, the Second, Third, and First Branches of The Mabinogion. Another work, The Sword Is Forged (1983), was founded on the Greek legends of Theseus. A series of three short stories about the legendary, foundered island of Ys also appeared in the late 1970s and early 1980s. But all these were originally drafted during the years following the publication of Virgin. The poor sales of Virgin had resulted in no publisher interest in the subsequent three Mabinogion volumes; Sword was shelved, apparently, due to the appearance of Mary Renault’s treatments of the Theseus legends, The King Must Die (1958) and The Bull from the Sea (1962)—despite the fact that Walton’s interpretation bears little resemblance to Renault’s. Walton’s Mabinogion remains her most well-known and highly regarded work, and it is a bit closer to the practice of Stephens and Morris than White’s work. The language is comparatively spare and notable for its lack

TWENTIETH CENTURY

153

of archaism, but it is a formal prose nonetheless, with a rhythmic quality that evokes something of the flavor of the original. Though Walton augments the original a great deal in terms of narrative detail and character motivation, her treatment is only minimally novelistic: structure is determined by the original, to which Walton is far more scrupulously true than is Morris, and fantasy elements (such as the size of Bran in Children) are presented unapologetically and never rationalized. Though in terms of publication, three quarters of the series date to the BAFS period, the original composition places the entire series three decades earlier. T. H. White wrote a fair number of both short and book-length works that are fantasy in the broad sense but rather removed from the BAFS template.40 His reputation, and his relevance here, stems mainly from his Arthurian series, written in the late 1930s and early 1940s, though the overall publication history is more protracted. The first three sections were published independently: The Sword in the Stone (1938), The Witch in the Wood (1939), and The Ill-Made Knight (1940). Two more sections were written during this period, intended for an omnibus edition of the entire work, which was shelved due to the wartime paper shortage.41 When the omnibus edition, The Once and Future King, appeared in 1958, the first three sections were revised (the first two heavily): material from the unpublished fifth section was spliced into Sword, and some of the original material removed; Witch was virtually rewritten, chopped back to half its original length, and retitled The Queen of Air and Darkness. The fourth section, The Candle in the Wind, was revised to conclude the sequence. That the fifth section was left unrevised and dropped from the sequence suggests it was deliberately omitted by White, and it was only posthumously (1977) that it appeared as The Book of Merlyn. White explicitly casts the work as a sort of embellishment/commentary on Malory, and passages discuss Malory directly. Though White’s prose can rise to exceptional eloquence, particularly in the tragic conclusion, the voice is informal, even colloquial; occasionally the prose will play with Malory’s wording, but there is no attempt at a sustained “bardic” voice. Humor tends to the slapstick, in contrast to the mannered irony of Cabell, Dunsany, or Stephens. Characters are treated with a familiar realism: Lancelot is “Lance,” Guenevere “Jenny.” White’s “archaism” lies entirely in his extensive antiquarian knowledge of medieval material culture; his treatment of his material is very much in keeping with the modern novel. At their initial publication, the first three volumes of the series were notably more popular than any of the other work discussed here—with the possible exception of the “obscene” Jurgen. This was no doubt partly due to the perennial attraction of the Arthurian legends, but it was also most certainly due to White’s engagingly informal and intimate novelistic

154

THE EVOLUTION OF MODERN FANTASY

manner. The 1958 compendium edition was a bestseller and inspired a Disney adaptation of its first part, as well as a popular musical, Camelot. The work’s commercial success was clearly bolstered by the burgeoning fantasy genre of the 1960s, and Lin Carter declared it “the single finest fantasy novel written in our time” (2Carter 125). Some readers have had a more mixed reaction to White’s work, but he has not lacked enthusiastic readers: the stir created by the publication of The Book of Merlyn in the fall of 1977 was only overshadowed by the near-simultaneous appearance of The Silmarillion. Beyond Walton and White, the period shows little more than a scattering of relevant material. Some of the short stories of Eric Linklater, usually humorous engagements of Scottish and Orkney folklore, are relevant here;42 the book-length A Spell for Old Bones (1949), set in legendary, preRoman Scotland, comes closest of Linklater’s work to the BAFS template. Naomi Mitchison’s Travel Light (1952) begins in the world of Norse legend and moves into the world of early Christian Constantinople; To the Chapel Perilous (1955) is an idiosyncratic treatment of the legends of the Holy Grail, satirizing the modern press. Some of the work of the great Welsh novelist John Cowper Powys is of some interest here, though primarily by virtue of subject: Porius (1951, uncut 2008) is a highly unconventional Arthurian novel; The Brazen Head (1956) is centered on the thirteenthcentury philosopher/alchemist Roger Bacon; Atlantis (1954) and Homer and the Aether (1959) play off of Homer. But like White’s work, these are very much modern novels, with a good deal less of the overtly miraculous than The Once and Future King and with far more emphasis on character interiority than external event. Finally, Mervyn Peake’s three Gormenghast novels, Titus Groan (1946), Gormenghast (1950), and Titus Alone (1959), should be mentioned here. Like Eddison’s work and Lindsay’s Voyage, Ballantine republished all three titles before Lin Carter entered the scene for the BAFS proper, and they remained consistently in print in tandem with the series during its life. But as I noted in Chapter 1, Peake’s work, while one of the high points of midcentury English language fiction, really stands apart from the main focus here. Though the castle of Gormenghast and the dystopian world Titus wanders in the third volume are invented, Peake’s world is not derived from readings of medieval romance, Icelandic saga, or Homeric epic. The castle is “Gothic,” but it is not particularly archaic; apart from exaggeration and an influx of the inexplicable in one sequence of the first book, there is nothing that would aptly be termed magic. The dystopian world of the third book is, in its apparatus (cars, machines, etc.), contemporary to Peake. The narrative approach of the first two books is distinctly character based and recalls Charles Dickens and Laurence Sterne rather than Morris

TWENTIETH CENTURY

155

or Eddison. The third novel, a quest story in which Titus leaves the castle of Gormenghast, is structurally more akin to some of the work discussed here, but, again, its world is constituted of contemporary elements. Peake’s prose, colorfully poetic and vivid, does not evoke the quality of a “story told.” Despite the association reinforced by the BAFS and frequent studies of fantasy, Peake’s achievement is peripheral to the main focus here. So we arrive at the period that saw the publication of LOTR. However, before turning to Tolkien, a couple brief observations are in order. Excepting Eddison’s Zimiamvia (and Tolkien), the key works published by the major writers of the twentieth-century literary canon appeared between 1905 and 1928—a period of merely 23 years. The five writers whom I have dubbed the major writers were, in age, all within five years of each other. Without supposing a genre in the conventional sense, it might nevertheless be appropriate to speak of this body of work not as the linear father-toson type of tradition Carter depicts but as the product of a period. The preceding pages clearly suggest some notable common features—the formal, archaized prose styles; the recourse to archaic narrative structures; the frequently pervasive irony (among the earlier writers)—which, except in the case of Eddison, decrease notably in the overall diminishing body of writers emerging from the 1920s on. In fact, the relation between Dunsany, Cabell, and Eddison, on the one hand, and White, Mitchison, and Linklater, on the other, may be seen as accidental to a considerable degree. It is difficult to see much of anything, beyond legendary subject matter, in common between The Fates of the Princes of Dyfed and The Once and Future King. If White read Kenneth Morris (or Dunsany, or Eddison, etc.), it did not affect his literary practice. And again, excepting Eddison (and Tolkien), the amount of fantasy amenable to the BAFS template by any of those major writers after 1930 was virtually nill. Tolkien It is in this context, then, that we find Tolkien. In age, his is in the middle: 14 years younger than Dunsany, 14 years older than White. Publication dates for the work appearing in his lifetime would seem to connect him more to White’s generation: The Hobbit appeared in 1937 and LOTR in 1954–55. Of the shorter, non-Middle-earth fantasies, we find Farmer Giles of Ham (1949), “Leaf by Niggle” in Tree and Leaf (1964), and Smith of Wootton Major (1967). But the publication dates of these works are deceptive: The Hobbit was largely complete by the early 1930s; LOTR was conceptualized and begun around 1936–37. Of the shorter works, only Smith was written close to its publication date: as discussed, the original short version of

156

THE EVOLUTION OF MODERN FANTASY

Farmer Giles was written in the late 1920s, “Leaf ” in the 1940s. These dates push Tolkien’s work back closer to the period of the earlier generation. But it is Tolkien’s posthumously published work, and particularly the contents of the first five volumes of The History of Middle-earth, which most clearly settles his initial imaginative impulse in the earlier period. The Silmarillion was known of previous to 1977, but beyond the fact that it dealt with the Elder Days, the First Age of Middle-earth, and the fact that Tolkien considered it to be the conceptual bedrock of LOTR, precisely what “it” was, was unclear. The streamlined version that appeared in 1977 answered many questions but left others unresolved: When was it written? What was the exact nature of Christopher Tolkien’s involvement in shaping the final text? The subsequent Unfinished Tales (1980), with its fragments of alternate versions of constituent stories of 1977’s Silmarillion (as well as other material), suggested something of Tolkien’s practice of composing distinct but coexisting recensions of individual stories. Even so, the selections in Unfinished Tales all dated to the period following the completion of LOTR, leaving the pre-LOTR period still a mystery: what was Christopher Tolkien talking about when he alluded to “the earliest phase of the mythology, ‘the Book of Lost Tales’: itself a very substantial work, of the utmost interest to one concerned with the origins of Middle-earth” (1Tolkien 6)? What emerged quite clearly when the History began to appear at yearly intervals in 1983 was that the Silmarillion, broadly speaking, was not really a single, discrete work at all but a mosaic of fragments of distinct recensions of the matter attached to the Elder Days. It gave body to Christopher Tolkien’s allusions to a “wealth” of material predating The Hobbit and LOTR. The Silmarillion, as a broad narrative structure, had been at least partly conceived before World War One.43 The “Book of Lost Tales,” written between 1916 and 1920, constituted the first two volumes of the History (1983, 1984). This is a textually incomplete but fully conceptualized version of the narratives that came to constitute the “Quenta Silmarillion,” or Silmarillion proper, but here told by various Elves in a frame to Eriol, an Anglo-Saxon mariner who has voyaged into the Western Seas of Faery. Like much of the other fantasy from this period, the language is highly mannered and extremely archaistic.44 But most important, these two volumes demonstrate that the bedrock narrative structure underlying LOTR had, in its essentials, been established before Dunsany had turned to his booklength fantasies—before The Worm Ouroboros had been published. The succeeding three volumes of the History—The Lays of Beleriand (1985), The Shaping of Middle-earth (1986), and The Lost Road (1987)—include the material composed between 1920 and the later 1930s: extensive fragments of two narrative poems (and brief fragments of a few others), verse adaptations of “Lost Tales”; three successive castings of

TWENTIETH CENTURY

157

the core narrative material of the “Book of Lost Tales,” much condensed, without the frame, gradually transforming into the Quenta Silmarillion; and two castings each of two sets of “Annals” embodying the Silmarillion material. All this, along with some shorter works and the beginnings of the material attached to the Second Age (including the frustratingly fragmentary “Lost Road”), was in existence by the late 1930s. When Tolkien returned to this material after completing LOTR, he was in large measure recasting and revising rather than composing in the strict sense: first with the frustrated hope that Quenta Silmarillion would be published together with LOTR, then in answer to the growing demand that accompanied the latter work’s unexpected popularity. There were only three “new” Middle-earth narratives that were not embellished accounts of issues directly pertinent to LOTR, two of them brief fragments.45 Tolkien’s debt to nineteenth-century scholarship (which he was intimately involved with advancing on a professional level) and to the texts that had surfaced during the century plus before he embarked on “Book of Lost Tales” has been amply elucidated by Tom Shippey, Verlyn Flieger, and others. The editorial persona Tolkien developed is simply a fictional extension of what he did professionally. The Eddas, The Kalevala, Geoffrey of Monmouth’s Historia Regum Britannia, and others account for the formal character of the First Age material. Beowulf and Volsunga Saga, among others, are intrinsic in the more heroically centered material. The parallel but overlapping worlds of Elves and Men finds a clear antecedent in Ireland’s Tuatha de Danaan and Milesians; the Undying Lands to the West find a prototype in Tir-Nan-Og. The hobbits are a children’s fairy-tale invention, inspired by English rural people, who inadvertently walked onto a massive mythological canvas. On a more conceptual level, Tolkien’s understanding of “story” as a dynamic, changing entity—where any one story is in fact inclusive of many tellings, oral and written—also emerges from the traditional material he was preoccupied with. The story of Sigurd and the Volsungs, for example, is told in continuous form in Volsunga Saga, in short self-contained songs narrating single episodes in The Poetic Edda, and in compressed “handbook” form in Snorri’s Prose Edda. The latter part of the tale is the focus of the Old High German epic Nibelungenlied. Thidrek’s Saga weaves the tale into the broader context of the tales surrounding King Thidrek (Deitrich in Old High German). Later ballads from Denmark and the Faroe Islands narrate episodes connected to the tale. Literary quality may vary from version to version, but no one version is authoritative: each represents a different stage or strand of tradition. In addition, the surviving texts are only the tip of a massive iceberg: other written versions no doubt existed, and many

158

THE EVOLUTION OF MODERN FANTASY

versions existed in oral tradition. This conception was built into the stories of Middle-earth and is particularly evident in the legends of the Elder Days: over and over, the texts repeat that such-and-such incident or event “is told of in many songs,” and one can clearly infer that the fragmentary “Lays” represent the physical remains of this (fictional) tradition. In the heading of the typescript of the “Quenta,”46 it is noted that the narrative is “drawn from the Book of Lost Tales” (1Tolkien 77). The later version is, thus, not a revised draft intended to replace the earlier but a different version of the same material. This also creeps into LOTR: the lay sung by the minstrel at the Field of Cormallen, about “‘Frodo of the Nine Fingers and the Ring of Doom’” (3Tolkien 950), is different than the tale told in Minas Tirith, summarized by Ioreth of the Houses of Healing, in which Frodo “‘went with only his esquire into the Black Country and fought with the Dark Lord all by himself, and set fire to his Tower’” (3Tolkien 966). Both are different from the practical narrative penned by Bilbo, Frodo, and Sam. Tolkien’s engagement of techniques and conventions from traditional literature goes beyond simply the style and structure of individual tales to include the nature of interrelated texts itself. Of the writers discussed here, only Cabell moves in this direction to any notable degree, and Tolkien alone went so far as to produce variants. While it is naturally to be supposed that later versions of a story came closer to Tolkien’s ideal conception of it, that ideal did not necessarily supplant earlier versions as valid in their own right. Given the formative role played by LOTR in shaping the post-1960s genre, it may seem a bit gratuitous to ask how Tolkien’s work sits with the BAFS template. Nevertheless, the question is worth asking. With regard to magic, little qualification is needed: magic coheres with the work and is an intrinsic part of it, though—as with Eddison and William Morris—it is not systematized or extrapolated on a speculative level. But the question of the invented world does call for some qualification—if somewhat ironically, since Middle-earth stands as the quintessential invented world for many readers. Middle-earth (a medieval term designating the part of the cosmos inhabited by humans) was not conceived as a bona fide invented world but as representing an earlier epoch of our world. As noted previously, the “Book of Lost Tales,” written before Tolkien had conceptualized the intervening Second and Third Ages, was framed by the story of Eriol, a mariner of the Anglo-Saxon age, which was to merge with English history at the time of Hengist and Horsa.47 This became more removed and impressionistic as hobbits entered the scene, but Tolkien maintained his conception of the tales of Middle-earth as an imaginary version of the lost, pre-Christian myths and legends of England. In other words, Tolkien’s “invented world” is invented as William Morris’s were, not as Ursula Le Guin’s Earthsea is invented.

TWENTIETH CENTURY

159

To return to the issue of Tolkien’s context, though the publication dates of his work are inevitably late, his work is clearly rooted in the period of Dunsany, Cabell, and Eddison rather than that of White. At the same time, those late dates precluded any notable contemporary reputation: The Hobbit was a popular children’s book, but the First Age material was not published until 1977 and following, and the reputation generated by LOTR, of course, had its inception in the period just preceding the emergence of fantasy as a genre. It would not be amiss to end this section with some mention of C. S. Lewis. His role in Tolkien’s work was crucial: he enthused on the First Age material Tolkien shared with him when they met in the late 1920s, encouraged Tolkien to finish and publish The Hobbit, and provided Tolkien’s chief audience throughout the protracted composition of LOTR—without which, according to Tolkien, it may never have been finished. In several essays, notably “On Stories” (1947) and “On Science Fiction” (lecture 1955, publication 1966), he provided some astute commentary on material later to become canonical in the fantasy genre and on the critical shortsightedness contributing to its neglect. It is only in these belated commentaries of Lewis that we begin to see the hazy outlines of a perceived collectivity involving some of the writers discussed here. Most of Lewis’s creative work hovers on the edges of the BAFS template, but much of it is worthy of note. Though conceived, written, and published for children, and hence treated somewhat peripherally when the burgeoning “Adult Fantasy” genre was struggling with the debilitating preconception that fantasy was “kid stuff,” The Chronicles of Narnia (1950–56) is a deft embodiment of the key elements that came to define the genre, as well as the first body of work to evince the influence of Tolkien. Lewis is secondarily known for the “Space Trilogy” or “Ransom Trilogy”: Out of the Silent Planet (1938), Perelandra (1944), and That Hideous Strength (1945). Though placed within science fiction frames, a la Lindsay’s Arcturus, the first two books resemble much fantasy in the development and depiction of their worlds (Mars/Malacandra in the first, Venus/Perelandra in the second) and in the development of an overarching interplanetary “mythology.” Lewis’s allusions to Tolkien’s “Legendarium” in the trilogy give an added interest, fleeting though they are. Though a novelistic, first-person narrative, the retelling of the Cupid and Psyche myth, Till We Have Faces (1956), is of interest here, and it is possibly Lewis’s finest work of fiction. The early The Pilgrim’s Regress (1933) employs much of the vocabulary of the work discussed here, though it is painstakingly allegorized. Last, there is the aforementioned Dymer and the three other verse fantasies collected in Narrative Poems (1969).

160

THE EVOLUTION OF MODERN FANTASY

Conclusion: The Question of Influence This, then, is the post–William Morris literary branch of the BAFS canon. However, as I have suggested, its status as “tradition” or “genre” is primarily the product of hindsight. This is not to suggest that these writers existed in a vacuum, entirely isolated from each other: most of them were probably aware of most of the others; some were certainly very enthusiastic about some of the others’ work. James Stephens wrote introductions to Eddison’s Worm and Fish Dinner; James Branch Cabell wrote a brief preface to the American edition of Worm. Surviving correspondence yields further evidence of recognition and enthusiasm: Eddison of Cabell, Cabell of Dunsany, and so on. On the other hand, there was certainly no collective writerly community, and it seems that those who knew each other’s work perceived each other as individual writers rather than mutual practitioners of a common genre. Contemporary reviews do not suggest any sort of collective genre or literary movement. The closest one can come to any writerly community would the Inklings, and this only included Tolkien and Lewis.48 However, the BAFS did not canonize this group of writers on no basis, and there are indisputable common elements running through their work. The question is, to what degree are these commonalities the result of interinfluence, and to what degree do they result from other factors? My discussion puts considerably more weight on the latter. But it would be appropriate to explore the question of interinfluence. By influence, I mean a substantial and conscious shaping force, not simply a piecemeal echo or appropriation49 or a vaguely impressionistic and specious resemblance in tone or atmosphere.50 A detailed cross-comparison of all the canonical writers would create an excessive labyrinth, so I will restrict the following to two examples: coming earliest, and considering the frequent assertions of his influence, it would be appropriate to consider Dunsany’s influence on the writers here. Coming latest, one might suppose Tolkien to likely embody the cumulative influence of the earlier writers, so this will be the second example. The question of Dunsany’s influence has been variously assessed. On the one hand, we have Lin Carter’s assertion that Dunsany “molded the writing styles of virtually every important fantasy writer of the first half of the twentieth century” (4Carter vii). On the other hand, while Ursula Le Guin would call Dunsany the “most imitated, and the most inimitable, writer of fantasy,” she added, “I have never seen any imitation of Dunsany that consisted of anything beyond a lot of elaborate made-up names, some vague descriptions of gorgeous cities and unmentionable dooms, and a great many sentences beginning with ‘And,’” concluding that “Dunsany is

TWENTIETH CENTURY

161

indeed the First Terrible Fate that Awaiteth Unwary Beginners in Fantasy” (Le Guin 89). Le Guin does not deny Dunsany’s influence, but she does downplay its constructive significance. As pertains to the writers discussed specifically in this chapter, I would also downplay his significance. While Cabell enthused on Dunsany’s work, and might appropriate from Dunsany for his own purposes (see note 49), formative or determining influence is doubtful. Cabell’s most significant works were book-length narratives, and he had already written, among others, Jurgen and Figures of Earth before Dunsany turned to longer works in 1922. Neither of Cabell’s two major short fantasies, The Music from behind the Moon and The Way of Ecben, are at all “Dunsanian.” In general, Cabell’s character-based dramatic focus is fundamentally distinct from Dunsany’s highly descriptive narrative emphasis. The painstakingly intricate genealogical webwork underscoring Poictesme would be alien in Dunsany’s whimsical dreamlands. Beyond the common tendency to archaize, there is little likeness between Cabell’s erudite prose and the King James sonorities of Dunsany. Really all that is left is the specious possibility that Cabell’s reading of Dunsany inspired his move from the romanticized but mundane past of The Line of Love and Chivalry to the more fantastical Poictesme that began to emerge in Domnei. This is possible, but if so, the connection is little more than a hint, and the hint was entirely sublimated to Cabell’s already developed literary style. We do not get much further with Stephens, Kenneth Morris, or Eddison. Stephens, like Cabell, has a strong, character-based dramatic focus rather than the descriptive focus of Dunsany, and the Gaelic-tinged syntax of his prose is alien to Dunsany, as is his explicitly Irish content. As a writer of short stories, one might suppose Kenneth Morris more apt to Dunsany’s influence, but Morris’s style and thematic substance was already evident in embryonic form in his early stories, published before even The Gods of Pegana. The philosophic substrata intrinsic to all Morris’s work is alien to Dunsany’s aestheticized fantasias, as is Morris’s placement of his narratives within extant mythological traditions. The blurb writer for The Secret Mountain aptly observed, “The author’s nearest literary relative is perhaps Lord Dunsany, but he stands entirely on his own feet” (Morris 9). It is hard to suppose any substantive influence of Dunsany on the highly dramatic, Renaissance-flavored, expansive canvasses of E. R. Eddison. This leaves, in an appropriate transition point, Tolkien. John Clute, in his entry on Dunsany in Clute and Grant’s Encyclopedia, contends that “the secondary world created by J.R.R. Tolkien . . . took shape and flavor from Lord Dunsany’s example” (Clute and Grant 302). This is an oftechoed contention, particularly with regard to the matter of the invented mythology: the presumption seems to be that since each writer invented a “mythology,” and Dunsany’s came first, Tolkien must have been following

162

THE EVOLUTION OF MODERN FANTASY

Dunsany’s lead. Post hoc ergo propter hoc. But while Tolkien’s sporadic comments on Dunsany are of a familiar nature, suggesting he knew his work, they also tend to be critical. While many have extolled Dunsany’s skill in coining evocative names, Tolkien objected to the fact that they followed no linguistic pattern and were, hence, unrealistic (Carpenter 26). L. Sprague de Camp wrote that he sent Tolkien a copy of his 1963 anthology Swords and Sorcery, which included Dunsany’s story “The Distressing Tale of Thangobrind the Jeweler.” When de Camp met Tolkien on a subsequent trip to England in 1967, the story came up: “[Tolkien] spoke of ‘Dunsany’s worst style,’ especially at the end, where Dunsany, for the sake of a joke, pricked his own illusion” (de Camp 243). While one would suppose that there might be, to Tolkien, a better Dunsany style, still it seems Tolkien would have been alienated by Dunsany’s frequent flippancy. And as already noted, Dunsany’s disparate, dreamy, and often whimsical “mythic” narratives are of an entirely different order than the thematically integrated and meticulously structured mythological history of Tolkien. Tolkien’s mythology has far more in common with the Eddas, The Kalevala, and Heimskringla than with Dunsany, and Dunsany does not really seem a necessary step between the older works and the Silmarillion. Let us now turn to Tolkien’s relation to the other figures of the twentiethcentury literary canon and William Morris. Tolkien did acknowledge the influence of William Morris.51 The frame of “Book of Lost Tales” echoes that of The Earthly Paradise to some degree; Morris’s Wolfings and Burgdalers were acknowledged prototypes of the Rohirrim; Tolkien, like Morris, drew on Anglo-Teutonic rather than Latinate language forms, evident particularly in archaisms; Ralph’s “scouring” of Upmeads in the latter pages of The Well at the World’s End is hazily echoed in the hobbits’ “scouring” of the Shire at the close of LOTR. Nevertheless, Tolkien’s debt to Morris was selective and measurable, not along the lines of Morris’s influence on Newbolt’s Aladore, and entirely appropriated to Tolkien’s distinct purposes. With the actual twentieth-century writers, we begin to draw blanks. While The Worm Ouroboros was often compared to LOTR when the latter was published in the 1950s—and, in a rather baffling footnote in his Modern Fantasy: Five Studies, C. N. Manlove insinuates that Tolkien got it all from Eddison (Manlove 222)—Tolkien’s categorical statement that Eddison “was certainly not an influence” (Carpenter 258) cannot be seen, on the evidence, as disingenuous. While Tolkien admired Eddison’s literary skill, “I disliked his characters (always excepting the Lord Gro) and despised what he seemed to admire . . . I thought his nomenclature slipshod and often inept” (Carpenter 258). Points of comparison between the two books are really points of difference: the Gondor/Mordor antagonism, in its nature and depiction, is utterly unlike the Demonland/Witchland

TWENTIETH CENTURY

163

antagonism. Likewise, the quest of the Lords Juss and Brandoch Daha for Goldry Bluzco bears no resemblance, beyond a mountain being involved, to Frodo’s quest to destroy the Ring. It might be contended that the tone and texture of some of the non-hobbit-centered passages in LOTR owes something to Eddison, but this is rather specious and impressionistic, and Tolkien had developed his high, heroic tone in “Book of Lost Tales” years before Worm was published. Finally, the unabashedly aristocratic world of Worm, and even more that of the Zimiamvia books, could scarcely be more different than the world of the hobbits. Of the rest, Dunsany has been discussed. There is little resemblance between Stephens’s work and Tolkien’s, and it would not be surprising if Tolkien had never encountered Kenneth Morris’s work. It is probable that Tolkien heard of Cabell, though there is no published evidence that he had read him. I would venture the guess that he would have disliked Cabell’s work. Again, to proceed with the question of forward-and-backward lines of influence until all possible combinations have been exhausted would be a pointless exercise. Some of our writers may have found a generalized inspiration in some of the others’ work, but generalized inspiration is not notable formative influence—it is not following the others’ leads in more than the vaguest sense. Only two of the minor authors from the pre-1930s canon, Henry Newbolt and Donald Corley, seem to be strongly indebted to work from within the canon. The six major authors stand firmly on their own ground, their respective bodies of work distinct and unique. If there is a “tradition” here, it is a tradition of mavericks. My contention is that the common elements in the work of these writers stem not so much from mutual influence as from a common inheritance from the preceding century and a half. The writers would have grown up with Paflagonia, Gwyntystorm, and Pantouflia and are as likely to have gotten the idea of “invented countries” from these works as from William Morris’s prose romances. Likewise, the idea of subject matter rooted in traditional narrative conventions could have been suggested by the literary fairy tales for children; it was no doubt also suggested both by the subject matter of Romantic and Victorian narrative poets and certainly by the texts and translations of traditional material that, by 1900, had become a considerable library. The implicit syncretism underlying much of the work in “invented” settings was clearly facilitated by the proliferation of this material. The artificial, elevated, archaized prose styles, as I have discussed, follow the practice of both narrative poets and translators. The movement to prose itself probably reflects a number of things: the baroque, diffuse excesses of much Victorian narrative fantasy in verse had pushed the form

164

THE EVOLUTION OF MODERN FANTASY

into a rather static corner;52 recognition that much traditional narrative was in prose; most other forms of narrative were done in prose. Again, both the dates of the authors’ lives and the dates of the publication of the most significant work suggest that this “tradition” was really the product of a period—of a generation. The important material (Zimiamvia and LOTR) appearing much past 1930 is rooted in work that extended back to before World War One or is really of a fundamentally different literary genus (The Once and Future King). Tolkien stands as something of the little brother here, and the late appearance of LOTR takes us to the eve of the creation of the genre. However, the publishing apparatus that underlay that creation is tied to the popular branch of the twentieth-century BAFS canon, which I will now turn to. Works Cited Anderson, Douglas A. Introduction, in Morris, Kenneth. The Dragon Path: Collected Stories of Kenneth Morris. Edited by Douglas A. Anderson. New York: Tor Books, 1995. Baum, L. Frank. The Wonderful Wizard of Oz. New York: Dover, 1960. Belloc, Hilaire. Preface, in Bramah, Ernest. Kai Lung’s Golden Hours. New York: Crown Pubs., n.d. (c.1960). Bramah, Ernest. Kai Lung’s Golden Hours. New York: Crown Pubs., n.d. (c.1960). Cabell, James Branch. Figures of Earth. New York: Robert M. McBride and Co., 1925. Carpenter, Humphrey, ed. The Letters of J. R. R. Tolkien. Boston: Houghton Mifflin Co., 1981. 1Carter, Lin, ed. The Young Magicians. New York: Ballantine Books, 1969. 2Carter, Lin. Imaginary Worlds. New York: Ballantine Books, 1973. 3Carter, Lin. “About At the Edge of the World and Lord Dunsany: The Dreams of Mana-Yood-Shushai,” in Dunsany, Lord. At the Edge of the World. Edited by Lin Carter. New York: Ballantine Books, 1970. Clute, John, and John Grant, eds. The Encyclopedia of Fantasy. New York: St. Martin’s Press, 1997. de Camp, L. Sprague. Literary Swordsmen and Sorcerers: The Makers of Heroic Fantasy. Sauk City, WI: Arkham House, 1976. Eddison, E. R. The Worm Ouroboros. New York: Ballantine Books, 1967. Le Guin, Ursula K. “From Elfland to Poughkeepsie,” in The Language of the Night: Essays on Fantasy and Science Fiction. Edited by Susan Wood. New York: Berkeley Books, 1982. Manlove, C. N. Modern Fantasy: Five Studies. New York: Cambridge UP, 1975. Morris, Kenneth. The Dragon Path: Collected Stories of Kenneth Morris. Edited by Douglas A. Anderson. New York: Tor Books, 1995. Stephens, James. Introduction, in Eddison, E. R. The Worm Ouroboros. New York: Ballantine Books, 1967.

TWENTIETH CENTURY

165

Tolkien, Christopher. Introduction, in Tolkien, J. R. R. Unfinished Tales. Edited by Christopher Tolkien. Boston: Houghton Mifflin Co., 1980. 1Tolkien, J. R. R. The Shaping of Middle-earth. Edited by Christopher Tolkien. Boston: Houghton Mifflin Co., 1986. 2Tolkien, J. R. R. The Return of the King. New York: Ballantine Books, 1965. 3Tolkien, J. R. R. The Lord of the Rings. Boston: Houghton Mifflin Co., 2004.

Other Secondary Work Consulted Attebury, Brian. The Fantasy Tradition in American Literature. Bloomington: Indiana UP, 1980. Carpenter, Humphrey. The Inklings. New York: Ballantine Books, 1981. Carpenter, Humphrey. Tolkien: A Biography. Boston: Houghton Mifflin Co., 1977. Mendlesohn, Farah, and Edward Jones. A Short History of Fantasy. London: Middlesex UP, 2009. Douglas A. Anderson’s editorial apparatus in the Kenneth Morris volumes listed in Works Cited and in Primary Source Editions is invaluable; Paul Edmund Thomas’s introductions and notes in the editions of E. R. Eddison’s work listed in Primary Source Editions are likewise invaluable.

Some Primary Source Editions A large proportion of the work discussed appeared in the BAFS or the Newcastle Forgotten Fantasy Library, as noted in the text. Many of the works have subsequently appeared in multiple editions, an exhaustive list of which would be of prohibitive length. What follows are some of the more notable editions and collections from the post-BAFS period. Anderson, Douglas A., ed. Tales before Narnia. New York: Ballantine Books, 2008. Anderson, Douglas A., ed. Tales before Tolkien. New York: Ballantine Books, 2003. Dunsany, Lord. In the Land of Time and Other Fantasy Tales. Edited by S. T. Joshi. New York: Penguin Books, 2004. Eddison, E. R. The Worm Ouroboros. New York: Dell Publishing, 1991. Eddison, E. R. Zimiamvia: A Trilogy. New York: Dell Publishing, 1992. Morris, Kenneth. Book of the Three Dragons. Cold Spring Harbor, NY: Cold Spring Press, 2004. Morris, Kenneth. The Chalchiuhite Dragon. New York: Tor Books, 1992.

6

Twentieth Century Popular Fantasy

Introduction

T

he writers constituting the popular strand of the twentieth-century BAFS canon are H. P. Lovecraft, Clark Ashton Smith, and Robert E. Howard, whose main market was Weird Tales during the Farnsworth Wright editorship (1924–40); Fritz Leiber, Fletcher Pratt, and L. Sprague de Camp, whose relevant work was initially published in Unknown/Unknown Worlds (1939–43); and Jack Vance and Poul Anderson, who published single works of fantasy in the 1950s, the period when the dominance of the pulps was giving way to the pocket paperback medium. Edgar Rice Burroughs and A. Merritt were often spliced into this context. The reasons for looking at this as, in crucial respects, a separate phenomenon from the focus of Chapter 5 can be outlined, beginning with factors attached to publishing venues. The core majority of these writers were publishing for popular periodicals. This meant that, first, short form was the norm. Weird Tales was particularly centered on short stories, with an occasional novella-length feature; Unknown published both short stories and novellas. Even in the latter case, what were midlength books to the literary writers—Figures of Earth, Don Rodriguez, Lud-in-the-Mist—would have been extremely long. The popular work operated with fundamentally different architectural expectations. Second, the popular writers were writing for markets with defined foci and therefore expectations that needed to be (at least nominally) met. The literary writers, by contrast, functioned more as self-defined mavericks. A larger proportion of the popular writers were professionals who relied (often precariously) on what they earned by their writing, reinforcing the need to adhere to perceived expectations and also the need to produce rapidly. The pulps occupied an ephemeral literary ghetto, and writers,

168

THE EVOLUTION OF MODERN FANTASY

regardless of quality, received one-time payment for work that would (as it then seemed) disappear with the issue of the magazine. By contrast, a minority of the literary writers relied on their writing to support them.1 The disparities in circumstances of publication and the nature of publishing venues naturally created differences in the work produced. In addition, the popular writers were, as a group, well-read, literate, and intelligent, but few had the kind of focused, scholarly interest in medieval and ancient literature that their “literary” compeers did.2 Their work does not, on the whole, seek to “emulate” Malory, Volsunga Saga, or the Fenian tales—to echo linguistic and narrative textures—but to adapt elements of content from such works to modern forms of popular romance. Those “modern forms of popular romance” were, of course, largely determined by the focus of a given periodical, though a writer submitting to, say, Weird Tales would likely already have some interest in horror fiction. In general, the work of the popular canon is far more determined by its proximity to horror fiction, action adventure fiction, and science fiction than by its debt to The Mabinogion or Beowulf. The imaginative allure of lost cities and forgotten empires, spurred by the archaeological excavations in the Middle East and elsewhere, was more significant here than the literary textures of ancient poems. As far as specific interinfluence between the literary and popular writers, I will address some of this in discussion of specific writers. But simply in terms of dates, possible influence moving from the popular sphere to the literary would be negligible. From the other direction, some influence from some of the literary writers (Dunsany, Cabell, and Eddison in rapidly descending significance) can be discerned in some of the popular writers’ work, but this must be seen within natural limitations. Something on the scale of The Worm Ouroboros, or even Jurgen, would have been inherently unworkable in the popular context.3 Elaborate, archaized, poetic language, a common feature of the literary work, was unwelcome in the pulps. This would impede on the degree and nature of the influence of even Dunsany, who is often accorded great influence here, especially on the Weird Tales writers.4 More can be seen of intrainfluence among the popular than the literary writers, and fairly tight writerly communities grew up around both the Wright-period Weird Tales and Unknown.5 At the same time, there was no consolidated fantasy genre along the lines of what was established by the BAFS, and what were, to the writers, defining features of their work would more likely have reflected their specific markets. The emphasis of Weird Tales was on what we would now call horror, not on stories set in invented magical worlds. The emphasis of Unknown was more speculative, after the manner of science fiction. Most of the work of Lovecraft and Smith not

TWENTIETH CENTURY

169

in keeping with the BAFS template tends toward horror. Leiber, Pratt, and de Camp were fairly prolific science fiction writers. The popular fantasy “tradition” through the Weird Tales/Unknown period breaks down into three fundamentally distinct groups: (1) the proto–Dark Fantasy of Lovecraft and Smith; (2) the swashbuckling proto– Sword and Sorcery adventure fiction of Howard and Leiber; and (3) the more speculative fantasy of the Pratt/de Camp collaborations. I will discuss each of these in turn, followed by an account of the post–World War Two period, including Jack Vance and Poul Anderson, which will take this study back to its beginning point. Dark Fantasy and Weird Tales: Lovecraft and Smith Lovecraft, Smith, and Howard, though they sold to a wide array of periodicals, found their most consistent market in Weird Tales.6 Weird Tales, again, specialized in stories we would now be apt to call horror, the latter-day descendant of the eighteenth-century Gothic and its progeny: nineteenth-century ghost stories and tales of the supernatural, Edgar Allan Poe, French Romanticism and decadence, Arthur Machen and Algernon Blackwood, and William Hope Hodgson. It is here, rather than in Ossian, Keats, and William Morris, that we find the significant ancestors of Lovecraft and Smith.7 Lovecraft’s significance here is founded on a small portion of his work, most of which was not published in Weird Tales.8 This material, comprising the “Dreamlands” cycle of stories, is fairly meager in quantity. The invented world is a dreamland, which, like the Pegana of Dunsany’s “Idle Days on the Yann,” operates on a collective level: other dreamers besides frequent protagonist Randolph Carter enter and explore it, and it has (more so than Pegana) a fixed, mappable geography. The exotic place-names—Ulthar, Celephais, and so on—also recall Dunsany, and the atmosphere is broadly “Dunsanian.” The short novel The Dream-Quest of Unknown Kadath was written in 1926–27; the “Dreamlands” setting was otherwise restricted to a small number of stories, all fairly brief, written in close proximity to Lovecraft’s first encounter with Dunsany’s work in 1919. These include, most notably, “The Cats of Ulthar,” “The Doom That Came to Sarnath,” and “Celephais,” all published initially in amateur journals before much later (posthumously with the latter two) finding places in Weird Tales. Other associated tales, including most of the handful centered on Randolph Carter, move away from the “Dreamlands” setting per se, and some, such as “The Silver Key” and “The Strange High House in the Mist,” are in fact more along the lines of Gothic supernatural tales and removed from

170

THE EVOLUTION OF MODERN FANTASY

the BAFS template. Again, the overall quantity of this material is meager: the two BAFS Lovecraft volumes, The Dream-Quest of Unknown Kadath (1970) and The Doom That Came to Sarnath (1971), include ample material both outside the BAFS template and (in the second volume) unrelated to the “Dreamlands” cycle. Lovecraft’s fantasy is usually equated with his Dunsanian work, for obvious reasons. But the nature and extent of Dunsany’s influence on Lovecraft is worth considering. As I have noted, the influence of Dunsany on the “Dream-Quest” and a few of the shorter tales can be seen in the exotic names, in the collective nature of the Dreamland, and in the atmosphere. At the same time, Lovecraft made no attempt to mimic the King James sonorities of Dunsany’s prose, nor is Lovecraft (though not humorless) an ironist in Dunsany’s manner. Even those tales most akin to Dunsany’s early work carry a heavier tone, a strong undercurrent of alienation, and a more marked tendency to the grotesque and the bizarre—all characteristic elements of Lovecraft’s non-Dunsanian literary heritage. It is interesting that Lovecraft in fact wrote a “Dunsanian” story (“Polaris”) in 1917, two years before discovering Dunsany, suggesting that other impeti were at work in his fantasy: Poe’s “Eleanora” and “Silence: A Fable” are likely ancestors; the New England Gothic of Nathaniel Hawthorne stands behind Lovecraft’s New England, which permeates the “Dreamlands” material as much as it does the more properly Gothic horror work of the Cthulhu cycle. It is also interesting that one of Lovecraft’s major debts to Dunsany is evident in his least “Dunsanian” work: behind the pantheon of the Cthulhu mythos is Dunsany’s pantheon of Pegana. Lovecraft appropriated selectively to his own ends: Dunsany’s preoccupation with heroic legend and the fairy tale did not particularly seep through into Lovecraft’s work; the idea of a collective “Dreamland” did. A higher proportion of Clark Ashton Smith’s output, a considerably greater quantity overall, stands pretty squarely within the BAFS template. Four full volumes of Smith’s work appeared in the BAFS;9 sufficient further material remained that at least another volume would likely have appeared had the series continued. Most of Smith’s relevant material (all short stories) belongs to “cycles” set in particular locales, though close dramatic interconnections within these are minimal. The most extensive of the cycles, set on the far future continent of Zothique, occupied the entirety of the first Ballantine collection. A slightly less extensive cycle, set on the ancient, legendary polar continent of Hyperborea, filled most of the second volume. Other, slighter cycles, centered on Poseidonis (an island of Atlantis), the planet Xiccarph, and the medieval province of Averoigne. Nearly all Smith’s more than one hundred tales were written during less than a decade, between the late 1920s and the mid-1930s.10 However,

TWENTIETH CENTURY

171

the stories were not dashed off. According to Donald Sidney-Fryer, “Smith characteristically wrote about five drafts of each story . . . he would subject each draft to considerable alteration and correction . . . taking the manuscript with him on a stroll and reading it aloud” (Sidney-Fryer 1–2). The preoccupation with language in his fiction was inherited from his earlier work: under the mentorship of George Sterling, Smith had published three poetry collections11 and built a minor reputation in the western United States on their basis. The fiction follows the poetry, both in its nuanced language and in theme and atmosphere. Smith’s language, however, while having frequent recourse to archaic constructions, does not emerge out of a medieval Anglo-Teutonic diction and syntax, or Welsh, or the King James Bible. Closer models figure here: the dense, evocative textures of Poe’s “The Masque of the Red Death,” Beckford’s Vathek,12 Flaubert’s Salammbo, Baudelaire,13 and the ornate, elaborate style of Burton’s translation of The Arabian Nights. Rooted in Poe and Beckford, Smith’s stylistic influences merge far more strongly with French Romantic decadence than those of the literary writers discussed here. As with Lovecraft, then, Smith’s literary ancestry descends from the Gothic and its subsequent offshoots. Within the context of his exotic invented settings, Smith’s thematic and aesthetic substance was grounded in the grotesque and the bizarre; most of his stories are indebted to the horror tale in structure and emphasis. But while Smith’s inspirations emerged from some wellsprings common to Lovecraft, they were not identical. Smith owed nothing to Hawthorne’s New England Gothic, and Lovecraft owed little to the post-Poe French Decadents. The “Oriental Gothic” of Vathek only sporadically touched Lovecraft’s work. Although few of Smith’s tales are narrowly quasi-Oriental, a large proportion of his fantasy, including the Zothique cycle, is thoroughly infused with an aesthetic drawn from Beckford and Burton. While a cursory reading of Smith might register “Dunsanian,” the resemblance is limited and superficial: the dark, dense textures of Smith’s prose are quite distinct from Dunsany’s ironic King James constructions.14 Nevertheless, despite these differences, Lovecraft’s slight “Dreamlands” corpus and Smith’s more substantial body of work belong to the same, broad line of literary descent and together form a block of the popular fantasy canon in which the main focus of Weird Tales would seem to be closer to the writers’ conceptions of their own work than the BAFS template, though this work is largely compatible with that template. But while Lovecraft and Smith loom large, the vein they mined was fairly narrow and was not much followed. The influence of Lovecraft emerged from his more famous horror stories, including the Cthulhu mythos cycle, and he is far more significant in the horror genre than in fantasy. Smith’s more

172

THE EVOLUTION OF MODERN FANTASY

meticulously wrought tales were perhaps too styled and too idiosyncratic to spawn a “school” as Lovecraft’s horror stories did, particularly in the context of the pulps. While Zothique would seem to have partly served as the inspiration behind Jack Vance’s later “Dying Earth” milieu, further influence is difficult to track. Sword and Sorcery: Howard, Leiber, and Some Others Robert E. Howard is generally credited as the de facto originator of Sword and Sorcery in its most strict form. Quite distinct from the work of Lovecraft and Smith, and in fact rather aside from the main focus of Weird Tales, these stories nevertheless incorporated enough of the “weird” through evil magicians and ghoulish monsters to suit the magazine, and they became quite popular. Howard was an enormously prolific writer within the confines of a brief writing career, and in addition to his proto–Sword and Sorcery tales, he produced westerns, boxing stories, horror stories, and historical fiction. The core S&S work comprises the tales of Bran Mak Morn the Pict (some half dozen tales, plus fragments) and of King Kull of Atlantis (some ten tales, plus fragments), written mainly between the later 1920s and the early 1930s, as well as, most famously, those concerning Conan the Barbarian, the warrior-king of Howard’s ancient, invented Hyborian Age, written from the early 1930s until Howard’s death in 1936. These last comprise some 21 tales (several novella- and one novel-length), plus fragments, sketches, and a “historical” essay, “The Hyborian Age.” While Howard’s stories shared the imaginary ancient age settings of Smith, the development was entirely distinct: where Smith was preoccupied with generating a rich, poetically textured atmosphere, Howard was interested in a combination of swashbuckling heroics and quasi-historical verisimilitude. Howard’s stories are unambiguously action driven, and his heroes are invariably powerful and able warriors whose strength and martial abilities are often their means of dealing with difficulties. The prose is direct and unadorned—heavily naturalistic, almost hardboiled, with none of Smith’s poetic texture and only occasional passages of Lovecraftian atmosphere building. The quasi-historical dimension, which became particularly pronounced in the Conan series, underlined by the “historical” essay, reflected something of Howard’s literary inspirations and aspirations: “There is no literary work, to me, half as zestful as rewriting history in the guise of fiction” (1Howard xiii). In a prefatory note to “The Hyborian Age,” Howard wrote that his “history” was intended “to lend him [Conan] and his sagas a greater aspect of realness . . . I have never violated the ‘facts’ or spirit of

TWENTIETH CENTURY

173

the ‘history’ . . . but have followed the lines of that history as closely as the writer of actual historical-fiction follows the lines of actual history” (2Howard 381). Howard’s literary ancestry stretches back to a different nineteenth century than that behind Lovecraft and Smith, and much of that ancestry is not fantasy at all. The adventure-cum-historical-romance finds its major wellspring in the widely read novels of Walter Scott, which spurred the work of, in varying capacities, James Fenimore Cooper, Alexander Dumas (very popular in English translation), Robert Louis Stevenson, and many others. Some of these turned to eras often associated with fantasy but developed according to Enlightenment notions of “historicity” and “reality”: Scott’s Ivanhoe (1820) turned to the Middle Ages; Bulwer-Lytton’s The Last Days of Pompeii (1834) turned to the ancient Roman Empire. The importance of this nonfantasy work to Howard should not be underestimated. The bridge between this nineteenth-century nonfantasy and Howard begins with H. Rider Haggard, whose “Lost Race” romances are in part an outgrowth of the historical romance.15 Haggard’s romances do concern themselves with history, and they do represent an extension of the Romantic depictions of the distant past that run through many earlier historical romances. Through the device of “lost cities” and “lost empires”—which were inevitably offshoots of ancient civilizations in remote regions (the African interior, central Asia, etc.) that formed cultural cul-de-sacs that preserved elements of those ancient civilizations intact—the ancient world was, in effect, brought into the nineteenth century. Ayesha, the goddesslike “She-Who-Must-Be-Obeyed” of She (1886) and its several offshoots, was an ancient Egyptian priestess of Isis, who narrates her early history in Wisdom’s Daughter (1923). Haggard obviously runs considerably afield from actual history in these tales, but much of their attraction at the time was akin to the attraction of historical adventure fiction. The romantic fascination with the past had been further whetted, during the second half of the nineteenth century, by archaeological investigation and discoveries in Egypt, Mesopotamia, Mexico, and elsewhere of ruined temples and cities and by the European colonial myth of the “Dark Continent.” Haggard’s depiction of the convergence of antiquity with the nineteenth century in “remote” regions hidden from “civilization,” crossed with an attractive dose of the mystical and the supernatural, curses and destinies, proved commercially irresistible: King Solomon’s Mines and She were bestsellers, spawning many sequels and offshoots. Haggard was widely imitated (and parodied), and the subsequent “Lost Race” adventure tales that proliferated throughout the first half of the twentieth century would fill a library. Nearly every corner of the globe “remote” from the Eurocentric perspective was exploited, and the basic

174

THE EVOLUTION OF MODERN FANTASY

pattern of Haggard’s initial two bestsellers—bold adventurer travels to unknown region and discovers forgotten people and (usually) some mystical secret—repeated itself ad infinitum. The significance to Howard was twofold: (1) the straightforward emphasis on adventure provided a (not the only) narrative model, and (2) the settling of the imaginative impulse within a context of presumed historicity spoke to Howard’s interest in history. A number of Haggard’s romances eschew the nineteenth-century point of reference and narrate events set entirely in the past—for example, the aforementioned Wisdom’s Daughter, The Saga of Eric Brighteyes (1890), and Cleopatra (1889). It is a short step from these tales to Howard’s “Hyborian Age.” Haggard fell into a related-but-tangential position within the BAFS, which reprinted two of his works,16 though eight titles were included in the Newcastle Forgotten Fantasy Library (fully one third of that series). Also in the line of Haggard’s influence, C. J. Cutliffe Hyne’s The Lost Continent (1899/1900), an adventure story of the last days of Atlantis, was included in the BAFS; Edwin Lester Arnold’s Phra the Phoenician (1890), whose protagonist passes through a series of awakenings from death from the time of Julius Caesar to the time of Queen Elizabeth, was reprinted by Newcastle. The decades between Haggard’s initial work and Howard’s work at Weird Tales saw notable related work by Talbot Mundy, E. Charles Vivian, and the American writer A. Merritt. The latter’s The Ship of Ishtar (1924) is perhaps the closest to the BAFS template. But Haggard’s work also led to offshoots more in the direction of science fiction.17 The ancient time referenced in Arthur Conan Doyle’s The Lost World (1912), for example, is “scientific”: rather than goddess figures from the ancient Mediterranean world, the modern world crosses with remnants of the dinosaur age. I will return more generally to science fiction in the next section, but here, probably the most significant stepping stone between Haggard and Howard—Edgar Rice Burroughs—likewise tends to the science fictional. Burroughs’s most famous work, the Tarzan series, set in an Africa replete with forgotten cities and remnants of Atlantis, is an exception. His other series, with locations ranging from the Lost World–style island of The Land That Time Forgot (1918), to the “Hollow Earth” setting of the Pellucidar series, to the planet Venus, are more apt to be considered in terms of science fiction. Nevertheless, the John Carter series, which takes its protagonist to Mars (or Barsoom), operates in many ways in terms we would associate with fantasy:18 the journey to Mars accomplished by something akin to astral projection; the ancient-style civilizations (replete with ruined cities) whose warriors tend to be swordsmen; battles with monsters; and so on. The John Carter tales can be read as Sword and Sorcery tales set on another planet.

TWENTIETH CENTURY

175

Like the work of Haggard, this later work, including that of Burroughs, tends to be seen as a fringe phenomenon in relation to the fantasy genre. Nevertheless, the basic elements of the BAFS template are usually present in at least qualified form. While wholly secondary worlds are restricted to other planets (which, of course, remain within the present cosmos) and are therefore apt to be considered science fictional, locations such as Kor in She are as much “invented landscapes” as Cabell’s Poictesme. Hyne’s Atlantis is as much a fantasy milieu as Clark Ashton Smith’s. The usual mystical/supernatural elements would certainly fill the magic prerequisite: Ayesha’s unleashing of the “Powers” against Atene in Ayesha: The Return of She (1905) is a quintessential fantasy scene. The compromising elements would probably be the too-solid connections to actual history, the moreoften-than-not contemporary settings, and the not infrequent recourse to rationales for phenomena that carry (to postgenre readers) science fictional overtones. Fringe or not, this material had a strong impact on popular fantasy,19 and along with the historical adventure fiction of the nineteenth century, it forms the literary ancestry of Howard and the S&S subgenre. Howard’s Hyborian and Atlantean Ages can be seen as an outgrowth of the lost races of Haggard and his followers; as the latter are invariably connected in some fashion to actual history, so Howard’s Picts, Cimmerians, and so on clearly and intentionally in their names suggest actual historical peoples. His development of protagonists to serve as a glue for series of stories follows Allan Quatermain; in a loose sense, Conan might be described as a cross between Tarzan and John Carter. Howard’s narrative style, unadorned and with a naturalistic focus on action, follows Haggard and Burroughs. Finally, many of these forerunners published in popular fiction periodicals: the protopulps Argosy and All-Story, catchalls for most types of popular genre fiction, regularly ran Burroughs and Merritt tales before their book publication. Many of the followers of Haggard had new work, or reprints of old work, appear in both Weird Tales and Unknown. A little more might be said of the more conventionally fantasy elements— wizards, dragons, and so on—that do pervade the work of Howard and his followers to a far greater degree than that of his forerunners. While the demands of Weird Tales may partly account for their presence, some further account of the form they take is in order. The fantastic beings of Burroughs’s Barsoom series perhaps served as something of a model for Howard. However, where Burroughs tended to invention here, Howard was more apt to draw on actual legendary traditions. Of course, Howard’s Hyborian Age was terrestrial, Burroughs’s Barsoom extraterrestrial. The often generalized linking of Howard’s heroes with the legendary warriors of antiquity—Beowulf, Achilles, Cuchulainn—suggests another

176

THE EVOLUTION OF MODERN FANTASY

strand feeding into the development of S&S, seemingly in common with early twentieth-century literary fantasy: heroic romance and epic. More has been made of this ancestry, perhaps, than of the nonfantastic nineteenthcentury historical romance, though this is in part owing to the tendency to view pregenre fantasy through a postgenre lens. There is, undeniably, a not insignificant debt to traditional heroic narrative here; the idea of an archetypal dragon-slayer warrior, for instance, Howard explicitly ties to the protagonist of “The Valley of the Worm”: “You have heard the tale before in many guises wherein the hero was named Tyr, or Perseus, or Beowulf, or Saint George. But it was Niord who met the loathly demoniac thing” (3Howard 146). The nature of the debt to traditional heroic narrative needs some qualification, however. Howard’s work (and S&S generally) operates at a far greater remove from the traditional material than does the work of Eddison or William Morris. Yes, Conan and Bran Mak Morn are powerful warriors like their forebears, and they engage in adventures superficially resembling those of their forebears. But whereas the textures of the older narratives were intimately woven into The Worm Ouroboros or Child Christopher, Howard and his followers more often simply transposed elements of surface content to modern adventure fiction. What Howard drew from these earlier narratives could as well have come from Bulfinch’s or Guerber’s as from Homer’s or Snorri Sturlusson’s works themselves. The modern conception of “barbarian,” so crucial to the depiction of Conan, is strikingly absent not only in heroic epics from Gilgamesh to Ramayana to Tain bo Cualgne but in the works of Tolkien, Eddison, and Morris. Before turning to Howard’s followers, one last note about Howard’s preoccupation with history is appropriate here. There are some easy-to-see similarities between Howard’s Hyborian Age and the kind of historicizing often evident in the literary fantasy discussed, developed most extensively by Tolkien. Certainly Howard’s prehistoric ages bear some likeness to Tolkien’s prehistoric Three Ages of Middle-earth. However, Howard cast his summary history in the form of an objective (modern) historical essay, chiefly concerned with setting down straightforward “facts” to be adhered to in the stories. Tolkien is more complex: the “history” of Middle-earth is not presented to the modern reader as “background facts,” or as a modern historical essay, but in the quasi-traditional forms of chronicle, annal, and so on. “History” is narrative—is “story”—and cannot be separated from the form and language in which and the perspective from which it is told: LOTR is the “history” of the end of the Third Age from the perspective of Bilbo, Frodo, and Sam, as the chronicles and annals are “histories” from the perspective of specific elves or humans. Howard’s “Hyborian Age” essay is not an “artifact” from the Hyborian Age; Tolkien’s “Tale of the Years,”

TWENTIETH CENTURY

177

“Annals of Aman,” and “Quenta Silmarillion” are presented as artifacts from the Three Ages. Howard assumes something like the modern sciencebased shape of time and evolution in his work; Tolkien assumes a nonscientific mythological cosmos in his work. In short, while Howard’s S&S may seem closer to the literary work than the proto–Dark Fantasy of Lovecraft and Smith, it is in conceptually quite distinct territory. Unlike the fantasy of Lovecraft and Smith, Howard’s tales bred followers, and a fair amount of work modeled on the Conan stories appeared in Weird Tales. Probably the most significant, and original, follower was Catherine L. Moore, who introduced her female warrior, Jirel of Joiry, with “Black God’s Kiss” (1934), followed by four more tales over the next five years. Henry Kuttner and Clifford Ball both published stories in Howard’s vein. Following more in the Haggard mode, H. Warner Munn’s King of the World’s Edge (1939) follows Myrdhinn (Merlin), Gwalchmai (Gawain), and the Roman centurion Ventidius Varo from the ruin of Arthur’s Britain to the “Lands of the West” (North America) and the Aztec empire. While the work of Howard and his immediate followers was thinly represented in the BAFS, Carter did not ignore it in his commentaries, and its absence, again, would seem to reflect that most of it was available from other paperback publishers. In his prefatory notes to both “The Valley of the Worm” (in the 1969 The Young Magicians anthology) and “The Garden of Fear” (in the 1971 New Worlds for Old anthology)—the only Howard stories to appear in the BAFS, and each a “stand alone” tale—Carter noted that Howard’s other work was already “well known and available” (Carter 145). Carter also included “Jirel Meets Magic” in New Worlds for Old, but the full Jirel sequence was already available from Paperback Library (1969). King of the World’s Edge was reprinted by Lancer Books in 1966; the still-living Munn followed it with two new sequels: The Ship from Atlantis (1967), also published by Lancer, and the voluminous Merlin’s Ring (1974), one of the last releases of the BAFS. With the end of the Wright period, Weird Tales turned away from the types of fantasy discussed here—the Ray Bradbury of October Country was probably the most significant author of the 1940s. Munn is the only author up to this point to produce fantasy during the period of the genre’s distillation: Howard died in 1936, Lovecraft in 1937, Kuttner in 1958; Smith had largely ceased writing by the late 1930s, Moore by the late 1950s. Fritz Leiber’s early tales of Fafhrd and the Gray Mouser were published by Unknown rather than Weird Tales, but they follow in the S&S pattern established by Howard.20 According to Leiber, the two characters were suggested in 1934 by friend Otto Fischer, and the series was to a considerable degree a collaborative

178

THE EVOLUTION OF MODERN FANTASY

effort. While standing clearly in Howard’s line of descent, with focus on the swashbuckling adventures of the protagonists, direct influence of Howard is not strongly evident. The world of Leiber’s heroes, Nehwon, is a fully secondary world, which, with its complex apparatus of doors to other times and worlds (including ours), perhaps recalls Cabell and Eddison rather than Howard. The sophistication of Leiber’s controlled, varied, often heavily ironic (if not notably archaized) prose also to some degree recalls those writers rather than Howard. At the same time, the series is very much a product of the world of the pulps, and Leiber’s debt to Howard’s ancestry—he particularly notes Haggard and Talbot Mundy (Leiber 7)—is clear, though he demonstrates exceptional control and sophistication in processing that debt. “Two Sought Adventure” (later retitled “The Jewels in the Forest”) was published in Unknown in 1939 and followed by four more Fafhrd and Gray Mouser tales over the next four years. After the demise of Unknown, Leiber wrote only two new tales of his heroic duo until the groundswell of interest in Sword and Sorcery (Leiber is credited with coining the name) nearly two decades later. A Leiber collection published by Arkham House, Night’s Black Agents (1947), contained two stories of Fafhrd and the Gray Mouser, neither new. All the eight then-extant stories of the series were gathered in Gnome Press’s Two Sought Adventure (1957). But it was only with the publication of four new tales in Fantastic in 1959 and following, and the subsequent demand brought on by paperback editions of the stories in the 1960s, that Leiber returned in a focused way to the series, continuing to add to it through the 1980s: the better part of the Fafhrd and Gray Mouser material in fact dates to the genre period. While the series was untouched by the BAFS, one surmises that this was simply due to the fact that all the stories were available in Ace Books editions at the time; Lin Carter ranked the Fafhrd and Gray Mouser series highly. Unknown Unknown published from 1939 to 1943. In contrast to the Weird Tales emphasis on the “weird,” Unknown was strongly touched by a speculative bent more often associated with science fiction. Editor John W. Campbell is most remembered as the founder and longtime editor of the science fiction magazine Astounding, where, with notable success, he had insisted that the “science” in the material he accepted be credible, at least on a theoretical level. Not surprisingly, some of the more well-known science fiction writers of Astounding (Asimov, Heinlein, van Vogt), thoroughly imbued with Campbell’s editorial emphasis there, regularly contributed less

TWENTIETH CENTURY

179

scientifically rigorous work to Unknown; a fair amount of material published in Unknown many readers would consider science fiction. Though the focus of Unknown was more explicitly on the fantastic, Campbell nevertheless placed a strong emphasis on logic and rules in dealing with magic and the supernatural. The result was that magical elements were often lifted into relief, so that magic itself became a theme in a way it had generally not been either at Weird Tales or in the work of the literary writers. More often than not, stories from Unknown focused on the imposition of what might more aptly be termed an alternate dimension than an alternate world on the lives of “normal” modern people. In H. L. Gold’s “Trouble with Water,” the protagonist, while fishing in Long Island, offends a water gnome who causes a drought; a mermaid participates in a New York swim race in L. Sprague de Camp’s “Nothing in the Rules.” The tension resulting from the juxtaposition of the “everyday” with the “improbable” is at the core of these and countless other stories from Unknown. There is a certain structural similarity between these stories and stories of the horror-based fantastic, but whereas the latter tended to see in the fantastic a crisis-fomenting breakdown of a fragile consensus reality (a common feature in Lovecraft’s most characteristic work), the stories in Unknown turned the tables and transformed the fantastic into a problem subject to logical scrutiny, solvable with the use of rationality and commonsense. Most frequently, the result tended toward humor rather than horror, and the recourse to speculative logic in resolving the tension links these stories more to science fiction than the “weird.” Of course, this places much of the content of Unknown outside the core focus of this study: Unknown was a “fantasy” magazine, but “fantasy” did not especially mean tales of quest and war in invented settings where magic works. The core Unknown template was not the BAFS template. Some of the Unknown work most amenable to the BAFS template was in fact a bit un-Unknown-ish. This includes the Fafhrd and Gray Mouser stories. Hannes Bok’s The Sorcerer’s Ship (1942), included in the BAFS, likewise lacks the usual speculative bent: set in a secondary world stumbled into by Gene, a primary world character, Bok’s work is reminiscent of A. Merritt’s The Ship of Ishtar.21 But Unknown did publish work that evinced its more characteristic speculative edge but that was simultaneously amenable to the subsequent BAFS template. For example, A. E. van Vogt’s The Book of Ptath (1943, expanded for book publication 1947), set on the far future continent of Gonwonlane, is constructed around a fundamentally action fantasy vocabulary, with its core focus on the conflict between an incarnation of the god Ptath and the goddess Ineznia. The magical elements, however, are posited as a sort of

180

THE EVOLUTION OF MODERN FANTASY

psychic offshoot of science at a point in history when lack of crucial metals has rendered advanced technology, as we know it, impossible. Probably the most significant of such work to appear in Unknown came in a series of collaborations by Fletcher Pratt and L. Sprague de Camp. Chief of these were the initial three adventures of Harold Shea: “The Roaring Trumpet,” “The Mathematics of Magic” (both 1940 and combined to form The Incomplete Enchanter in 1941), and “The Castle of Iron” (1941, expanded for book publication 1950). In these, Harold Shea, a New York psychologist, travels, respectively, to the worlds of Norse myth, Edmund Spenser’s The Faerie Queene, and Ariosto’s Orlando Furioso. These unequivocally fantasy domains, however, are rationalized on the scientifically based supposition that all imagined worlds are in fact glimpses of parallel universes. Shea travels to these worlds by means of a technological invention, the “syllogismobile.” Despite the tongue-in-cheek (the Shea stories are unambiguously comic fantasies), the speculative dimension of the stories links them more to H. G. Wells (the “syllogismobile” certainly carries an echo of the Time Machine) than to Howard or Smith. Finally, and this emerges rather naturally with Unknown’s combined tendency to modern protagonists and humor, the Harold Shea stories were the first major fantasy work to deliberately, for comic effect, juxtapose colloquial language—especially in the speech of the characters—with the settings of heroic fantasy. Pratt and de Camp wrote two other collaborative novels during the Unknown years. Land of Unreason (1941) uses less tongue-in-cheek rationalization in its setup and is darker than the Harold Shea stories. Its journey into the realm of Oberon is less founded in fast-paced adventure but no less amenable to the BAFS template.22 The Carnelian Cube (1948) was intended for Unknown, but with the magazine’s demise, it did not appear until Gnome Press’s edition several years later. This title moves significantly away from the BAFS template. After Unknown to the Beginnings of the Genre Notably less successful commercially than its sibling Astounding, Unknown fell victim to the World War Two paper shortage in 1943, and the period between this and the emergence of Sword and Sorcery in the early 1960s is checkered. For the remainder of the 1940s, there was no successor to Unknown that lasted, which meant that there was no sustained, identifiable periodical market for fantasy. That the next major new magazine to emerge, The Magazine of Fantasy, became The Magazine of Fantasy and Science Fiction in its second issue in late 1949 serves to underscore the

TWENTIETH CENTURY

181

disproportionate popularity of science fiction compared to fantasy, as does the fact that science fiction greatly outweighed fantasy in its contents. It is also important to note that fantasy had different connotations at this point than it came to have with the BAFS: very little of what fantasy did appear was in the vein of the BAFS template. Given the postwar situation, some of the core Unknown fantasy writers (notably de Camp and Leiber) turned more exclusively to science fiction. Nevertheless, some significant work did squeeze through the cracks during the 1940s and 1950s. Most of this embodied a distinctly Unknown rather than Weird Tales sensibility. The Pratt/de Camp team produced two more Harold Shea tales: “Wall of Serpents” (1953), which takes Shea to the world of the Finnish Kalevala, appeared in Fantasy Fiction; “The Green Magician” (1954), which takes Shea to the Ireland of Tain bo Cualgne, appeared in Beyond. The two were combined in hardcover as Wall of Serpents (1960). However, as the foregoing has implicitly suggested, book editions of work by the popular authors—unheard of during the Lovecraft/Smith/ Howard years of Weird Tales—became increasingly common during and after the Unknown years. There were variations in this process. Some of the work whose initial magazine publication in Unknown had elicited an enthusiastic reception was published soon after in book form by Holt.23 Some small circulation, hardcover, amateur presses devoted to the fantastic began operation during the Unknown period and after. The most famous and long lasting (still in operation) of these was August Derleth’s Arkham House, founded in 1939 specifically to publish Lovecraft’s work in book form. The first book publications of Lovecraft’s “Dreamlands” tales (1943) and of the tales of Clark Ashton Smith (1942 and following) were by Arkham House. Less long lived, the Gnome Press, founded in 1948, published the first book editions of Howard’s Conan stories (1950 and following), albeit with additional material and posthumous “collaborations” by de Camp and Bjorn Nyberg intended to flesh out a full “biography” of the hero. The first Fafhrd and Gray Mouser collection and the Pratt/de Camp Carnelian Cube (as well as the expanded Castle of Iron) were published by Gnome. In addition, some fantasy titles would occasionally be released by more generic popular fiction publishers. Fletcher Pratt’s two more ambitious, post-Unknown solo fantasy novels appeared in this last context. The Well of the Unicorn (1948) was published by William Sloane Associates, under the pseudonym George U. Fletcher. Certainly the most ambitious single work, in both scale and complexity, to be produced by any of the pregenre popular fantasy writers, it is closer to the work of William Morris, Eddison, and Tolkien than anything otherwise produced by a writer to emerge from the world of Weird Tales and Unknown.

182

THE EVOLUTION OF MODERN FANTASY

It was still more than half a decade until the publication of LOTR, but the influences of both Morris and Eddison lurk on various levels behind The Well of the Unicorn. Some of this is finite and piecemeal: the eponymous “Well,” which brings peace to its drinkers, conceptually echoes Morris’s “Well at the World’s End,” though, insofar as the protagonist Airar declines to drink of it, the significance is inverted; the mysterious enchanter Meliboe seems almost directly modeled on Eddison’s Vandermast of the Zimiamvia books. On a more impressionistic level, Pratt’s modestly conceived hero Airar and his wandering but practically and precisely chronicled adventures, in which magic plays only a subdued role, recall many of the protagonists of Morris’s romances, while the meticulously evoked world of political intrigue he finds himself in recalls Eddison’s Zimiamvia.24 But the influences of Morris and Eddison were limited; “inspiration” was probably more of a factor than close “influence” in both cases. Pratt was also intimately familiar with some of the same traditional wellsprings that influenced Morris and Eddison: de Camp notes that Pratt “read Norse sagas in the original” (de Camp 178); the excerpts from The Poetic Edda in the text of “The Roaring Trumpet” were translated by Pratt. Some of the techniques of the saga writers are apparent in the “precise, dramatic and factual story,” as Well was called in the “Publisher’s Forward” (Pratt ix). There are distinct differences between the work of Morris and Eddison, on the one hand, and of Pratt, on the other. While the prose of The Well of the Unicorn is formal, if not elevated or archaized (there are none of the Harold Shea colloquialisms here), embodying a certain rhythm and some syntactic idiosyncrasies, it is not “poetic” in the romantic sense of Morris or Eddison. While Airar is modestly conceived as Morris’s Hallblithe and Osberne are, he is depicted in much more strongly novelistic terms, not in broad romance-style brushstrokes. And while Pratt’s narrative is as wrapped up in intrigue as Eddison’s are, it is an intrigue shorn of its aristocratic elevation and Elizabethan nuances. Pratt’s depiction of magic, while muted, nevertheless betrays an Unknown-ish sensibility in his attention to its practical ramifications in the lives of his characters—something neither Morris nor Eddison are particularly interested in. Finally, Pratt’s own interests as an historian,25 which included an expert knowledge of medieval Danish history (this informs Airar’s world), are evident in the meticulously developed sociopolitical fabric of The Well of the Unicorn. Pratt’s second fantasy, The Blue Star, appeared with two supernatural novels26 in an omnibus, Witches Three (1952), published by Twayne Publishers. The Unknown-ish sensibility is more pronounced here. This is partly due to a rather mechanical frame, faintly reminiscent of Wells’s Time Machine, in which a group of armchair intellectuals speculate about the possibility of other worlds and posit a matriarchal world in which magic

TWENTIETH CENTURY

183

is real—then collectively dream the main narrative of the book. The magical, or “fantasy,” element of this world is the eponymous “Blue Star,” which binds the male that holds it under the power of its witch but also gives him the ability to read minds. The Unknown-ish sensibility is evident in the way Pratt develops this finite magical element in terms of the specific practical impact it has on individuals in a world in which, once again, the sociopolitical fabric is meticulously developed. The setting is not a variation on the ancient- or medieval-style societies that typified most of the core BAFS titles but, with its high society balls and classical orchestras, a world whose prototype is, de Camp suggests, “the Austrian or Holy Roman Empire of the eighteenth century” (de Camp 190). The Blue Star is distinctly unheroic and, indeed, more akin to a historical novel with a magical theme than it is to most of the other work discussed in this chapter. Neither of Pratt’s solo fantasies sold well. In addition to the fringe status of such fantasy at the time, the insistence of William Sloane Associates that Pratt use a pseudonym meant that Well did not benefit commercially from the fact that its author was well-known both for his genre fiction and for his historical writing. The Blue Star probably did not benefit from being the third part of a large omnibus. Of course, both books are demanding and considerably less commercial in nature than Pratt’s collaborations with de Camp. Nevertheless, both books did have their admirers and reappeared with the emergence of the genre in the late 1960s. Well was reprinted by Lancer Books in 1968 (and therefore not included in the BAFS); The Blue Star was the first release in the BAFS proper in 1969. Though de Camp had turned primarily to science fiction during the 1940s, his belated discovery of Howard’s Conan stories through Gnome Press’s Conan the Conqueror (1950), given to him by Pratt, who disliked it, spurred him back into the action fantasy direction. A large part of the result was the Conan spin-off work he and Boris Nyberg did for subsequent Gnome Press Conan volumes, something he would return to in collaboration with Lin Carter in the 1960s. Other work by de Camp evincing Howard’s influence includes The Tritonian Ring (1953), released in hardcover by Twayne Publishers, followed by some shorter tales set in the same milieu (“Pusad”) over the next few years, as well as the Novaria sequence, beginning with The Goblin Tower (1968). Two new writers began careers with fantasies in the postwar, pre-1960s period, but like Leiber and de Camp, they turned predominantly to science fiction after these initial publications. Poul Anderson’s The Broken Sword (1954), like Pratt’s Well, was published by a popular genre house, AbelardSchuman. Less so than Well, Sword is nevertheless closer to the literary work than the other popular fantasy discussed in this chapter, though this does not seem to be the result of direct influence. Set in the Scandinavia,

184

THE EVOLUTION OF MODERN FANTASY

Great Britain, and Ireland of the Viking Age, Anderson evokes his Northern atmosphere partly by direct recourse to saga narrative conventions, echoed immediately in the direct, factual account of lineage in the opening paragraph. The subsequent story’s focus on dramatic action, and Anderson’s weaving of strands of Fate as an overbinding element of the plot, creating a sense of tragic inevitability, also echoes saga conventions. Further nuances, including Anderson’s use of traditional skaldic poetic form in the spoken verses, also underscore this effect. While Anderson does not stint on description, and his prose is less spare than that of the sagas, the language is rooted in Anglo-Teutonic forms, and its mildly archaistic, rhythmic force evokes Northern tradition. Yet there are elements of Sword quite distinct from the sagas, which betray something of an Unknown-ish sensibility. For example, the sagas rarely move overtly into the Faerie realm, and even a “mythic” saga, such as Volsunga Saga, remains emphatically centered in the human sphere. Magical creatures are rarely treated familiarly. In contrast, comparatively little of Sword takes place in the human sphere, and some of its magical characters are treated familiarly, even intimately. The Faerie realm is treated in an implicitly speculative fashion, like a parallel dimension in the same geographic space as the human world but invisible to humans. The structure of the Faerie world is also, Unknown-style, neatly diagrammatic. Aesir and Tuatha de Danaan, Elves and Sidhe folk, coexist in the cosmos of Sword, rooted in, respectively, Germanic/Scandinavian and Gaelic dominated areas—something alien to either traditional Germanic or Celtic sources.27 The Germanic end of Faerie breaks clearly down into Elf, Dwarf, Troll, and Goblin, each with their respective societies. In his depiction of the specifically magical qualities of his Faerie peoples, Anderson himself, writing in 1971, acknowledged that he had “amused himself with a bit of rationalization in the grand old Unknown Worlds manner” (Anderson xiv). Anderson wrote one other fantasy in the 1950s, prior to turning almost exclusively to science fiction until the 1970s. An early, short version of Three Hearts and Three Lions (1953) appeared in The Magazine of Fantasy and Science Fiction, which was expanded for book publication in 1961. More unambiguously in the Unknown mode, Three Hearts protagonist Holger Carlsen, like Harold Shea, provides a twentieth-century viewpoint, facilitating a speculative edge to the story; in transporting Carlsen to an actual traditional locale (the world of medieval Carolingian romance), Anderson echoed the Shea stories. Three Hearts and Three Lions was consistently available from the early 1960s through the BAFS period in non-Ballantine editions and hence did not appear in the series. The Broken Sword appeared in the series in 1971, though Anderson took the opportunity of a new edition to, outside

TWENTIETH CENTURY

185

of leaving the plot largely intact, heavily revise—virtually rewrite—his original text. The BAFS also published Anderson’s new Hrolf Kraki’s Saga (1973), derived from the Icelandic King Hrolf ’s Saga and other sources. Surprisingly little of the fantasy from this period appeared in the rapidly expanding pocket paperback medium, and Jack Vance’s The Dying Earth (1950), published by Hillman Books, may be alone in this regard. The Dying Earth is in fact a collection of short stories, unified not by any intrinsic dramatic structure (though common characters do occasionally crisscross between stories) but by a common setting. The world of the “Dying Earth” echoes Clark Ashton Smith’s Zothique insofar as it is a far future milieu in which the aging sun has turned red and sorcery has been reborn. Apart from Smith, Vance is the only popular writer (possibly excepting Leiber) whose prose carries the deliberately crafted artificiality of the early twentieth-century literary writers, though his style is quite distinct from Smith’s. While Smith’s prose and narrative textures are dark and ponderous, strongly rooted in Beckford and Poe, Vance is lively and humorous: the influence of Cabell is evident in the droll, ironic detachment with which he depicts his characters; the frequent calculated indirection at times recalls Bramah’s Kai Lung tales. With their frequent speculative edge (though this is invariably tongue-in-cheek), the Dying Earth stories would not have been out of place in Unknown. Like The Well of the Unicorn, The Blue Star, and The Broken Sword, The Dying Earth made little impression when it was originally published. During the 1940s, Vance had tried to sell some of the stories singly to the science fiction magazine Thrilling Wonder Stories, whose editor, Sam Merwin, considered them “fascinating but, alas, unpublishable pseudo-Cabellian fantasies” (Carter 151). Hillman Books itself was a short-lived publisher, with virtually no distribution beyond the West Coast—both of which, in addition to the limited commercial status of fantasy at the time, conspired against The Dying Earth, which, until the Lancer Books edition (1961), became an inaccessible legend. Like Anderson, Vance turned almost exclusively to science fiction after his first work, returning to the “Dying Earth” milieu only in the wake of the Sword and Sorcery revival with Eyes of the Overworld (1966) and, after the BAFS period, with Cugel’s Saga (1983) and Rhialto the Marvelous (1985). Here we reach the transition point at which fantasy began to coalesce into the discrete genre we now recognize. In commercial terms, that genre grew out of the publication venues that were the media for the twentiethcentury popular writers of the BAFS canon. Between the wars, pulp periodicals were the major outlets for genre fiction, with Weird Tales and Unknown of greatest significance here. After World War Two, the mass market pocket paperback industry emerged, complementing, then vying

186

THE EVOLUTION OF MODERN FANTASY

with, and then, by the early 1960s, supplanting the pulp periodicals as the major medium for genre fiction. Throughout this latter period, there was little commercial demand for fantasy of any kind, and certainly not for the kind of work that came to typify the BAFS; even during the Weird Tales/ Unknown period, there was not a sufficiently coordinated popularity to draw together the different strands on the basis of their common elements and create a discrete genre. There was no commercial fantasy genre beyond what might be appended to science fiction or horror, or simply generic popular fiction—nothing resembling what was staked by Ballantine in the late 1960s. When popular fantasy began its more focused renaissance in the early 1960s, it was as a subcategory of science fiction, by publishers (Ace, Pyramid, Lancer, and later Ballantine) whose mainstay was science fiction. The connecting thread is the writers themselves: as noted before, Leiber and de Camp, after their beginnings at Unknown, and Anderson and Vance, after their initial fantasies, turned primarily to science fiction, becoming staples at the new paperback firms (Ace, Ballantine, etc.) with science fiction lines. When, with the renewed and rapidly increasing interest in pulp fantasy in the early 1960s, their older work was reprinted, it was reprinted by these same publishers. These four writers in particular represent a bridge between the pregenre and genre periods: they all published fantasy during the pregenre period, then turned predominantly to science fiction, then returned to fantasy (albeit with varying frequency) after 1960—two continuing series begun long before.28 That these writers took so naturally to the new genre itself suggests an organic connection between it and their pre-1960s publishing venues. But again, it is important to reiterate that “Sword and Sorcery,” “Heroic Fantasy,” and Ballantine’s “Adult Fantasy” were creations of the 1960s, and even the popular strand that provided the impulse for the initial steps of the genre’s creation in the half decade before the Tolkien boom of 1965–66 was itself compounded of diverse threads, though it undeniably had more substantial “glue” scattered among its parts than the literary strand. Conclusion The work discussed in this chapter and Chapter 5, then, constitutes the pregenre twentieth-century fantasy canon enshrined by the BAFS. That both the popular and literary strands, and particularly the latter, were internally marked not by generic uniformity but by variety, the preceding has made clear. But a few reflections on the relation and ultimate interpenetration of the two strands, and their relation to the BAFS template articulated by Lin Carter, are in order.

TWENTIETH CENTURY

187

The genre itself emerged directly out of the popular strand. The literary strand was, in effect, spliced in: as discussed in the introduction, Don Wollheim of Ace Books turned to LOTR only when the supply of S&S from his usual (popular) sources became thin. Tolkien rapidly became a far bigger seller than the rest of the S&S subgenre combined, and while his popularity no doubt served to expand the reading base of the popular work, not even Conan transcended the bounds of genre readership so fully. So ultimately, Ballantine, having acquired publishing rights to Tolkien’s work, began tracking down other maverick literary work “like Tolkien” in a focused way, which process was accelerated by the arrival of Lin Carter and beginning of the BAFS in 1969. Thus work that had emerged in a “highbrow” context, alien to the pulps and pocket paperbacks of popular American fiction publishing, was suddenly appearing side by side with work by Howard, Smith, Leiber, de Camp, Vance, and so on, which had been written for that market. That the connection between the two strands was limited has been amply developed here, but I will close this chapter by noting one genre period disagreement that implicitly reinforces my point, concerning language. In her essay “From Elfland to Poughkeepsie” (1973), Ursula Le Guin insists that a certain formal decorum is appropriate to the language of fantasy.29 She lambasts the journalistic language of Katherine Kurtz’s Deryni Rising (1970) and criticizes Fafhrd and the Gray Mouser’s behind-the-scenes colloquializing as the characters “peeking out of the story to say ‘We’re just plain folks really’ to the reader,” though she grants Leiber considerable skill as a writer. She cites Eddison, Kenneth Morris, and Tolkien as the great prose masters of twentieth-century fantasy, with Dunsany and Cabell mentioned for high marks (Le Guin 90–99). However, in an introduction to an edition of the Harold Shea stories, David Drake asserts that “bad writers since William Morris have insisted that their characters speak pseudo-Elizabethan” and that the use of colloquial language emphasizes to the reader that the characters are “real people making real decisions” (Drake ix–x). It is interesting that what is distracting to Le Guin is a virtue to Drake. What is clear is that Le Guin would likely prefer the writers discussed in Chapter 5, while Drake would prefer those discussed in this chapter. Works Cited Anderson, Poul. The Broken Sword. New York: Ballantine Books, 1971. Carter, Lin. Imaginary Worlds. New York: Ballantine Books, 1973. Connors, Scott, and Ron Hilger. “Appendix One: Story Notes,” in Smith, Clark Ashton. The End of the Story: Volume One of the Collected Fantasies of Clark Ashton Smith. Edited by Connors and Hilger. San Francisco: Night Shade Books, 2006.

188

THE EVOLUTION OF MODERN FANTASY

de Camp, L. Sprague. Literary Swordsmen and Sorcerers: The Makers of Heroic Fantasy. Sauk City, WI: Arkham House, 1976. Drake, Dave. Preface, in de Camp, L. Sprague, and Fletcher Pratt. The Complete Compleat Enchanter. New York: Baen Books, 1989. 1Howard, Robert E. The Bloody Crown of Conan. New York: Ballantine Books, 2003. 2Howard, Robert E. The Coming of Conan the Cimmerian. New York: Ballantine Books, 2003. 3Howard, Robert E. “The Valley of the Worm,” in Carter, Lin, ed. The Young Magicians. New York: Ballantine Books, 1969. Le Guin, Ursula K. “From Elfland to Poughkeepsie,” in The Language of the Night: Essays on Fantasy and Science Fiction. Edited by Susan Wood. New York: Berkley Books, 1985. Leiber, Fritz. Ill Met in Lankhmar. Clarkston, GA: White Wolf Pub., 1995. Pratt, Fletcher. The Well of the Unicorn. New York: William Sloane Associates, 1948. Sidney-Fryer, Donald. Introduction, in Smith, Clark Ashton. The Last Incantation. Edited by Sidney-Fryer. New York: Pocket Books, 1982.

Other Secondary Work Consulted Both the Carter and de Camp volumes in Works Cited provide detailed background concerning the popular fantasy discussed in this chapter. The editorial apparatus in the Ballantine Howard volumes listed in Works Cited and in Primary Source Editions, and in the Night Shade Books edition of Smith’s work, is invaluable.

Some Primary Source Editions Lovecraft’s relevant work has recently appeared in the following: Lovecraft, H. P. The Dreams in the Witch House and Other Weird Stories. Edited by S. T. Joshi. New York: Penguin Books, 2004. Smith’s complete fantasy has appeared in the five-volume Collected Fantasies of Clark Ashton Smith. The first volume is listed in Works Cited under Connors; the others are The Door to Saturn (2007), A Vintage from Atlantis (2007), The Maze of the Enchanter (2009), and The Last Hieroglyph (2010). There is also a good, less exhaustive selection: Smith, Clark Ashton. The Dark Eidolon and Other Fantasies. Edited by S. T. Joshi. New York: Penguin Books, 2014. All Howard’s Conan stories, and those of King Kull and Bran Mak Morn, were recently collected by Ballantine Books. The first two of the Conan volumes are listed in Works Cited; the third is The Conquering Sword of Conan (2005). The other cycles are collected in Bran Mak Morn: The Last King (2005) and Kull: Exile of Atlantis (2006).

TWENTIETH CENTURY

189

Fritz Leiber’s Fafhrd and Gray Mouser stories have been collected by White Wolf Press. The first is listed in Works Cited; the others are Lean Times in Lankhmar (1996), Return to Lankhmar (1997), and Farewell to Lankhmar (1998). The complete Harold Shea stories of de Camp/Pratt are included in The Complete Compleat Enchanter, listed in Works Cited. Jack Vance’s Dying Earth stories are collected in the following: Vance, Jack. Tales of the Dying Earth. New York: Orb Books, 2000. The following give a substantial cross section of work from Weird Tales and Unknown: Kaye, Marvin, ed. Weird Tales. Garden City, NY: Doubleday, 1988. Schmidt, Stanley, and Martin H. Greenberg, eds. Unknown Worlds: Tales from Beyond. New York: Galahad Books, 1988.

7

Conclusion 1960–80

The “First” Fantasy Writers

I

will conclude with a brief overview of the new work that appeared during the period of the genre’s construction, from the Sword and Sorcery vogue of the early 1960s through about 1980, the point at which the bestseller genre that is still with us had fully consolidated itself. In a certain literal sense, it would be somewhere among these writers that we would find the “first” fantasy writers: writers who conceived of their work as fantasy in something closely approximating the terms of the BAFS template. I have chosen not to discuss work from the “bestseller” period for a variety of reasons. In crucial respects, the material from this period embodies some profound differences from the earlier work; often, likenesses are cosmetic. I will suggest some of these in the following pages. But more than the likenesses and differences, the sheer volume of work, coupled with the fact that the publishing phenomenon is still ongoing, would render any attempt at an overview a dubious endeavor at best. In Chapter 1, I outlined the stages of the genre’s development through the 1960–80 period as follows: (1) the nascent period of the Sword and Sorcery and Tolkien booms (1960s); (2) the BAFS period (1969–74); and (3) the post-BAFS transition to the bestseller genre (1975–80). The first two stages were as or more characterized by reprints of the material that this study has focused on (the “BAFS canon,” as I have called it) as by new material, though there was certainly new material. The third stage is marked by a decided shift in the direction of new work and the consolidation of a formulaic approach very much geared to “bestsellerdom.” I will, appropriately, break the following discussion into two sections: (1) the 1960–74 period, where the new material existed very much

192

THE EVOLUTION OF MODERN FANTASY

in relation to the pre-1960 canon, which flanked it readily on bookstore shelves, and (2) the 1975–80 period, when the works of the pre-1960 canon became gradually less accessible and new material came to compose a more self-contained phenomenon. 1960–74 As I discussed at the end of Chapter 6, the Sword and Sorcery vogue of the early sixties was facilitated by publishing venues that were a direct outgrowth of the pulps, where much of the popular fantasy of the BAFS canon had originally appeared. A fair amount of the new material of the 1960s was in fact written by surviving Weird Tales and Unknown writers who were still active. As noted, H. Warner Munn and Fritz Leiber continued series that had begun in those periodicals. Vance’s The Dying Earth, after being rejected piecemeal by a science fiction pulp, had been published by a short-lived science fiction publisher. The Eyes of the Overworld (1966) was not a sequel, per se, but followed the setting and style of the first book. L. Sprague de Camp, now in collaboration with Lin Carter, continued with Conan spin-offs; in original work, such as The Goblin Tower (1968), the tone and substance were very much in keeping with his pre-1960s work. Among the newer writers here, the most significant is Michael Moorcock, whose Sword and Sorcery tales of Elric of Melniboe began appearing in Science Fantasy magazine in the early 1960s.1 Though this work is originally and recognizably Moorcock’s, it is very much of a piece with the work of the older writers, with Leiber and Poul Anderson’s The Broken Sword seeming to have exerted particular influence.2 The continuity between this and the corresponding pre-1960 work is, for obvious reasons, particularly strong, and the focus remained fairly fixed throughout the 1960–75 period. There is little sign that the work of Tolkien affected this material at all.3 By contrast, literary fantasy, in the sense connoted by the material discussed in Chapter 5, was sparse at the time LOTR was published and had all but evaporated by the 1960–75 period. All the earlier major authors were dead, except Tolkien, who was in his seventies and puttering around with the massive labyrinth of Silmarillion drafts, which he was never able to give a final form. Only his brief, elegiac Smith of Wootton Major (1967) dates in composition to the post-1960 period.4 Peter S. Beagle’s The Last Unicorn (1968), published by Viking a year prior to the Ballantine edition, is one of the few significant fantasies from the period to have first appeared as a literary work. Though not sharing many of the common characteristics of the earlier literary work (formal archaized language, engagement of heroic

CONCLUSION

193

romance conventions), The Last Unicorn is certainly a uniquely idiosyncratic, maverick work in its own right. Equally idiosyncratic is the world of the rabbits explored by Richard Adams in Watership Down (1972), with its realism of setting and Tolkienian backdrop of myth and legend. Adams turned to the human sphere in Shardik (1974), set in the invented Beklan Empire, notable for its development of political context and for the extremely muted nature of its supernatural elements.5 While both Beagle’s and Adams’s works follow the literary canon in their maverick character, it is difficult to find unambiguous signs of direct influence beyond a vaguely impressionistic Tolkienian flavor. In general, discernible influence coming from the pregenre literary work was very much Tolkien dominant and primarily evident in work published for young readers, by authors who were now more often American and women. The Tolkien influence in this context stretches back to C. S. Lewis’s Narnia series during the earlier 1950s, though since Lewis belongs to Tolkien’s generation, I appended it to the discussion of Tolkien in Chapter 5. American writer Carol Kendall’s The Gammage Cup (1959) and its sequel, The Whisper of Glocken (1965), evince a fairly thick layer of Tolkien influence, specifically of The Hobbit, in their tone and setting and in their hobbit-ish Minnipins. The Tolkien influence, combined with that of T. H. White, is evident in American writer Lloyd Alexander’s Prydain Chronicles (1964–68).6 Here it is the shadow of LOTR rather than The Hobbit, however: a very Tolkienian good-versus-evil struggle underscores the series, and the culminating volume seems directly patterned on LOTR. Despite Alexander’s use of a Welsh-derived nomenclature, there is little Welsh in the thematic and narrative substance of Prydain: it is not really a “Celtic Fantasy.” Ursula Le Guin’s Earthsea trilogy (1968–72)7 is probably the most original of the work from the period standing in the stream of Tolkien influence. Rather than direct borrowings or heavy-handed echoes, the Tolkien presence is felt in more generalized elements: the narrative voice, while simpler (and not archaized), carries much of the tone associated with Tolkien’s work; the quasi-historical depth of the canvas, with its allusions to traditional myths and legends, echoes Tolkien. However, these elements are very much turned to Le Guin’s own purposes. Perhaps the most notable development has to do with magic: similar to much popular work going back to Unknown, there is an immediate consciousness of the practical ramifications of magic, but the further development of a metaphysical structure underscoring magic moves Earthsea into (then) quite original territory.8 Of the new work from the 1960–75 period, Le Guin’s trilogy has probably been the most widely influential.

194

THE EVOLUTION OF MODERN FANTASY

Other Tolkien-influenced work for younger readers from the period includes English writer Joy Chant’s Red Moon and Black Mountain (1970). Though flawed by the all-too-obvious shadow of Tolkien in some of its parts,9 Chant’s first work also includes much that is original and carries an eloquent narrative force. Her world of Khendiol is developed with a vivid immediacy and depth.10 Patricia McKillip and Jane Yolen, both American, should also be mentioned here: the former’s The Forgotten Beasts of Eld (1974) echoes Tolkien in its nuanced prose, though not in its narrative and dramatic substance; the latter’s The Magic Three of Solatia (1974) bears a similar relationship to Tolkien, and Le Guin’s Earthsea seems to lurk behind it to some degree as well.11 At somewhat of a tangent, something of E. Nesbit’s Five Children and It and its sequels emerged in Alan Garner’s The Weirdstone of Brisingamen (1960) and its sequel, The Moon of Gomrath (1963), which focus on the impingement of the magical on otherwise “normal” children in contemporary Cheshire. However, where the magical in Nesbit’s books was little elaborated beyond its direct relation to her children’s experience and depicted humorously, in Garner’s two books the magical embodies a quite earnest struggle between good and evil, developed in terms and characters clearly indebted to Tolkien. A like dynamic informs Susan Cooper’s later Dark Is Rising series,12 though the magical is given a more directly Welsh and Arthurian point of reference. In certain key respects, the work of Garner and Cooper, both English, can be seen to anticipate important aspects of Rowling’s Harry Potter series. As we reach 1974–75, then, the bulk of new fantasy either followed the action-based work rooted in pulp conventions or processed, to varying degrees, the influence of Tolkien in work directed to young readers. Some of this appeared in the BAFS: Munn’s Merlin’s Ring (1974) was one of the last titles published in the series; Red Moon and Black Mountain, despite its appearance in England as a title for young readers, was included in 1971.13 Particularly in the more Tolkienian work for children, we can see magic being lifted increasingly into relief as a defining theme;14 an increased consciousness of invented worlds is evident in more structured, detailed mytho-historical background and in the frequent presence of features such as maps.15 Moreso, here we find emerging a consolidated, and increasingly conscious, generic vocabulary—the kind of multigenre eclecticism marking the work of the popular canon, and the direct dialogue with traditional forms of romance, saga, and epic marking the work of the literary canon, become increasingly removed.

CONCLUSION

195

1975–80 The first factor in the transition to the second half of the 1970s was the termination of the BAFS, the hub of the reprints that had retroactively been assembled to create the new genre. I have discussed the ramifications of this in Chapter 1 and will not pursue the issue of reprints in detail here. Suffice to say, the end of the BAFS resulted in something of a void for fantasy publishing, and after a brief lull in 1975, the void began to be filled not by reprints16 but by new work by new writers. The shift to a focus on new work was a departure from what the BAFS had represented, though the new material the BAFS had published can be seen to signal some of what followed during the 1975–80 period.17 The more formal, literate direction embodied in Joy Chant’s Red Moon and Black Mountain—flanked by the non-BAFS work of Le Guin, McKillip, and Yolen—advanced during the decade’s second half, though less commonly appearing in the garb of work for young readers. Among the emerging writers in this vein were Tanith Lee, an English writer whose Birthgrave Trilogy18 and opening volumes of Tales of the Flat Earth19 are marked by a poetic, darkly lyrical style. The American Nancy Springer’s The Book of Suns (1977)20 and The White Hart (1979), the first two of five “Books of the Isle,” are also marked by poetic prose and a broad, Tolkienian canvas. This kind of material carried over into the bestseller period of the 1980s and following, attracting a strong readership, though it has only sporadically made its way onto bestseller lists. Among the more popular works published in the BAFS were the initial volumes of Katherine Kurtz’s Deryni series.21 Kurtz combined the more elaborate historicizing emerging in much of the work of the period (with chronological tables and, of course, maps) with a very practical, detailed magical schema, in a way not entirely alien to what Le Guin, Chant, and others were doing. On the other hand, the focus of the stories tends to settle on the fairly mundane world of political and ecclesiastical intrigue, depicted in a plainspoken, rather bald prose style. The sense of atmosphere so much a part of the work of Le Guin, Chant, McKillip, and Yolen is strikingly absent in Kurtz’s world of the Eleven Kingdoms:22 the Deryni books are along the lines of engaging political thrillers in which magic is involved. This combination of elements, as well as an openness sufficient to breed countless sequels and “prequels,” can perhaps mark the Deryni series as an effective prototype of many of the series to appear during the bestseller period. The issue of language perhaps serves as a barometer for the direction taken by Ballantine when, under the new editorship of Lester and

196

THE EVOLUTION OF MODERN FANTASY

Judy-Lynn del Rey, the publisher returned to fantasy in 1976. As noted in Chapter 1, further reprints of BAFS titles were limited. Those that were reprinted tended disproportionately to the popular rather than the literary end of the canon; “new” reprints were exclusively culled from the popular canon. As the previous chapters have discussed, the work from the popular canon tended away from the more involved forays into archaistic, poetic language, and it is perhaps significant here that the one popular writer to be omitted entirely from the Ballantine fantasy publications of the del Rey period is the one most preoccupied with language: Clark Ashton Smith. The new emphasis on more plain written work seems related to the new, more viscerally “realistic” look adopted by the del Reys, seen in the cover art of the Brothers Hildebrandt, Darrell Sweet, and others, quite distinct from the dreamy, surreal BAFS covers of Gervasio Gallardo, Bob Pepper, and others. The new work published by Ballantine followed this emphasis. Gordon R. Dickson’s The Dragon and the George appeared in late 1976, followed by Brian Daley’s The Doomfarers of Coramonde and Jaan Kangalaski’s The Seeking Sword in early 1977. All three were decidedly rooted in American popular fantasy conventions,23 written in unadorned, action-focused prose. But whereas most of the book-length material from the popular canon, and most of the new material that followed it during the 1960–75 period, had been of fairly humble dimensions, with 60–70,000 word volumes the norm, the new works by Dickson, Daley, and Kangalaski were well over 100,000 words. Essentially, Ballantine now favored a style and approach with roots in the popular canon but sought to cross that with the more expansive type of canvas associated with Tolkien.24 As I suggested in Chapter 1, the major turning point in fantasy’s final vault to bestseller-genre status occurred in 1977 and 1978, with the first publications of Terry Brooks and Stephen R. Donaldson. These combined some of the core new emphases of Ballantine—increased dimensions and the use of plainer, more action-focused prose—with a baseline blueprint formula derived clearly and specifically from LOTR. Brooks’s The Sword of Shannara appeared in late spring 1977 in a handsome trade edition, with cover and illustrations by the Brothers Hildebrandt25 and, of course, a map. The blurbs proclaimed it the book everyone had been waiting for since they finished LOTR. That many responded positively is amply indicated by the fact that Sword hung at number two on the New York Times trade paperback bestseller list for the entire summer and into the fall and that more than two dozen sequels/prequels have followed since. Donaldson’s first Chronicles of Thomas Covenant trilogy26 appeared in the late fall of 1977, in hardcover editions published by Doubleday; Ballantine’s mass market editions of the following summer, as hyped as The

CONCLUSION

197

Sword of Shannara had been, became bestsellers. A second trilogy appeared in the 1980s, as well as a further tetralogy in the new millennium. While both works generated large readerships and some critical enthusiasm, there were also dissenting voices. Lin Carter, who for a number of years after the folding of the BAFS edited The Year’s Best Fantasy anthologies for Daw Books, created a new category in the year’s summary section for 1977 called “The Worst Book Ever” specifically for The Sword of Shannara, writing, “This atrocity is the single most cold-blooded, complete ripoff of another book that I have ever read”: “Crom knows I have nothing against the fine art of pastiche, or writing in another author’s style, since I practice the craft myself. But Terry Brooks wasn’t trying to imitate Tolkien’s prose, just steal his story line and complete cast of characters, and did it with such clumsiness and so heavy handedly, that he virtually rubbed your nose in it. Everyone concerned with this war crime of a novel really ought to be ashamed of themselves” (2Carter 207–8). Regardless of whether one sees Carter’s ire as justified, it simply cannot be denied that The Sword of Shannara follows LOTR, episode by episode, character by character. If there’s Frodo, there’s Shea; if there’s Sam, there’s Flick; if there’s the Shire, there’s Shady Vale; if there’s Gandalf, there’s Allanon; if there are Black Riders, there are Skull Bearers; and so on. While there are piecemeal elements that are original to Sword, they exist within a canvas that is sourced directly and virtually exclusively from LOTR. Donaldson’s trilogy had more original elements in it, though it is not difficult to discern LOTR quite specifically in the basic narrative structure, in the “Dark Lord” (Lord Foul) who returns to power at the end of an age, and in the deployment of Tolkienian motifs such as the fellowship. (And of course, there are maps of “The Land.”) These bestselling works by Brooks and Donaldson stand as the direct prototypes of the later expansive “epic” fantasies by authors from David Eddings to Michael Jordan to Steve Erikson.27 But while the impetus behind this particular bestselling branch of the contemporary bestselling genre was clearly Tolkien and LOTR, it must be said that a book like The Sword of Shannara is a drastically different work than its model. A frequent counter to charges that Brooks’s work, Donaldson’s work, or many of the “epic” trilogies to come are overly derivative of Tolkien is that Tolkien borrowed, too.28 But what Tolkien “borrowed” from was millennia of quite varied material, which he knew in the original languages: he had virtually the entire body of traditional legendary and mythic European narrative in his head. His work was not modeled on that of any one modern writer, and the modern influences he would grant—William Morris and H. Rider Haggard—are fully appropriated to his own ends. They do not loom over his work in anything like the way LOTR looms over The Sword of Shannara

198

THE EVOLUTION OF MODERN FANTASY

and much of the work it and its progeny inspired. The importance of the traditional material to Tolkien was that it provided a vocabulary, which was then sifted through an enormously original imagination.29 By contrast, the “source” of Sword is a bit too singularly one quite recent modern work. Despite the surface similarities, Sword and LOTR are profoundly different books. Conclusion The basic trajectories that began to define themselves during the 1975–80 period, with the bestsellers of Brooks and Donaldson at the commercial core, are what drove the mushrooming of the fantasy genre during the early 1980s into the full-grown sibling of science fiction and horror that it now is; those trajectories are still easily discernable. And the phenomenon is not simply literary: with bestsellerdom, fantasy began to permeate mainstream popular culture through the media of games, films, and graphic novels. By itself, sheer bulk, as I have said, renders any “overview” of the post-1980 material necessarily provisional; it is also difficult to separate the strictly literary from the broader popular culture phenomenon. I suggested in the preceding pages some of the ways in which the work of the 1960s and 1970s, and particularly of the 1975–80 period, differed from the pregenre work. The bulk factor itself contributed to the escalation of these differences after 1980. As ever greater numbers of new fantasy works appeared, as ever greater numbers of new fantasy writers emerged, as the pregenre work canonized by the BAFS went out of print once more, and as Morris, Dunsany, and Eddison receded to effectively legendary status, the contemporary genre became increasingly contained and self-referential. Rather than pushing the horror story in a direction embodying elements that we would now recognize as fantasy elements (as Clark Ashton Smith did), the interface with horror now comes from the standpoint of an established genre and represents a conscious melding of elements belonging to two distinct genres. The similarities between the fantasy of the last few decades and the BAFS canon are often more cosmetic than otherwise. Nevertheless, the earlier work gathered together by Ballantine and others now more than forty years ago stands as a necessary prelude to the contemporary genre. My goal here has been to divest the earlier material from a too-ready identification with the terms of the current, established genre and to look at it in its own context and on its own terms. Simply moving from something like the BAFS template backward until one finds the “first” example (whether that is Tolkien or William Morris or Sara Coleridge or Eliza Haywood) creates a reductionist and ultimately inaccurate picture of

CONCLUSION

199

literary history. The evolution of fantasy is far more complex and engages areas of literature that have on the whole been neglected by studies to date. Works Cited 1Carter, Lin. Imaginary Worlds. New York: Ballantine Books, 1973. 2Carter, Lin, ed. The Year’s Best Fantasy Stories: 4. New York: Daw Books, 1978. Daley, Brian. The Doomfarers of Coramonde. New York: Ballantine Books, 1977. Flieger, Verlyn, ed. Commentary, in Tolkien, J. R. R. Smith of Wootton Major. London: Harper Collins, 2005. Kay, Guy Gavriel. The Summer Tree. New York: Berkley Books, 1986.

Notes

Introduction 1. Weird Tales, during the Farnsworth Wright editorial years (1926–39), and Unknown, later Unknown Worlds (1939–43), are the most noteworthy of these, though there were a host of others. This will be covered in some detail in Chapter 6. 2. Oddly, little of this work had appeared in the rapidly growing mass-market pocket paperback medium prior to the 1960s; Jack Vance’s The Dying Earth seems alone in this regard. 3. Though other publishers did release relevant titles, the three most noteworthy here are Lancer Books, Pyramid Books, and Ace Books. 4. Circa 1960, Fritz Leiber, L. Sprague de Camp, Jack Vance, and Poul Anderson were most well-known (and in demand) as science fiction writers, and while they did return to fantasy with the Sword and Sorcery revival, their output of science fiction, possibly excepting de Camp, remained far greater. 5. By contrast, releases by Ace, Lancer, and Pyramid—and, later, DAW and Pocket Books—were tagged “SF” even into the second half of the 1970s. 6. Carter was not, technically, the first to demarcate the terrain so: L. Sprague de Camp does so almost identically in the introductions to his two Pyramid Books anthologies, Swords and Sorcery (1963) and The Spell of Seven (1965), though he uses the labels “Sword and Sorcery” and the qualified “Heroic Fantasy.” But Carter’s accountings go into far greater detail than de Camp’s two-page introductions and were repeated over and over again through dozens of BAFS introductions and two books, supported by a publisher who was trying to do something “different.” And with Carter, the label became, simply, “fantasy.” 7. New titles did appear in the series, were welcomed when they were considered to be of quality in keeping with other series titles, and were—on the basis of frequency of reprintings—among the better-selling titles. But while Ballantine had cornered the market on reprinted material, most often by deceased authors, they had to compete with other publishers for new work by active authors. There were 11 new titles published in the BAFS, including the first three volumes of Katherine Kurtz’s voluminous Deryni series, Joy Chant’s Red Moon and Black Mountain, three of Evangeline Walton’s four books treating The Mabinogion, and Poul Anderson’s Hrolf Kraki’s Saga.

202

NOTES

8. I include here the works that Ballantine published before the series proper but were kept in print in tandem with the series—appearing with them on the advertising pages in Ballantine releases and so on—and of which Carter wrote, “They are all books I would certainly have urged Ballantine to publish” (1Carter 268). 9. One novel and one novella by Pratt/de Camp, as well as one solo Pratt novel; a couple of scattered short stories by Howard, de Camp, and Vance; nothing at all by Leiber. 10. The BAFS included three MacDonald volumes, though Carter favored Morris’s romances as fantasy’s “first” rather than MacDonald’s Phantastes and Lilith due to the latters’ visionary frames; Lewis and Merritt were often cited by Carter (and rotated into his “list”), but since their relevant work was available elsewhere, little was included in the BAFS. 11. The assumption that stories involving wizards or dragons or magic swords were by their nature suited to children, often coupled with an aggressively derisive attitude, was something that the phenomenon that culminated in the BAFS had to deal with. No doubt Ballantine’s label “Adult Fantasy” reflects this. Nevertheless, works from Victorian author George MacDonald’s Curdie books, to The Hobbit and the Narnia series, to Ursula Le Guin’s initial Earthsea trilogy have been seminal in the development of the genre. Of course, the “children’s” aspect of these books was often tiptoed around (genre paperback editions would appear with no note of YA origins, as with the Bantam editions of Le Guin’s Earthsea trilogy) or even dismissed (the three stories included in Evenor, a BAFS George MacDonald collection, were published for children in the nineteenth century, but Carter insists they are “too serious, too meaningful,” to be considered children’s fairy tales; 3Carter ix). 12. Among the reasons for this, probably the most notable was the sale of Ballantine Books and its ultimate acquisition by Random House, which took bottom-line publishing decisions out of the hands of genre afficianados like Betty and Ian Ballantine and put them in the hands of the executives of what was emerging as one of the major corporate publishing conglomerates of the United States. While the BAFS had not been a huge money-maker (it had not, for example, found the secret to following up on Tolkien’s mass success), it had done reasonably well—but reasonably well was clearly not good enough for Random House. In addition, reprints by dead authors are a nonrenewable resource, and once the stores of Morris, Dunsany, Cabell, and Eddison began to deplete, interest and sales began to flag somewhat. The “slowdown” in the series did precede the arrival of Random House on the scene. 13. Abetted by the folding of Lancer Books, after a decline, in the mid-1970s and the realignment of Pyramid Books following its sale to Harcourt Brace Jovanovich (another up-and-coming corporate publishing conglomerate) in 1975. 14. Ursula Le Guin’s Earthsea trilogy was published by Bantam in the fall of 1975, and Patricia McKillip’s The Forgotten Beasts of Eld by Avon, also in the fall of 1975.

NOTES

203

15. Pratt/de Camp’s The Compleat Enchanter, comprising all the Unknown Harold Shea stories, published in two volumes previously; Pratt’s The Well of the Unicorn; and H. Warner Munn’s Merlin’s Godson, comprising King of the World’s Edge (originally published in Weird Tales in 1939) and The Ship From Atlantis, to which the BAFS Merlin’s Ring had been a sequel. 16. The selections were clearly culled from the more successful titles, and the pre-Tolkien works included Dunsany’s The King of Elfland’s Daughter and The Charwoman’s Shadow, a single-volume edition of William Morris’s The Well at the World’s End (an exception in maintaining the BAFS cover art and Carter’s introduction), Eddison’s The Worm Ouroboros and Zimiamvia trilogy, Hope Mirrlees’s Lud-in-the-Mist, and, in the early 1980s, James Branch Cabell’s six titles. Tolkien’s work, of course, remained perennially available. 17. A slight hedge—The Island of the Mighty (originally titled The Virgin and the Swine), Walton’s treatment of the Fourth Branch, had been published in 1936, and the other three were drafted, but not published, in the years immediately following. But even though drafted decades before, these latter three—The Children of Llyr, The Song of Rhiannon, and Prince of Annwn— were of course revised and became BAFS first publications. 18. Terry Brooks’s The Sword of Shannara, which hovered near the top of trade paperback bestseller lists for most of the second half of 1977, and Stephen Donaldson’s first Thomas Covenant trilogy, whose paperback editions were issued in the summer of 1978. Both were published by del Rey/Ballantine, and both spawned countless sequels and prequels; together, the two launched fantasy’s final jump to bestseller-genre status. 19. The Boyer and Zahorski anthology, Fantasists on Fantasy (1984), is a good barometer: less than half of the authors precede the genre period, most of those do not work within a conceptual framework that more than rubs elbows with the BAFS template, and the specific literary form discussed tends to be fairy tale/story or romance rather than “fantasy”—a point to which I will return. 20. Works such as Andrew Lang’s Pantouflia books, Barrie’s Peter Pan, and MacDonald’s “The Golden Key” and “The Giant’s Heart.” MacDonald’s Lilith is the only modern work not published for young readers, and associated in some capacity with the genre (it was included in the BAFS), to which he refers. Other references tend toward older, traditional works, such as the Egyptian “Tale of Two Brothers,” Volsunga Saga, and the contents of Andrew Lang’s colored fairy-tale books. 21. With chapters on William Morris, Dunsany, Lovecraft, Eddison, Howard, Pratt, Smith, and Tolkien, framed by more general opening and concluding chapters. 22. Kingsley’s The Water Babies is a formal amalgam, though it does include significant elements associated with the BAFS template; Lewis’s Space trilogy, particularly Out of the Silent Planet and Perelandra, would formally belong to the science fiction subgenre of the planetary romance, though the vocabulary of that subgenre has much in common with BAFS fantasy; MacDonald’s

204

23. 24.

25.

26.

27. 28.

29.

30. 31. 32.

33.

34.

NOTES

Lilith and Phantastes and Peake’s Gormenghast trilogy are distinct in crucial ways (I will return to this) from most of the BAFS canon but were published in the series. The most recent edition appeared in 2004. I don’t remember seeing The Game of the Impossible until many years after its publication; I do, however, remember seeing Jackson’s book around the time it came out—at a point when fantasy meant Morris, MacDonald, Dunsany, Eddison, and Tolkien to me—and finding myself bemused. Jackson would certainly fall into Shippey’s academic category. While he does not name Jackson or her book, he alludes to “a recent book on fantasy” with a list of authors on its back cover that is exactly the same as that listed on the back cover of Jackson’s book (Shippey xxi). On the basis of frequency of reprints and the fact that there are no other easier-to-find anthologies of “fantasy” stories from the period on the current used-book market. The same can be said of the bulk of the “fantasy” that appeared, for instance, in The Magazine of Fantasy and Science Fiction during the 1950s. In fact, qualifiers are still sometimes used: Michael Moorcock speaks of “epic fantasy,” while “high fantasy” is frequently used to describe the kind of material at the core of the BAFS. The scenarios of Robert M. Coates’s “The Hour after Westerly” and John Cheever’s “The Enormous Radio,” chosen randomly from Ray Bradbury’s deliberately chosen anthology Timeless Stories for Today and Tomorrow. Horace L. Gold’s “The Trouble with Water,” which appeared in Unknown in 1939. Common in, for example, Lovecraft. Genres are constructed and, as such, tend often to be rendered in more unequivocally absolute terms than their attendant texts warrant. The two “fantasies” here are, of course, separated by a broad grey area, not a clear dividing line. Early on, many of the tales of E. T. A Hoffmann (“The Golden Pot,” “The Nutcracker,” “Meister Floh”) maintain the kind of ambiguity central to Irwin’s and Jackson’s conceptions yet are imbued with the kind of fairy-tale elements common in BAFS-style fantasy. Robert Holdstock’s more recent tales of Ryhope Wood are similarly set in this grey area. Of course, many fantasies for young readers, from Nesbit’s Five Children and It to Alan Garner’s The Weirdstone of Brisingamen to J. K. Rowling’s Harry Potter series, also occupy the grey area. Others, however, that initially play on the “reality” question move clearly and unambiguously into an objectively depicted “other” world as real as ours: Lewis’s Narnia series is an example of this. See particularly Dark Imaginings: A Collection of Gothic Fantasy (1978), the only of the five of the Boyer and Zahorski anthologies to contain a section specifically devoted to “Low Fantasy”—here, specifically, “Gothic Low Fantasy.” With a few apt additions of authors whose revival postdated the BAFS (such as Kenneth Morris) or who have only been sparsely touched on in this context (such as James Stephens).

NOTES

205

35. The last two contrasts particularly may be seen to relate to a contrast in external circumstance: most of the literary writers were not reliant on their writing as a primary means of support (Eddison worked for the British Board of Trade, Tolkien was a professor at Oxford, etc.), while most of the popular writers were. The latter needed to be more aware of “the market” and hence of current popular narrative forms; their outputs on the whole tend to contain much higher proportions of work that is not fantasy (in our BAFS sense) than their literary compeers. The literary writers were less pressed to adhere to the perceived demands of the market. 36. Haggard’s influence, direct and indirect, on pulp fantasy will be elaborated in Chapter 6. Howard, of course, had many other influences than Haggard, though some of the most important—Edgar Rice Burroughs, for example— were themselves strongly influenced by Haggard. Of our literary fantasists, Morris read Haggard with great enthusiasm and wrote him a letter of introduction when he travelled to Iceland; E. R. Eddison sent Haggard a copy of The Work Ouroboros when it was published; Tolkien read all the Haggard that he could get his hands on, and (along with Morris) Haggard was the only modern writer he would concede as an influence. 37. Morris, Dunsany, Cabell, Eddison, Tolkien, and Mirrlees. 38. Lovecraft, Smith, Howard, de Camp, Pratt, Leiber, Vance, and Bok. 39. Though I include “worlds” corresponding to those of actual tradition (Walton’s Wales, Anderson’s Scandinavia, etc.) under my qualification of the BAFS template, I am being more literal here. 40. Whose “mythology,” as is now well documented, was not developed as background to The Lord of the Rings: the latter work was an outgrowth of the mythology. 41. I am restricting myself to Lovecraft’s “Dreamlands” work here. A more organized “pantheon,” of course, is developed in the Cthulhu Mythos cycle, but these tales belong to the horror and Gothic traditions. 42. While allusions to, and sometimes appearances of, deities occur fairly frequently in Eddison, Smith, Howard, and Leiber, they do not really stack up to what may properly be termed a “mythology.” The shapeshifting fluidity and ambiguity that characterize Cabell’s mythological elements keep them from ever solidifying into much of a structure. 43. Who is actually rather meagre in terms of strict sociohistorical detail, except in the case of the Shire. His annals and such, modeled on ancient and medieval works such as The Anglo-Saxon Chronicle, Geoffrey of Monmouth’s Historia Regum Britannia, and Saxo Grammaticus’s Gesta Danorum, tend to legendary narrative in their form, not to the practical and “realistic” concerns of modern historians. 44. It is also noteworthy that these examples are from popular writers who wrote for Unknown (Pratt and de Camp, though The Blue Star was a postUnknown work) or who were strongly influenced by the work published in it (Anderson). Editor John W. Campbell, most famous as the editor of Astounding who demanded adherence to genuine theory and a logical rigor

206

45.

46.

47.

48.

NOTES

to the “science” in the science fiction he published, likewise expected a certain critical logic with regard to the treatment of magic in the stories for Unknown. A result of this was that magical elements were lifted into relief, and magic often became a theme in itself. For example, William Morris’s Child Christopher and Goldilind the Fair, where the magic consists of one premonitory dream and a few passing allusions to fairy folk in the woods, who never appear. Unknown is perhaps most closely related, of all the subsections of pregenre fantasy, to the actual beginnings of the genre. Chronologically, it is the latest, and work drawn from it constituted a large chunk of the reprinted Sword and Sorcery fiction during the 1960s. This will be discussed further in Chapter 6. Can one look at, say, the naturalistically developed world of The Lord of the Rings and the explicitly metaphysical world of Phantastes and conclude that Tolkien and MacDonald were really doing the same thing? The claustrophobic and in fact generally nonmagical world of Gormenghast Castle and the magical parallel world of Narnia? I would say not. The presumed collective goal, equally central to all five writers, is “to make their fantastic worlds as real as our own” (12). But the degree to which Manlove adapts his framework from author to author raises the question, again, as to whether this goal reflects “what they were trying to do,” or whether it simply reflects a critically extrapolated lowest common denominator more geared to what Manlove is trying to do. For example, after using a brutally reductionist reading of George MacDonald’s essay “The Fantastic Imagination” to conclude that “the very provision by him of a theoretic background for the comprehension of works supposedly incomprehensible is a contradiction in terms” (68), he uses his conclusion as a stepping stone to evaluating MacDonald’s “fantasy,” whose goal is to evoke “only the unconscious and imaginative side of the mind. But though he tries to shut out the conscious selves of science and law, intellect and will, they keep coming back to interrupt the proceedings” (98). In Phantastes, the key cited evidence of this includes a series of quotations in which the narrator jumps from being “a would-be musician of the emotions” to being “the Aberdeen chemistry graduate” (78–79). To Manlove, these must be seen as incompatible and hence symptomatic of MacDonald’s failure to make his fantasy world “real.” But in the essay, MacDonald does not reject the use of the intellect in making meaning (which Manlove affirms but does not engage) but suggests that “a genuine work of art must mean many things” (MacDonald 317) and that the reader should discover their own meaning through reading, not by using the author’s intended meaning (if any) as the prime determinant. In his comments on the passages he cites from Phantastes, Manlove seems to assume that the ruminations of the fictional character Anodos are simply interchangeable with MacDonald’s perspective. He neglects the possibility that the “contradiction” he asserts as undermining the success of Phantastes might in fact be a deliberate tension, itself of thematic significance to

NOTES

49.

50.

51.

52.

53.

54.

55. 56.

207

the work. This said, Manlove’s criticism of MacDonald does overlap with something many readers (including Tolkien) have criticized: MacDonald’s tendency to sermonize and the occasional to frequent (depending on the reader) forcing of his material to make a predetermined point. But these flaws are in fact more glaringly apparent in MacDonald’s realistic fiction: Manlove’s attempt to couch it as symptomatic of MacDonald’s failure as a “fantasy writer” seems too deliberately contrived, his evidence manipulated to turn the author’s arguable flaws into a broad, generic failure. If Manlove had used his fundamental framework to discuss Le Guin’s initial Earthsea trilogy, Donaldson’s first Thomas Covenant trilogy, James Blaylock’s The Elfin Ship and its sequels, Yolen’s Great Alta trilogy, and Rowling’s Harry Potter series, he would have been on firmer ground. I can only find two instances of the term fantasy being applied generically by pregenre writers to their work. One is in the title of Clark Ashton Smith’s self-published chapbook collection of stories, The Double Shadow and Other Fantasies (1933), where it is rather doubtful Smith had something like the BAFS template in mind; the other, more marginal, is in C. S. Lewis’s preface to That Hideous Strength: A Modern Fairy Tale for Grown-Ups (1945), where Lewis states, “I have called this a fairy-tale in the hope that no one who dislikes fantasy may be misled by the first two chapters into reading further” (2Lewis 7). Here the term is attached to a story that has considerably more in common with the supernatural novels of Charles Williams than with Tolkien or Morris. Lewis’s subsequent discussion suggests that he was aware that the book strayed from the vocabulary of the traditional fairy tale. The chronological focus of Swinfen’s study, as indicated by its subtitle, is centered later than the focus of this. The passage I will discuss, however, speaks precisely to the present issue. Tolkien also does not say that talking beasts cannot appear in fairy-stories. The difference would be that in fairy-stories the beast form is not a “mask upon a human face”: Gwaihir the Eagle Lord is, in fact, an eagle. As one might look back to Mary Shelley’s Frankenstein and Poe’s The Narrative of Arthur Gordon Pym to find manifestations of elements which became “defining” when science fiction coalesced early in the twentieth century. Two Carter-edited anthologies published in the BAFS, Dragons, Elves, and Heroes (1969) and Golden Cities, Far (1970), drew specifically on this type of “ancient fantasy,” as Carter termed it. However, Carter’s distinction between “ancient” and “modern” is a bit murky: these anthologies include selections by Voltaire, Robert Browning, Kenneth Morris, Anatole France, and others that are as “modern” as Morris. Peake, as I have noted, is marginal in relation to the BAFS template also, while Williams is even further afield. Mainly translations and retellings of Norse myths (see 2Manlove 2). Some of the “fairy” collections he mentions, such as Keightley’s The Fairy Mythology (1828), also epitomize material outside the reach of the traditional fairy tale proper.

208

NOTES

57. Which particularly coalesced in the aforementioned French contes de fees, with relevant ancestors in the Italian collections of Straparola (1550/1553) and Basile (1634–36). 58. The daughter of William Morris who edited and wrote the commentary for the 24 volumes of The Collected Works of William Morris (1910–15). 59. Of course, the authors were well aware of this. Tolkien, for instance, certainly would have been conscious that, in “On Faerie Stories,” he was using “romance” in a more expansive sense than the more precise formal sense with which it could have been applied to Sir Gawain and the Green Knight. 60. The two last, products of the Renaissance and of consciously “literary” artists, tend in important respects away from the other work noted here. Ariosto treated his traditional matter, the Carolingian legends, with a free hand; in key respects, The Faerie Queene at least is perhaps closer to The Well at the World’s End than to Malory. On the other hand, the traditions they both drew on were, in the sixteenth century, still “living.” 61. It is true that, in some areas, oral traditions did last through the nineteenth century and beyond: many tales of Finn MacCumhail were collected by folklorists in Scotland and Ireland not more than a century ago; parts of the Volsung legends were still told in ballad form in the Faroe Islands in the early twentieth century; The Kalevala was processed from material collected in the nineteenth century. But these living remnants were/are remote from the worlds of modern scholarship and literary endeavor, and the form in which our authors invariably encountered them was the mediated text. 62. After Jesus Christ Superstar and Life of Brian, this has changed—though such interpretations of biblical material can still generate newsworthy controversy. 63. In this light, it is interesting to note that none of the contents of Boyer and Zahorski’s Visions of Wonder: An Anthology of Christian Fantasy (1981) in fact retell biblical stories. Several move outside the “stable, known quantity” and draw on apocryphal Christian legend. The only major “biblical” fantasy I know of, MacDonald’s Lilith, connects with Genesis not through the actual biblical book but via the apocryphal Judaic legends of Lilith, Adam’s first wife. But even this much is mainly background to a largely invented story in an invented world, and at any rate, MacDonald’s visionary intentions (Lilith, he said, was written under a “mandate from God”) liken his choice of subject to Milton’s choice for Paradise Lost rather than, say, White’s choice of the Arthurian legends. I am being deliberately narrow here and restricting myself to works that explicitly engage biblical material—that is, retell a story from the Bible. In a broader sense, of course, one can see obvious biblical influences: the imprint of the Old Testament can be discerned on the shape of Tolkien’s “Quenta Silmarillion,” the influence of the King James Bible on Lord Dunsany’s prose is readily apparent, and so on. 64. Evangeline Walton’s The Sword Is Forged, the beginning of a fantasy version of the Theseus legends, may be seen as a partial exception—though

NOTES

65.

66.

67.

68.

69.

70.

71. 72.

209

not published until 1983, it was originally drafted in the 1940s. The World’s Desire (1890), a continuation of the story of The Odyssey by Andrew Lang and H. Rider Haggard reprinted in the BAFS, may be seen as borderline. This is evident in the uniform, blanket exoticism that characterizes virtually all pseudo-Oriental fantasy tales—where there is little to distinguish Persia from Arabia from Kashmir from Cathay from Tartary and so on. That this is predicated on distance is undeniable. Though to varying degrees: for instance, while Smith was familiar with The Thousand and One Nights, and the Burton translation may have had some impact on his prose style, Beckford’s Vathek and Poe’s “Masque of the Red Death” are more apt sources for what are ultimately traditional romance elements. At the other end of the spectrum, both Pratt’s The Well of the Unicorn and Anderson’s The Broken Sword do draw to some extent on narrative techniques derived from the sagas—a form both had an expert knowledge of. But even here, popular adventure and historical romance fiction are arguably greater determinants of narrative style and strategy. De Camp characterized Morris’s romances as “combining the antiquarian romanticism of Scott with the supernaturalism of Walpole” (2de Camp 9)—a not terrifically apt description of Morris, but considerably more apt if applied to much popular fantasy—in the introduction to the 1965 Sword and Sorcery anthology The Spell of Seven. In Clute and Grant’s sketch of the authors representing the “heart of this enterprise” (in 1997’s The Encyclopedia of Fantasy, noted earlier), the canonical authors are preceded by a specious “the kind of fantasy that evolved from a few decades before the beginning of the nineteenth century,” followed by Hoffmann and Poe (Clute and Grant viii). The openings, in order, of The Sundering Flood, The Well at the World’s End, The Roots of the Mountains, and The House of the Wolfings. The remaining four romances, as well as the four fragments included in volume 23 of the Collected Works of William Morris, have similarly conventional openings. Particularly in the “medieval” portion of the Biography of Manuel. For example, Jurgen opens with “It is a tale which they narrate in Poictesme . . .” and Figures of Earth with “They of Poictesme narrate . . .” For example, Domnei opens with a “Critical Comment” that attributes the work to “Nicolas de Caen, one of the most eminent of the early French writers of romance” (Cabell xv); in the “Note on the Shire Records” preceding the text of The Lord of the Rings, an account is given of the manuscript history from which the work itself is derived. The scenarios of C. J. Cutliffe Hyne’s The Lost Continent and H. Warner Munn’s King of the World’s Edge, respectively. This was facilitated by a degree of expertise: Tolkien, a linguist, was professionally preoccupied with the traditional literature that informs his Middle-earth corpus; though not an academic, William Morris was, in his day, a notable expert on many aspects of medieval and ancient Germanic culture, including their literatures, of which he produced many translations;

210

73.

74.

75. 76. 77. 78.

79. 80.

81.

NOTES

Kenneth Morris and James Stephens respectively knew the Welsh and Irish originals of the narratives they interpreted in their fiction, and Morris was an academic with an expert knowledge of “traditional narrative forms” the world around; Eddison’s expertise was a relatively private affair, though he did publish a translation of Egil’s Saga, and the translations from Greek and Latin incorporated into his fiction were his own; one would surmise, simply on the basis of their work, that both Cabell and Dunsany had an intimate familiarity with our “traditional narrative forms.” These include The Life and Death of Jason (1867), The Earthly Paradise (1868–70), and The Story of Sigurd the Volsung and the Fall of the Niblungs (1876). The Prologue to The Earthly Paradise framing the retold tales is an original story in imaginary settings. This reflects a number of things: (1) the emergence of modernism after World War One, the influence of T. S. Eliot on poetic theory, and subsequent emphasis on compact, allusive verse forms coupled with a dislocation/ fragmentation of perspective; (2) a corresponding rereading of Romantic and Victorian poetry, with a new emphasis on work embodying elements anticipating modernism, and an eclipse (in some cases, virtually total) of the extended narrative poetry so highly regarded through the end of the nineteenth century; and (3) a resulting assumption that extended narrative—at least modern extended narrative—found its natural medium in prose. This third assumption perhaps accounts for the neglect of Romantic and Victorian narrative poetry in accounts of the development of modern fantasy. Such as Beowulf, Sir Gawain and the Green Knight, The Poetic Edda, The Kalevala, Orlando Furioso, and The Faerie Queene. Such as Malory’s Le Morte d’Arthur, Volsunga Saga, The Tain, The Mabinogion, and Snorri Sturlusson’s Prose Edda. The Wood beyond the World, Child Christopher and Goldilind the Fair, and The Water of the Wondrous Isles. Moorcock’s discussion also posits a clear distinction between commercial, or popular, work intended to please the masses, and artistic, or poetic, work. The sensationalism of his “popular fantasy” beginning with Amadis and Palmerin naturally allies this work with the former and distinguishes it not only from traditional work but from Spenser, Milton, Anatole France, and Italo Calvino as well. I have some problems with what seems to me Moorcock’s reductionist and simplistic division along these lines, but this part of his discussion is rather outside the present concern here. As well as the basis for uncounted sequels. Which is not to say that a great deal of what appears in Montalvo’s text is not invented; however, there is a great deal of invention in, say, the Arthurian literature of the Middle Ages, which Moorcock, as noted before, distinguishes from Amadis and its followers as intended historically. The narrative textures, the interlace structure, and many characters and episodes are clearly derived from the French Arthurian Vulgate Cycle and its fourteenth-century Iberian adaptations.

NOTES

211

82. The sequels—the first of which, the Sergas de Esplandian, concerning Amadis’s son, was written by Montalvo during the years he worked on Amadis proper—amount collectively to many times the bulk of the parent work, itself of considerable dimensions (just as the continuations of Chretien’s incomplete Perceval dwarf the parent work in length). Of the many spinoffs, Palmerin of England is the most well-known (and, together with Amadis, exempted from the flames in Cervantes’s Don Quixote) and the only to become widely available in English. 83. An incomplete translation of Amadis appeared during the course of the 1590s, and Anthony Munday’s full translation appeared in 1618–19, both from the French version; Munday’s translation of Palmerin preceded Amadis. 84. The elaborate interlace structure and chivalric atmosphere of The Faerie Queene would seem to have been informed by Amadis, though Ariosto and the cyclic Arthurian romances must figure here, too. The allegorical dimension of Spenser’s work is alien to Amadis and its followers. Sidney was no doubt indebted to Amadis for his atmosphere and setting, but Arcadia is not driven by fast-paced martial adventure, largely eschews magical elements, and is more concerned with courtly intrigue and manners and political philosophy. 85. The so-called English Spenserians, such as Giles and Phineas Fletcher, made little use of those elements The Faerie Queene had in common with Amadis; later prose romances, such as Mary Wroth’s Urania (1621) and Margaret Cavendish’s The Blazing World (1666), follow Arcadia in curtailing emphasis on sensationalist martial and magical content while maintaining a setting and atmosphere akin to that of Amadis. 86. A severe abridgement—to roughly one sixth the length of the original— appeared in 1702. This summary condensation, which we might liken to a Cliff ’s Notes plot synopsis, would have been the form in which Amadis was known to most eighteenth-century English readers. 87. One might in fact argue that the “artificial” Iberian chivalric romances exist in relation to earlier medieval romance in much the same way that contemporary bestseller genre fantasy exists in relation to the pregenre canon. 88. Even Southey’s Thalaba the Destroyer and The Curse of Kehama, products of the same period during which the poet worked on Amadis and Palmerin, show little in the way of unique influence stemming from those works. 89. As briefly noted before, the widespread popularity of the Iberian chivalric romances was made possible in large part due to the introduction of the printing press. The early years of the printing press in England saw a considerable proliferation of romances, mainly in prose. Malory’s Le Morte d’Arthur, printed by William Caxton in 1485, is probably the most famous of these; Lord Bernier’s translation of the Carolingian faerie romance Huon of Bordeaux, which appeared in 1535, has often been cited in the fantasy context. But Caxton and his successor, Wynkyn de Worde, issued quite a few other romances as well. These included romance-style redactions of the legends of Troy, the Carolingian Four Sonnes of Aymon, both adapted

212

90.

91.

92. 93. 94.

95.

NOTES

from French sources by Caxton, as well as a number of “artificial” romances, adapted mainly from French by Caxton and others. The latter, including Valentine and Orson, the Historye of Olivere, and others, tended to be considerably less extravagant and shorter than their Iberian counterparts. No doubt due to the relative dearth of magical incident, these romances have been largely ignored with relation to fantasy. However, though it is well-known that, in the books designed for the Kelmscott Press in the closing years of his life, William Morris drew heavily on the work of Caxton and de Worde in the visual aspects of book design, it tends to go unnoticed that Morris paid close attention to the literary content of more than Malory. In many respects, Morris’s romances owe more to these less-known prose romances than to Malory. While Arthurian romance was already in decline when Caxton produced his printed edition of Malory in 1485 and continued to decline throughout the sixteenth century, it was not yet fully eclipsed, despite the new emphasis on the “historical” emperor rooted in Geoffrey of Monmouth. Malory’s Le Morte d’Arthur had not yet receded to the status of a rare antiquarian volume, as it had by the first half of the eighteenth century, and was very much a part of the literary culture to which Spenser belonged. It is no accident that Spenser, and The Faerie Queene in particular, in fact went through a revival in the period of Warton. Though The Faerie Queene had never been “lost” or “forgotten,” it had been viewed critically through the Restoration and Augustan periods, regarded as a negative example, rather than a model, for practicing poets. To the extent that the poem was revered, it was with suspicion—like something you can’t get rid of. By contrast, the post-Augustan eighteenth century, as I will discuss in Chapter 2, embraced Spenser enthusiatically, and the enthusiasm was partly linked to his association with the world of Arthur and “ancient” romance. Of which Shakespeare’s “Venus and Adonis” and Marlowe’s “Hero and Leander” are probably the most well-known examples. Such as Robert Greene’s Pandosto, or the Triumph of Time, the basis of Shakespeare’s A Winter’s Tale. In fact, a story by Tutuola appears in Terri Windling and Mark Alan Arnold’s 1984 anthology Elsewhere Vol. III: Tales of Fantasy, in the company of Robin McKinley, Angela Carter, James Blaylock, and others. For The Mists of Avalon and The Druid’s Tune, respectively.

Chapter 2 1. In terms of what most English majors from my generation got, it is not really reductionist at all. The second half of the “Survey of English Lit” course that I took in the spring of my freshman year (1978) began with Pope, included a few poems by Johnson, and then jumped straight to Blake. The course did not really touch on the novel, and the upper-level courses that included

NOTES

2. 3. 4. 5. 6.

7. 8.

9.

10. 11. 12.

13. 14.

15.

16.

213

eighteenth-century work did not venture much beyond Defoe, Fielding, and Richardson. For the full text, see Dryden 458. For the full passage, see Pope 55–56. For the full passage, see Thomson 222–23. For the full text, see Lonsdale, 46–51. This is not the place to detail the tangled textual history of The Arabian Nights. Suffice to say, there was no authoritative edition in Galland’s day, and the possibility of a genuinely authoritative version is dubious. Husein Haddawy gives a concise account of the pre-Galland history in the introduction to his Arabian Nights, which does a credible, if necessarily speculative, job of distilling the original “core.” Galland did add spurious stories to the cycle, some heard from a Syrian priest living in Paris, some possibly of his own invention. The latter was included in the BAFS anthology Golden Cities, Far. And partially explains why Rasselas has been granted particular critical attention in the canonical narrative. Of course, the fact that it is by Johnson, and of high literary quality, are also factors. Only one version of the first episode, “The History of Prince Alasi and the Princess Firouzkah,” was published; the second, “The History of the Princes and Friends Alasi and Firouz,” was suppressed, no doubt due to its homosexual subject, and not published until 2001, in Kenneth Graham’s Vathek with the Episodes of Vathek. In that Beckford was English, though he wrote Vathek in French. For instance, Penguin Classics included it, along with The Castle of Otranto and Frankenstein, in Three Gothic Novels. Which is not to draw too incontrovertible a line between the two: Vathek certainly contains Gothic elements that its quasi-Oriental compeers lack; a number of authors (Beckford, Walpole, and Clara Reeve) wrote in both modes. More broadly, both were forms of popular romance that drew on the conventions of sentimental fiction, and both stood apart from the developments of Defoe, Fielding, and Richardson. This should not be taken as precisely identical to nineteenth-century conceptions of the two, but there is an obvious connection. Thomas Parnell’s “A Fairy Tale in the Ancient English Style” (1721) is initially situated “in Britain’s Isle in Arthur’s Days,” but the tale corresponds to the Irish folktale “A Legend of Knockgrafton,” collected by T. Crofton Croker and included in Fairy Legends and Traditions of the South of Ireland (1825). It seems most likely that Parnell, an Irishman, simply patched the tale, with English names, into an Arthurian frame—evidence of Arthur’s lack of currency. It was only with Thomas Tyrwitt’s edition of Chaucer (1775–78) that it was recognized that the closing e was pronounced, enabling recognition of metrical regularity. A short extract appeared in the BAFS anthology Dragons, Elves, and Heroes.

214

NOTES

17. For instance, though Pratt and de Camp send Harold Shea to the world of FQ in “Mathematics of Magic,” nothing of Spenser’s narrative textures is evident: the work is a straightforward adventure story. 18. Pratt and de Camp send Shea to the world of Orlando Furioso in The Castle of Iron; an extract, translated into prose by Richard Hodgens, appeared in the BAFS Golden Cities, Far (1970) and then the first volume of a projected full translation (1973). 19. Thomas Warton was one of these, and the fruits of his researches appeared in his voluminous History of English Poetry (1775–81). Though scarcely reliable by present standards (and criticized by some contemporary scholars), it is, in its early sections, a remarkable attempt to make sense of a wide array of then largely unconsidered material, with some sharp intuitions despite the built-in limitations of knowledge and perspective. 20. Such judgments, usually fueled by combined lack of adequate understanding and simple prejudice, continued long after Percy’s time: a century and a half later, Tolkien would address (with eminent “antiquarian” expertise) similar judgments of Beowulf, by scholars who tended to view the poem as a linguistic and cultural “specimen,” in “The Monster and the Critics.” 21. This is the context of the historical Arthur. 22. Thomas Tyrwitt at first believed Chatterton’s claims. 23. Walpole’s Castle of Otranto was also initially published as a translation of an Italian work by Onuphrio Muralto originally published in 1529, though this seems to reflect Walpole’s fears that it would be ridiculed: as soon as it garnered positive responses, he affirmed his authorship. James Ridley’s Tales of a Genii was published as a translation from the Persian of Horam, the Son of Asmar, who is given a lengthy biography in the introduction. 24. Which he was the first English literary figure to make use of. 25. Ariosto and Tasso, translated into English during the time of Spenser, were translated anew by John Hoole: the latter’s Jerusalem Delivered in 1763 and Rinaldo in 1792, the former’s Orlando Furioso in 1773–83. 26. Ossian was considered an indigenous “Homer”; Walpole would write, “Read Sindbad the Sailor’s voyages, and you will be sick of Aeneas’s” (Mack xviii–xix). 27. Poems, though they appeared as prose on the page.

Chapter 3 1. Like Percy before him, Scott freely revised (or “improved”) his texts, so though Minstrelsy is a seminal collection, it is an imperfect one from an exactingly “scientific” viewpoint. 2. Which also included collections from abroad, such as Southey’s (1808) and Lockhart’s (1823) translations of Spanish ballads. 3. Probably the most exhaustive collection until the encyclopedic work of Francis James Child later in the century.

NOTES

215

4. Ritson had also edited several ballad collections during the closing decades of the eighteenth century, in which he was quite meticulous with regard to “authenticity”; his virulent critiques of the shortcomings of Percy’s and Thomas Warton’s work made him notorious. 5. The long Ibero romances also returned to print: both William Stewart Rose’s verse and Robert Southey’s prose translations of Amadis of Gaul appeared in 1803 and the latter’s Palmerin of England in 1807. Rose also published translations of Orlando Furioso (1823–31) and Berni’s revision of Boiardo’s Orlando Innamorato (1823). 6. This would include material collected orally, as was the case with much of the contents of Scott’s Minstrelsy: while versions collected would inevitably have the stamp of their present-day tellers/singers, they were in origins quite old. 7. “A Gest of Robyn Hode,” about three times the length of “Rime,” appearing in Ritson’s 1795 collection of ballads about the outlaw, is one of the rare exceptions. 8. As a few negative reviews at the time averred: one reviewer opined that it was “the extravagance of a mad German poet,” and Southey suggested that “it is a Dutch attempt at German sublimity” (Gardner 21). 9. Who had also translated some of Burger’s literary ballads from the German during the 1790s. 10. Many of Hogg’s ballads were included in The Mountain Bard, whose first edition appeared in 1807. 11. Many of Southey’s ballads were included in Metrical Tales (1805). Slight when considered in the context of Southey’s prolific output, these shorter pieces nevertheless stand among his most successful poems. 12. He did anticipate accurately: in the introduction to the 1830 edition, he wrote that “the goblin story [was] objected to by several critics as an excrescence upon the poem” (Scott 5). 13. Like the traditional ballad, not all medieval metrical romances included magical or supernatural elements. 14. The “Ettrick Shepherd,” as he was called, had begun his adult life as a shepherd. 15. A performance for Mary Stuart following her arrival in Scotland in 1561. 16. Though Hogg did append his longer poems with Scott-style annotation. 17. His class status, combined with a literary sensibility lacking the genteel cultivation of Scott and others—evident in a directness in addressing shocking subjects (“indelicacy”) and unapologetic recourse to the supernatural—led to his being alternately patronized and ridiculed by many of his contemporaries, including Scott; later editions of his work, including the collected editions of both poetry and prose, were freely bowdlerized and sanitized, with the result that, without access to rare original editions, Hogg was known until recently mainly through filtered versions of his work. 18. Which, arguably, connects him more strongly to Silko and Tutuola, whom I discussed at the end of Chapter 1, than to Morris or Tolkien—or Scott.

216

NOTES

19. Matthew Arnold’s Sohrab and Rustum (1868) is the only like example I am aware of. 20. In 1838, he wrote that “there was nothing oriental in the style. I had learnt the language of poetry from our own great masters and the great poets of antiquity” (Southey 16). 21. It is worth noting here that, whereas the first-generation Romantics were adults when the flood of publications of old and traditional material began and grew up without it (and, of course, Scott and Southey were themselves instrumental in bringing much of it into new and/or first printed editions), the younger generation would have first encountered such material as children and were perhaps more apt to take it for granted and less apt to judgments like Scott’s. 22. The poem (like Beattie’s) is not cast as a dream, it should be noted. Nevertheless, the progression of the narrative and the symbolic import of the protagonist’s encounters amount to a “pilgrim’s progress” of the poetic imagination. 23. The poem did have an effect on George MacDonald, the first chapter of whose Phantastes takes its heading from the poem. Some of the structural contours of Shelley’s plot can be discerned, hazily, in Phantastes as well. 24. The former in ballads such as “The Mask of Anarchy,” the latter in the early Queen Mab, written in 1810 and circulated among friends but not published in Shelley’s lifetime. 25. For the full text, see Keats 39. 26. See Shelley 323. 27. Shakespeare’s Venus and Adonis and Marlowe’s Hero and Leander are the most famous of these, though there are numerous others by Thomas Lodge, Michael Drayton, and others; George Chapman completed Marlowe’s unfinished poem. 28. These flourishes particularly touch the poems on Greek subjects. 29. It is Beddoes “Pygmalion” that George MacDoanld turned to for the heading of chapter 7 of Phantastes. 30. Outside the Keats line of influence, this period also saw the lone, brief poetic fantasy by Wordsworth, the Arthurian The Egyptian Maid, composed in 1830 and published in 1835. Like “Lyulph’s Tale” in Scott’s Bridal of Triermain, while the poem alludes to traditional Arthurian legend, incorporating characters and motifs derived from medieval sources, the narrative itself is invented. 31. The Brownie of Bodsbeck (1818) is historical, and the “Brownie” of the title is a human being; the possibly supernatural elements in The Three Perils of Woman (1823) and The Private Memoirs and Confessions of a Justified Sinner (1824) are ambiguous. 32. It has been entirely neglected in studies of fantasy and is not even mentioned in the entry for Hogg in Clute and Grant’s Encyclopedia. No doubt this is partly due to the bowdlerization (see note 17), which hit TPM especially hard: the novel was not well received, partly, it would seem, due to its fantastic content, objections to which were sufficiently strong that Hogg pared down the story to its bare historical narrative, retitled The Siege of Roxburgh,

NOTES

33. 34. 35.

36.

37.

38.

39. 40. 41.

42. 43. 44. 45. 46.

47.

217

for his collected tales in the 1830s. It was this form of the story that most readers from later generations would have encountered, and the text of the original edition was not republished until 1973. Such as Headlong Hall (1815) and Nightmare Abbey (1818). Which had appeared in the Myvyrian Archaiology of Wales (1801–7). Gawain is Gwalchmai, Kay is Cei, and so on. It might be noted that Misfortunes features the same cast of characters, and some plot parallels, with the Welsh “Taliesin” (which is sometimes included in translations of The Mabinogion, though it is found independently in a manuscript later than those in which the tales of The Mabinogion are found). But unless Peacock had access to a manuscript, this could not have been a source, since Lady Guest’s translation did not appear until 1838. It is only mentioned in passing in theme-based entries in Clute and Grant’s Encyclopedia; Peacock’s entry comprises only references to those themebased mentions. Further afield, Disraeli’s Swiftian The Voyage of Captain Popanilla (1828) deserves passing mention here, as does his historical romance The Wondrous Tale of Alroy (1834), set in twelfth-century Baghdad, which incorporates the supernatural, though annotated after the manner of Scott: “Authority may be found in the traditions of the Hebrews for the introduction of all these spiritual agencies” (Disraeli vii). “Ixion” has appeared twice in post-BAFS anthologies: Boyer and Zahorski’s The Phoenix Tree (1980) and Stableford’s The Dedalus Book of British Fantasy (1991). Rather oddly, he is remembered primarily as the subject of a biography by his friend Thomas Carlyle. Though tending to the Gothic, Bulwer-Lytton’s Asmodeus at Large (1833) and Zanoni (1842) also deserve passing mention here. Disraeli’s two tales bear some resemblance to the tales on classical subjects in Richard Garnett’s Twilight of the Gods (1888/1903), though it is not clear this results from influence. Ruskin’s King of the Golden River (1839) and MacDonald’s Phantastes, both German influenced, might bear some comparison to Bulwer-Lytton and Sterling, but this may result from the common influence of German Romanticism. The tale of Cosmo at the center of Phantastes is thoroughly Hoffmannesque. “Trusty Eckardt” (1801) adapts the Tannhauser legend. Again, Eovaai’s Eastern orientation and topical/satirical elements partly compromise it in this capacity. There is no “pre-Adamitical” age or Chinese intermediary as in Eovaai. A 348-page (1874 edition) fairy tale not targeting children—though John Duke Coleridge suggests it is suited “especially for the young” (Coleridge vii)—would not have been expected to draw much attention. Ashley’s statement that it is modeled on FQ is one example; designation of the work as “for children,” though neither of the nineteenth-century editions suggests this, is another.

218

NOTES

48. It does have its flaws: its hero is saved a little too often by magical allies rather than by his own wits or courage; he too frequently just happens to be in the right spot, magically disguised or concealed, to overhear enemy conversations that reveal crucial information; and some readers may be disconcerted by the total lack of any geographical sense of the invented world.

Chapter 4 1. Jonathan Cott’s Beyond the Looking Glass (1973), Jack Zipes’s Victorian Fairy Tales (1987), and Michael Patrick Hearn’s Victorian Fairy Tale Book (1988). 2. With a century of development behind it to build on, it is worth noting that the catchall antiquarianism began to subdivide into the more specialized disciplines that we now recognize as philology, linguistics, archaeology, anthropology, comparative mythology, and folklore. 3. Around the beginning of the Victorian period, various “Clubs” and “Societies” began to appear specifically dedicated to editing and publishing old texts. Foremost among these, and still in operation, is Oxford University Press’s Early English Text Society, which commenced publishing in 1864. 4. There had been paraphrases and translations of extracts prior to this, but the only full translations had been into Danish and Latin. 5. The mythological dimension so clear in Volsunga Saga is almost imperceptible in Nibelungenlied, which is also layered over with French romance elements. 6. Though, in the highly compact saga style, considerably shorter than the Nibelungenlied, which nevertheless treats only the latter part of the story. A series of missing manuscript pages falls right in the midst of the sequence of related poems in The Poetic Edda. 7. To this, one might append Geoffrey of Monmouth’s Latin Historia Regum Britannia, which appeared in English translation by J. A. Giles in 1848. It is clear that Geoffrey drew heavily on ancient Welsh tradition, but it is equally clear that those traditions were freely altered, embellished, added to, and ordered to suit his intentions. It was Geoffrey’s expansive treatment of Arthur that was (and is) regarded as the most significant part of the Historia, yet the emperor depicted by Geoffrey is a considerable distance from the insular Welsh king that appears in the “Triads” or in “Culhwch and Olwen,” the only of the Arthurian stories in The Mabinogion predating French courtly tradition. 8. The Kalevala appeared in English in selections at various points throughout the period; the only full translation, by John Martin Crawford (1888), was by way of a German translation. W. F. Kirby’s translation (1907) was the first made directly from the Finnish. The Russian bylini appeared in English in Ralston’s Songs of the Russian People (1872) and Hapgood’s Epic Songs of Russia (1886).

NOTES

219

9. While significant parts of their conceptual underpinnings are no longer tenable, Max Muller and Frazer’s Golden Bough (first ed. 1890) is still cited in more than a period-piece capacity—unlike Jacob Bryant’s work. 10. German Popular Tales, Edgar Taylor’s English translation, appeared in 1823. 11. Not all “folktales” are “fairy tales,” nor do all—as with ballad and romance— contain magical or supernatural elements. Nevertheless, the term fairy tale was, and is, often used loosely and imprecisely: though many of the tales in the Grimm collection are not fairy tales, this does not prevent editions (such as Pantheon’s edition, which I now stare at on my bookshelf) from being titled The Complete Grimm’s Fairy Tales. 12. The popular image of the brothers combing the rural villages of remote regions of Germany and painstakingly transcribing the words of aging informants, which they present verbatim to their readers, has been dismantled: they did have informants, but they were generally educated urbanites who may have grown up in a “storytelling” context; the stories were freely tampered with, especially with the young audience in mind. See Zipes. 13. For example, The Boy’s King Arthur (1883). 14. For example, Charles Kingsley’s The Heroes (1856)—subtitled Greek Fairy Tales. 15. For example, Dalziel’s Illustrated Arabian Nights Entertainments (1865). 16. This would be evidenced by Andrew Lang’s famous 12-volume colored fairy book series, which began with the Blue Fairy Book (1889) and continued into the twentieth century, concluding with the Lilac Fairy Book (1910). The reader will find tales from folktale sources such as Grimm, tales from The Arabian Nights, adaptations from epic and saga (“The Tale of Sigurd”) and from classical myth (“Cupid and Psyche”), as well as literary tales from the French contes de fees, Hans Christian Andersen, and even Jonathan Swift. Needless to say, many of these were extensively bowdlerized. 17. There were exceptions to this, and both Robert Browning’s “Childe Roland to the Dark Tower Came” (1852/1855) and Christina Rossetti’s “Goblin Market” (1859/1862) are brief (204 and 567 lines, respectively) and their content invented. Both embody a compressed intensity shared by a number of Romantic but few other Victorian poetic narrative fantasies. Some connections might be made between these poems and later fantasy (the barren landscape of Browning’s poem and Tolkien’s Mordor, the goblin fruit of Rossetti’s poem and the Fairy Fruit of Hope Mirrlees’s Lud-in-the-Mist), but both also anticipate a more modernist consciousness than is generally associated with fantasy. 18. Thomas Bulfinch’s Age of Fable (1855), Age of Chivalry (1858), and Legends of Charlemagne (1863) are probably the most famous of these. 19. As well as the earlier Victorian period “The Vision of Launfal” (1843), by the American poet James Russell Lowell, and the voluminous King Arthur: An Epic Poem (1848), by Edward Bulwer-Lytton. Despite its title, the former has nothing to do with Thomas Chester’s fourteenth-century romance or Marie de France’s earlier lai, placing its eponymous hero in the context

220

20.

21. 22. 23.

24.

25. 26.

27.

28.

29. 30.

NOTES

of the grail quest. The matter of Bulwer-Lytton’s poem is almost entirely invented. “Guinevere,” narrating the queen’s last meeting with Arthur, was Tennyson’s invention. Most of the others are based on Malory, but Tennyson also turned to French (“Merlin and Viviane”) and Welsh (“Enid”) sources. Which also saw the publication of “The Lotos Eaters” (1832), noteworthy in the present context in drawing, like Keats, on ancient Greek matter. Based on an Italian poem rather than Malory, the basis of his later treatment of the same legend. The discovery of the Winchester manuscript in the twentieth century suggests that the final shape (Caxton’s continuous 21-book text) was, like Tennyson’s Idylls, arrived at and that the initial “plan” was for separate but related tales. Lapses in quality can be seen in the war with Rome and in the Tristram section, which simply loses its way (and is by far the longest of the tales). Swinburne wrote a few shorter Arthurian poems early in his career (unpublished in his lifetime), including “Queen Iseut” and “Joyeuse Gard,” both early assays of the Tristram legend. The later Tale of Balen (1896) follows Malory and is sparer and less given to the excruciating emotional and descriptive excesses of Tristram. As well as fantasy not actually Arthurian but informed by Arthurian elements: Cabell’s and Tolkien’s work would be among these. Twain’s A Connecticut Yankee in King Arthur’s Court (1889) savagely lambasts not Tennyson directly but the whole fabric of Tennysonian medievalism; Clemence Housman and T. H. White, with their realistic, psychologically penetrating character studies, in effect invert Tennyson (though White simultaneously writes against Twain). “The Death of Dermid” and “The Tain-Quest,” Fenian and Ulster tales respectively, appeared in Lays of the Western Gael (1864); the book-length Congal (1872), based on a non-“cyclic” legend, followed. The inaccurate “Paul Revere’s Ride” in fact supplanted history in the popular conception of Revere. Schoolchildren were set to memorize it: I remember my grandfather (b.1899) reciting it in the 1960s. One of the first reasonably reliable collections of American Indian oral tales and, in Longfellow’s time, the most extensive. Manabozho (Nanabozho, Nanabush, and Wenebojo are among the numerous variants) is the protagonist of hundreds of Chippewa (Canadian Ojibwa, self-named Anishnabe) oral tales. The core tale of Manabozho in Schoolcraft distills major episodes that turn up frequently elsewhere, and it is distinctly sanitized compared to many more risqué tellings. Some have accused Schoolcraft of bowdlerization, which, when “naughty bits” were routinely translated into Latin even in scholarly publications into the twentieth century, is certainly possible. But generally, Schoolcraft was scrupulous in such “scientific” matters, and it is more likely that his chief “informant,” his wife and a Christianized Chippewa, herself censored the tales in the telling.

NOTES

221

31. Tolkien admired Longfellow and Hiawatha, and the verse form of the fragmentary “Lay of Leithian” was in part inspired by Hiawatha. 32. With the possible exception of “Sir Galahad, A Christmas Mystery,” which depicts its hero, among other things, envying his father’s affair with Guenevere. Of the arrogantly self-righteous protagonist of Tennyson’s “Galahad,” published in the 1842 collection, Morris drily remarked, “Tennyson’s Galahad is rather a mild youth” (Thompson 79). 33. In Middle English, poetic adaptations include the alliterative Geste Hystoriale of the Destruction of Troy, John Lydgate’s massive Troy Book, as well as Troilus and Creseide and “The Knight’s Tale” by Chaucer. Prose works adapted from French by Caxton include The History of Jason and The Recuyell of the Historyes of Troy. 34. For instance, Constantinople is referred to by its Norse designation, Micklegarth. 35. Matthew Arnold should be given passing mention here for three shorter narrative poems: “Tristram and Iseult” (1852) conveys a more psychologically based emotional power than Swinburne, and his treatment of Iseult of Brittany is notable. “Balder Dead” (1853), drawing on the Prose Edda, is founded on Norse myth, and “Sohrab and Rustum” (1853) was based on plot abstracts of Ferdowsi’s Shah Nameh. Both of the latter see Arnold attempting to cast his material in terms of the classical epic. 36. EP was published in three large volumes; some editions were in four volumes. It is, possibly, the longest poem in the English language. 37. With roots going back to Italy: Straparola’s Facetious Nighrts (1550/3) and Basile’s Pentameron (1634). 38. Jack Zipes’s account of this in the introduction to his Victorian Fairy Tales is particularly sharp and succinct. 39. For instance, the famous illustrator George Cruikshank, a teetotaler, introduces alcohol as the culprit vice (the giant in “Jack and the Beanstalk” is a drunkard, etc.) in his adaptations collected in George Cruikshank’s Fairy Library (1865). 40. See Anne Thackeray Ritchie’s “Beauty and the Beast” (1874) and Juliana Horatia Ewing’s “Amelia and the Dwarfs” (1870), included in Auerbach and Knoepflmacher’s Forbidden Journeys (1992). 41. And still one of the most widely read and frequently reprinted—it is the only tale to appear in all three of the recent anthologies in Note 1. 42. Sufficiently unobtrusive that one R. H. Coe, introducing a 1916 American edition of the story, would, after reflecting unapprovingly on Ruskin’s later preoccupation with social reform, write, “There is nothing of all this in ‘The King of the Golden River’ . . . it was written merely to entertain” (Coe iv). 43. Many of his most famous tales, including “The Emperor’s New Clothes,” “The Ugly Duckling,” and “The Little Mermaid,” are straightforward moral allegories. 44. Though this could also have been suggested by Hoffmann and Fouqué, as well as Ruskin.

222

NOTES

45. Sometimes titled “The Day Boy and the Night Girl.” 46. Other, less successful short tales by MacDonald include “Cross Purposes” (1867) and “The Carosoyn” (1866/71); “The Wise Woman” (1874) is similar in length to “Photogen and Nycteris.” 47. Stockton’s tales appeared and reappeared in many collections, of which the most significant were The Floating Prince and Other Fairy Tales (1881), The Bee-Man of Orn and Other Fanciful Tales (1887), and The Queen’s Museum (1887). 48. No doubt partly due to Tolkien’s several references in “On Fairy-Stories.” 49. Though the Welsh name Gwyntystorm and the rugged terrain surrounding castle and city evokes a Welsh/Scottish feel in the Curdie books, this is only a feel. 50. The story begins at the time of the battle of Flodden Field in 1513 and proceeds to the years following; the exactly described location is in fact the area Lang grew up in. 51. Mike Ashley’s contention that the work “was the first English children’s fantasy novel which did not draw from folklore roots” (Clute and Grant 540) seems to be accurate. 52. Ashley notes (Clute and Grant 540) that it was the success of The Water Babies, along with the recommendation of George MacDonald, that prompted Macmillan to publish the first Alice book. 53. In Imaginary Worlds, Lin Carter guardedly allows discussion of some descendants of the Victorian fairy tale, including Baum’s Oz books and Lewis’s Narnia series, though “being children’s fantasies, [they] really do not come within the scope of this history” (1Carter 107). In Literary Swordsmen and Sorcerers, L. Sprague de Camp entirely omits anything published for young readers; Michael Moorcock devotes a short section of Wizardry and Wild Romance (the “Epic Pooh” chapter) mainly to children’s work, though Victorian material is largely ignored. Richard Mathews mentions some fantasy for young readers (and without any of Carter’s disclaimers) in passing, but it is mainly post-Victorian work, and he describes none save Le Guin’s Earthsea in any detail. All the work discussed here (as well as much that I did not mention) is duly affirmed in Clute and Grant’s Encyclopedia, though the “heart of this enterprise” list of writers includes only the peripheral Carroll and MacDonald. Of the major studies with something of an historical canvass, only those of C. N. Manlove include any extended discussion of Victorian children’s work without especial qualification. 54. “The Golden Key,” “The Wise Woman,” and “The Carasoyn” appeared in Evenor (1972); “The Romance of Photogen and Nycteris” was included in New Worlds for Old (1971). 55. Tolkien discusses Lang in this light in “On Fairy-Stories” (Tolkien 136). 56. MacDonald’s essay “The Fantastic Imagination,” a precursor to Tolkien’s essay, gives his more detailed articulation of these ideas. 57. In contrast, the descendants of the Gothic (ghost stories, tales of the supernatural, occult tales, etc.) were amply prolific.

NOTES

223

58. For example, “The Story of Bhanavar the Beautiful,” narrated by protagonist Shibli Bagarag, is a romance, while “The Punishment of Shahpesh,” told by Fehsnavat the Vizier, is a teaching tale. 59. See Caracciolo 201ff for a brief discussion on how The Arabian Nights continues as a factor in Meredith’s later realistic novels. 60. See 1Carter x and 2Carter 19–20. Carter maintains his contention that Morris is rightfully “first” on the grounds that the settings of Shagpat and Vathek are “real, but romanticized” and constitute “‘literary’ versions of the actual Middle East” (1Carter). Of course, one could argue that the settings of most of Morris’s romances are “literary versions” of the actual Middle Ages, and while the landscape of Shagpat is identified as Persia, Arabia, and so forth, this is impressionistic, and the geography of the text is far more imagined than real. 61. Two of Twilight’s tales, “The Poet of Persepolis” and “The City of Philosophers,” appeared in the BAFS Discoveries in Fantasy (1972). While Garnett’s work has been represented reasonably frequently in fantasy anthologies during the post-BAFS period, there does not seem to have ever been a full edition of Twilight attached to the fantasy genre. 62. Most notably, the very brief, compact story of the Poet, a Novalis-esque parable, inset in the verse drama Within and Without (1855), MacDonald’s first book. The more developed “The Castle,” behind which the shadows of both Novalis’s “Klingsohr’s Tale” and Goethe’s “Fairy Tale” are easily discerned, initially appeared in Adela Cathcart. 63. The evil ogre, the magical prohibition, the supernatural helper, to note a random few. 64. The story of the winged women with no arms has a distinct Novalis flavor; the tale of Cosmo is unadulteratedly Hoffmannesque. The chivalric figure of the Knight seems inspired in part by Fouqué. 65. Including Lady Byron and Charles Dickens. 66. Interestingly, Wells’s Time Machine appeared the same year. 67. One supposes that three decades of novel writing affected MacDonald’s approach to dramatic action. 68. In my introductory “Science Fiction and Fantasy” class, Lilith, though some find it objectionably intractable, is a great discussion piece—and it does occasionally elicit “This is the best thing I ever read.” The one time I tried Phantastes, it got the most exclusively thumbs-down response I’ve ever seen. 69. To these can be appended a number of fragmentary romances, four of which were included in volume 21 of The Collected Works of William Morris (1914). One of these, “The Folk of the Mountain Door,” Carter included in his postBAFS anthology Realms of Wizardry (1976). I omit A Dream of John Ball (1888) and News from Nowhere (1890), which are utopian narratives. 70. Oakenrealm, “so much a wood-land . . . that a squirrel might go from end to end, and all about, from tree to tree, and never touch the earth” (2Morris 3), finds its description in a common medieval description of England. The story was also a loose play off of the Anglo romance “Havelock the Dane.”

224

NOTES

71. Morris finished it as he was dying and in fact dictated the closing pages from his deathbed. 72. This has unfortunately been taken to mean that the romances, and especially the post-Germanic romances, are idle pleasantries with little substance: a blemish, for instance, in E. P. Thompson’s otherwise superb William Morris: From Romantic to Revolutionary. Studies by Carol Silver and Frederick Kirchhoff implicitly repudiate this notion. 73. Out of print as individual works marketed for the “general reader,” though the 22-volume Collected Works, expensive and more intended for libraries, of course included the romances. 74. Though at least Phantasmion and Eovaai embody all the key elements of the BAFS template and are not metaphysical narratives or published for children, Morris was the first to develop these elements in a sustained way through more than one work. 75. Michael Faletra has compared the linguistic underpinnings of Morris’s approach to archaism to Tolkien’s in an unpublished essay. 76. While Morris did not write or publish any fairy tales in the Victorian juvenile sense, he did read them aloud to his children: one would suspect that he was as interested in what he was reading as they were. 77. Carter included this in Great Short Novels of Adult Fantasy (1972). 78. Which can be seen implicitly in one reviewer’s reflections on The Sundering Flood: “If worked out in plain and simple language it would have made a pleasing essay in the genre of fairy tales for children . . . But if the author’s interest had lain in the growth and development of mental qualities, the Cotswold Hills of the nineteenth century would have afforded a better stage” (Faulkner 430–31). In other words, for Morris’s subject to have been acceptable, it should have been cast as a children’s story; if he was writing for adults, he should have written a Victorian three-decker.

Chapter 5 1. This is partly attributable to newer ideas on child psychology, including the idea that fairy tales would lead to fundamental confusions about reality. The proliferation of the ideas of Freud had also put a considerable damper on the perception of the fairy tale as a region of innocent wonder. 2. Tolkien’s poem appeared in 2013, Lewis’s in Narrative Poems (1969), which collected this and the three poems mentioned following. 3. Published in The Lays of Beleriand (1985), which also contains the beginning of a recasting of the “Lay of Leithian.” 4. The magical disenchanting of a group of statues in “The Nameless Isle” looks forward to the corollary scene in The Lion, the Witch, and the Wardrobe; the Turin and Beren and Luthien fragments are of course parts of the Middleearth “Legendarium.” 5. Lewis, at least, explicitly wished to generate a movement to counter Eliot and modernism.

NOTES

225

6. The translation-cum-adaptations of Lady Gregory’s Cuchulain of Muirthemne (1902) and Gods and Fighting Men (1904) were widely distributed; an ample number of more scholarly translations, by Eleanor Hull, A. H. Leahy, Joseph Dunn, and others, also appeared during the early twentieth century. 7. Some of E. Nesbit’s less well-known work, such as The Magic City (1910), and L. Frank Baum’s non-Oz work, such as Queen Zixi of Ix (1905), are among these. Later examples include A. A. Milne’s Once on a Time (1917), in the Thackeray/Lang vein, and Owen Barfield’s lone fantasy, The Silver Trumpet (1925). 8. The Marvelous Land of Oz (1904) takes place entirely in Oz; all the others follow the heroine Dorothy. 9. Walter de la Mare’s The Three Mulla-Mulgars (1910) also descends from the beast fable, though in setting and narrative style it is closer to the fairy tale. 10. These tales, including The Hobbit, were written for his children and not initially with publication in mind. 11. Which, perhaps, leaves this question: is Giles, ultimately, a children’s fantasy? As a small volume laced with illustrations by Pauline Baynes, so it appears. But by this point, Tolkien felt that writing “for children” was, in effect, writing to a fictitious audience and not entirely respectful to children. 12. Chronicles straddles LOTR in terms of publication dates, but Lewis had spent years hearing Tolkien’s drafts read aloud at Inklings gatherings. 13. Lewis’s intimate familiarity with non-children’s material discussed here is also evident: the very Scots world of Ettinsmoor; the Arabian Nights feel of Calormen; the Renaissance courtly atmosphere of parts of Prince Caspian and The Silver Chair; the list could go on. 14. Howard Pyle published a four-volume retelling of the Arthurian legends, largely derived from Malory (1903–10); Lewis Spence and H. A. Guerber both produced Bulfinch-style redactions of varied bodies of myth and legend during the opening decades of the twentieth century. 15. Opposing, for instance, Irish independence from England. 16. Apparently motivated in part by Yeats’s omission of Dunsany from the Irish Academy of Letters in 1932 on the grounds that he had written nothing about Ireland. 17. Carter included plays in Beyond the Fields We Know (1972) and Over the Hills and Far Away (1974), the latter of three BAFS collections from Dunsany. 18. The first was At the Edge of the World (1970). 19. There were three further tales: the aforementioned “Idle Days,” followed by “A Shop in Go-By Street” and “The Avenger of Perdondaris” in Tales of Three Hemispheres, where all three appeared under the subtitle “Beyond the Fields We Know.” 20. This can be seen in comparing the tone and approach of “Idle Days” to the earlier Pegana tales. The high decorum of the latter is mitigated by the intimacy of the narrator, a modern Irishman who travels to the “Lands of Dream,” in the former.

226

NOTES

21. Some of what followed is fantasy in the broad sense; the BAFS Dunsany anthologies included a few of the tall tales of Jorkens, which began appearing in The Travel Tales of Mr. Joseph Jorkens (1931). 22. The title is a bit misleading, insofar as, despite fidelity to the content of the originals, the voice and texture are entirely Stephens’s. 23. According to Douglas Anderson, nearly all were written before 1919, when Morris’s health suffered a breakdown (Morris 18). Insofar as Morris frequently adopted pseudonyms, the number of stories is not conclusive. 24. About the last third of the book was simply lopped off for the original edition; the library edition by Arno Press (1978) is a facsimile of this cut edition. 25. Domnei (revised from The Soul of Melicent, 1913); The Cream of the Jest (1917); Figures of Earth (1921); The High Place (1923); The Silver Stallion (1926); and Something about Eve (1927). The shorter The Music from behind the Moon (1926) was appended to Domnei; The Way of Ecben (1928) appeared in The Young Magicians (1969). Jurgen was not included, as it was already widely available in a pocket edition from Avon. 26. Summarized in the formal mock-genealogy The Lineage of Lichfield (1921). 27. Excepting most of Cream of the Jest, set, as noted, in the twentieth century; The High Place (late seventeenth / early eighteenth century); and Something about Eve (first half of the nineteenth century). Also, the medieval material written prior to Domnei and the flowering of the idea of Poictesme, most notably the stories in The Line of Love (1905) and Chivalry (1909), is not fantasy but historical romance set in an idealized Middle Ages—though, with some revision, it was roped into the Biography. 28. It alone deals with Manuel himself. 29. Including the fantasy trilogy The Nightmare Has Triplets, comprising Smirt (1933), Smire (1935), and Smith (1939). 30. The Clemence Housman entry in Clute and Grant’s Encyclopedia does not mention it. 31. Published in the United States as The Return of Kai Lung (1937). 32. After the drama of Kai Lung in the hands of the bandit Lin Yi before the first story in Wallet, Kai Lung’s presence reduces to banter with his audience and then to simply being identified as the teller. 33. Only a handful of Morris’s tales fall unequivocally into this vein; Dunsany very often employs a quasi-Oriental sensibility but does not set tales in Baghdad, Persia . . . , or China. 34. Cabell’s best fantasies date to the decade following the war, as do what are arguably Bramah’s two finest Kai Lung volumes. 35. It should be noted that he was from same generation as Dunsany, Stephens, and so on, and the major characters and basic plot pattern of The Worm Ouroboros, his first work, date to his childhood. 36. It is interesting that Styrbiorn appeared in a library edition in Arno Press’s Lost Race and Adult Fantasy Fiction series in 1978 (and that Eric Brighteyes was included in Newcastle’s series in 1974). 37. See the works of Poul Anderson.

NOTES

227

38. “Lud” was an ancient name for London; the “Fairy Hills” resonate with the Welsh Hills. 39. Leslie Barringer and Donald Corley might be mentioned here. The former’s trilogy, comprising Gerfalcon (1927), Joris of the Rock (1928), and Shy Leopardess (1948), was included in the Newcastle series. The setting is invented—an imaginary French province, Neustria—though magic is sparse and the narrative style novelistic. Corley’s more mannered work, echoing Cabell and Dunsany, was collected in The House of Lost Identity (1927) and The Haunted Jester (1931); two stories were included in the 1972 BAFS Discoveries in Fantasy, and Carter anticipated a possible Corley collection that never materialized. 40. Including Mistress Masham’s Repose (1946) and The Elephant and the Kangaroo (1947), as well as many of the stories included in The Maharajah and Other Stories (1981). 41. As well as to White’s explicitly pacifistic stance. 42. Collected in God Likes Them Plain (1935), Sealskin Trousers and Other Stories (1947), and A Sociable Plover (1957). None of these were included in the BAFS, though “The Abominable Imprecation,” from the first volume, appeared in Boyer and Zahorski’s The Fantastic Imagination II (1978), and “The Goose Girl,” from the second, appeared in their The Phoenix Tree (1980). 43. In 1914, Tolkien wrote to his wife, “I am trying to turn one of the stories [the story of Kullervo, from The Kalevala] . . . into a short story somewhat along the lines of Morris’ romances” (2Tolkien 7). This led to the story of Turin. 44. It is interesting that, as the century wore on and the archaistic approach to translations of medieval and ancient texts lost ground in favor of more idiomatic approaches (compare Jesse Byock’s 1990 translation of Saga of the Volsungs to the Morris/Magnusson translation), Tolkien’s tendency to archaism became more muted: this is evident in LOTR. It is also interesting to compare the “Book of Lost Tales” version of the Turin tale and the postLOTR version published as The Children of Hurin (2007). 45. “The New Shadow” (set in Gondor about a century into the Fourth Age) and “Tal-Elmar” (set in the Second Age) are fragments; “Aldarion and Erendis” (set in Second Age Numenor) is “complete,” but, in Christopher Tolkien’s words, by way of “a degree of editorial rehandling that made me doubtful of the propriety of including it” (1Tolkien 7–8). The first two appeared in The Peoples of Middle-earth (1996), the latter in Unfinished Tales. 46. One of the early compressed versions of the core Silmarillion narrative, included in The Shaping of Middle-earth. 47. A couple of generations prior to Arthur in Geoffrey of Monmouth. 48. And two honorary visits by E. R. Eddison in 1944, when LOTR was substantially underway and Eddison was involved in his last work, The Mezentian Gate. 49. Such as Manuel’s journey to challenge the magician Miramon Lluagor with the sword Flamberge in Figures of Earth, which clearly echoes Alveric’s

228

NOTES

journey to challenge the magician Gaznak with the sword Sacnoth in “The Fortress Unvanquishable.” Insofar as Cabell’s novel-length story bears no other notable resemblances to Dunsany’s short tale, this is more of a deliberate conceit than “influence.” 50. As the heroic sections of LOTR may be said to resemble those of Worm. 51. I have already suggested that it is difficult to discern signs of influence of Morris on Dunsany, and it is likewise difficult to discern such influence in the work of Stephens, Kenneth Morris, or Cabell. Eddison praised Morris, but the possibility of actual substantial influence would be restricted to Styrbiorn the Strong, where it would seem more likely deriving from Morris’s translations than his prose romances. 52. Some of the scathing modernist critiques were to some degree justified.

Chapter 6 1. Tolkien, Lewis, and Kenneth Morris were academics; Dunsany was a member of the landed aristocracy, and only with the Depression did he begin to rely on earnings from his work—after all his major fantasy had been written; Eddison worked for the British board of trade; William Morris was by trade a designer. 2. Fletcher Pratt, who knew Icelandic and read the sagas in the original, and Poul Anderson, who also knew Icelandic, would be partial exceptions to this. Their respective The Well of the Unicorn (1948) and The Broken Sword (1954) come closest to the work discussed in Chapter 5. 3. Of the popular work, only Pratt’s The Well of the Unicorn operates on an Eddison-scaled canvas. 4. Clark Ashton Smith, the only of the popular writers to cultivate a deliberately dense, poetic style, frequently had his stories ruthlessly edited, and on one occasion, Wright temporarily rejected a story (“The Tale of Satampa Zeiros”) with the note that it “reminded me of Lord Dunsany’s stories in The Book of Wonder. However, I fear that Lord Dunsany’s stories would prove unpalatable to most of our readers” (Connors and Hilger 263). 5. Living respectively in Rhode Island, California, and Texas, Lovecraft, Smith, and Howard never met, but they did actively correspond, sharing drafts of their work, complaints about Wright, and so forth. Both Smith and Howard contributed stories to Lovecraft’s Cthulhu mythos. 6. There were, of course, countless “pulp” periodicals during the period between the wars, with varied specialties (SF, detective stories, etc.). Weird Tales and Unknown are among the most highly regarded in general; as far as venues that published stories in the BAFS vein, they were, as Lin Carter often reiterated, the most important. 7. The BAFS included editions of Machen’s The Three Imposters and Hodgson’s The Boats of the Glen Carrig (1908) and The Night Land (1912)—all among the works Ballantine published skirting the BAFS template.

NOTES

229

8. The major work, The Dream-Quest of Unknown Kadath, was only published posthumously in the Beyond the Wall of Sleep (1943) omnibus from August Derleth’s Arkham House. 9. Zothique (1970), Hyperborea (1971), Xiccarph (1972), and Poseidonis (1973). 10. Apparently as a means to augment his income to support his ailing parents. 11. The Star-Treader (1912), Ebony and Crystal (1922), and Sandalwood (1925). 12. Smith translated the incomplete episode “The Story of the Princess Zulkais and Princess Kalilah” from Beckford’s French and wrote a conclusion. This was included in the BAFS anthology New Worlds for Old (1971). 13. Smith translated Le Fleurs du Mal. 14. L. Sprague de Camp notes that, while Smith had read Dunsany, he claimed Poe and Ambrose Bierce had influenced him more (de Camp 207). 15. The first, King Solomon’s Mines (1885), one of many to feature that precursor of Indiana Jones, Allan Quatermain, was in fact spurred by the wish to write a better adventure story than Stevenson’s Treasure Island (1883). 16. The People of the Mist (1894), a non-Quatermain African-set lost race tale, and The World’s Desire (1890), a sequel to The Odyssey written with Andrew Lang. 17. There is often an ambiguity in the seemingly miraculous in Haggard and some of the other writers here: is Ayesha’s life prolonged by magic or by some advanced scientific method? 18. Lin Carter discusses the Barsoom series at some length in Imaginary Worlds. Widely available elsewhere at the time, nothing from the Barsoom series appeared in the BAFS. 19. It would not be amiss to point out that Haggard was enthused on by many of the literary writers: William Morris wrote Haggard a letter of introduction when the latter travelled to Iceland prior to writing Eric Brighteyes; E. R. Eddison sent a copy of The Worm Ouroboros to Haggard upon its publication; apart from William Morris, Haggard is the only modern writer Tolkien would concede as an influence. In the latter instance, I would see this influence in the development of atmosphere—the view of Minas Morgul, the gates of Argonath—rather than in terms of language (of which Haggard was notoriously ambivalent) and narrative strategy. 20. According to Leiber, all the Fafhrd and Gray Mouser stories submitted to Weird Tales, during Farnsworth Wright’s editorship and after, were rejected. John W. Campbell accepted them at Unknown, but he “more than once remarked in accepting a [Fafhrd and Gray Mouser] story, ‘This is more of a Weird Tales piece than Unknown usually prints. However . . .” (Leiber 14). 21. Bok wrote a more Unknown-ish tale, published in cut form as “The Blue Flamingo” in Startling Stories (1948). The full text appeared in the BAFS under the title Beyond the Golden Stair (1970). 22. Land of Unreason appeared in the BAFS in 1970. Most of the Pratt/de Camp collaborations were available elsewhere at the time, and the only Shea tale to appear (in 1972’s Great Short Novels of Adult Fantasy I) was the later “Green Magician,” mentioned following.

230

NOTES

23. The Incomplete Enchanter and Land of Unreason were instances of this. 24. Pratt identifies the world of the Well with that of Dunsany’s play “King Argimenes and the Unknown Warrior.” But insofar as the tale and the play share no characters, and there is no direct or significant connection between the events of the tale and those of the play (which takes place off of Pratt’s map and at a considerably earlier time), it is difficult to call this actual “influence.” In its construction and narrative style, Well is distinctly un-Dunsanian. 25. Pratt was most well-known as a popular historical author when he died in 1956. 26. James Blish’s There Shall Be No Darkness and Leiber’s Conjure Wife. 27. A faun astray from southern Europe appears early in the story! 28. H. Warner Munn should be added—the only Weird Tales writer from the Wright period to return to fantasy in the genre period. 29. Not necessarily archaism, and certainly not incompetently executed archaism.

Conclusion 1. Other S&S series by Moorcock with roots in this period include those of Erekose, Dorian of Hawkmoon, and Corum. 2. Other writers working within the ultimately pulp-rooted frame of reference, sometimes in work with overtones of science fiction, include Marion Zimmer Bradley and Lin Carter. Roger Zelazny’s Amber series, beginning with Nine Princes in Amber (1970), is of particular note here. 3. The pulp-rooted writers often seemed somewhat ambivalent of Tolkien: see 1Carter 115–17 for Leiber’s assessment; Moorcock has in fact been somewhat notorious in his condemnations of Tolkien’s work. 4. Though, even here, there is no sense of a consciousness of the burgeoning genre. Smith is very much a “literary fairy tale” and was in fact spurred by MacDonald’s “The Golden Key”: Tolkien had agreed to provide an introduction to a new edition of MacDonald’s tale, but upon rereading it, he felt a strong distaste for it. Smith was written instead of the introduction (see Flieger 69–75). 5. Adams returned to the world of Shardik in Maia (1984); Tales from Watership Down (1996) included some of the myths and legends alluded to in its parent work. 6. The Book of Three (1964), The Black Cauldron (1965), The Castle of Llyr (1966), Taran Wanderer (1967), and The High King (1968). The Foundling and Other Tales of Prydain (1973, exp. 1999) collected backstory narratives. 7. A Wizard of Earthsea (1968), The Tombs of Atuan (1971), and The Farthest Shore (1972). These were preceded by two short stories published in Fantastic Stories of Imagination: “The Word of Unbinding” and “The Rule of Names” (both 1964). Le Guin later returned to Earthsea with Tehanu (1990), The Other Wind (2001), and a number of shorter tales collected in Tales of Earthsea (2001).

NOTES

231

8. The School for Wizards on Roke, it might be mentioned, stands as an antecedent of J. K. Rowling’s Hogwarts. 9. For example, Fendarl as a rather passive echo of Sauron. The scene presaging the final confrontation with Fendarl, corresponding to “The Last Debate” chapter, is rhetorically and dramatically almost lifted from LOTR. 10. Chant’s two subsequent books set in Khendiol, The Grey Mane of Morning (1977) and When Voiha Wakes (1983), postdate the 1960–74 period. Less successful commercially, both works, while maintaining a Tolkienian tone and decorum, dispense with direct borrowings and are, all told, more original. 11. Both McKillip and Yolen have subsequently published a considerable number of fantasies, some published for young readers, some not. McKillip’s Riddle Master trilogy, comprising The Riddle Master of Hed (1976), Heir of Sea and Fire (1977), and Harpist in the Wind (1979), remains her most wellknown work, though since the late 1980s she has been very prolific. Yolen’s work, often drawing on her expertise as a folklorist, has been more various, and a high point in her fantasy is the White Jenna trilogy, comprising Sister Light Sister Dark (1989), White Jenna (1990), and The One Armed Queen (1998). 12. Over Sea Under Stone (1965), The Dark is Rising (1973), Greenwitch (1974), The Grey King (1975), and Silver on the Tree (1977). 13. This last again underscores the arbitrariness of the adult/juvenile distinction in fantasy. As noted in Chapter 1, both the Earthsea Trilogy and Forgotten Beasts appeared in non-age-specific fantasy editions in 1975; Alan Garner’s two aforementioned books would do so in the early 1980s. 14. Most emphatically in Le Guin, but also in Chant, McKillip, Garner, and Cooper. 15. Which, by the 1970s, were beginning to be almost obligatory. 16. The Newcastle Forgotten Fantasy Library took up the BAFS gauntlet from 1974 to 1979, but Newcastle was a small independent press publishing tradesized editions and cannot be seen as a barometer for what was happening in the commercial expansion of the new genre. 17. As noted in Chapter 1, during 1973–74 the BAFS itself had become increasingly dominated by new titles. 18. The Birthgrave (1975); Shadowfire (1978; Vazkor Son of Vazkor in the United States); and Quest of the White Witch (1978). 19. Night’s Master (1978) and Death’s Master (1979); three further volumes appeared in the 1980s. 20. Revised to its benefit as The Silver Sun (1980). 21. Deryni Rising (1970), Deryni Checkmate (1972), and High Deryni (1973), comprising Chronicles of the Deryni, appeared in the BAFS; Camber of Culdi (1976) and Saint Camber (1978) followed, with some dozen more volumes from the early 1980s to the present. 22. Which, despite the Welsh nomenclature and vague historical parallels, is perhaps even less “Welsh” than Alexander’s Prydain.

232

NOTES

23. The first-page blurb of Daley’s work includes an endorsement by Poul Anderson: “A highly entertaining fantasy in the manner of the old Unknown Worlds” (see Daley). Dickson’s is in fact an expansion of a story published in 1957 in The Magazine of Fantasy and Science Fiction. 24. A prototype of this direction from the BAFS would be Munn’s Merlin’s Ring, considerably longer than its two predecessors combined (which Ballantine published as Merlin’s Godson in 1976). 25. Who, in addition to being favored cover artists for Ballantine during this period, were also the artists of the very popular, “official” Tolkien calendars. 26. Comprising Lord Foul’s Bane, The Illearth War, and The Power That Preserves. 27. Some of which are less specifically Tolkien derivative than Brooks and Donaldson. 28. A reviewer quoted for the blurb of Guy Gavriel Kay’s The Summer Tree (1985) wrote that Kay was “borrowing the feel of Tolkien’s original trilogy, but only in the same manner that Tolkien borrowed the feel of the eddas, epics, and romances that preceded Lord of the Rings” (see Kay). 29. The same can be said of William Morris, Cabell, Eddison, and so on.

Index

Ace Books, 2, 3, 178, 186, 187, 201 action adventure fiction, 34, 36, 168 Adams, Richard, 193, 230 Addison, Richard, 51 AE (George William Russell), 141 Alexander, Lloyd, 131; Prydain Chronicles, 193, 230, 231 Amadis of Gaul, 39–42, 59, 92, 210, 211; Munday translation, 211; Rose translation, 215 “ancient fantasy,” 29, 143, 207 Andersen, Hans Christian, 107–8, 110, 111, 129, 219, 221 Anderson, Douglas, 143, 226 Anderson, Poul, 167, 169, 183–85, 201, 205, 226, 228, 232; The Broken Sword, 27, 29, 183–84, 185, 192, 209, 228; Hrolf Kraki’s Saga, 12, 185, 201; Three Hearts and Three Lions, 17, 29, 184 Anglo-Saxon Chronicle, 28, 205 antiquarian(ism), 29, 31, 37, 70, 121, 122, 128, 129, 132, 214, 218 Arabian Nights. See Thousand and One Nights Argonautica (Apollonios Rhodios), 35 Ariosto, Ludovico, 29, 42, 55, 57, 208, 211, 214; Orlando Furioso, 29, 58, 180, 209, 214, 215 Aristotle, 22 Arkham House, 13, 178, 181, 229 Arnold, Edwin Lester, 174 Arnold, Matthew, 38; “Balder Dead,” 221; “Sohrab and Rustum,” 216, 221; “Tristram and Iseult,” 221

Arthurian literature (legend, tradition, etc.), 26, 27, 29, 36, 38, 41, 75, 214; children’s retellings, 95, 113; lack of currency in the eighteenth century, 59–60; medieval Arthurian Romance, x, 210, 212; in modern poetry, 73, 91, 97–99, 105, 127–28; romantic and Victorian editions of medieval works, 70, 92. See also Bradley, Marion Zimmer; Bulwer-Lytton, Edward; Housman, Clemence; Lewis, C. S.; Masefield, John; Morris, William; Peacock, Thomas Love; Robinson, Edward Arlington; Scott, Walter; Swinburne, Algernon Charles; Tennyson, Alfred Lord; Tolkien, J. R. R.; Twain, Mark; Warton, Thomas; White, T. H.; Williams, Charles; Wordsworth, William Ashley, Mike, 86, 217, 222 Asimov, Isaac, 178 Astounding, 178, 205 Atlantis, 36, 174 Auerbach, Nina, and U. C. Knoepflmacher, 221 Augustans, 49, 57, 69 Austen, Jane, 47, 151; Pride and Prejudice, x BAFS. See Ballantine Adult Fantasy Series Ball, Clifford, 177 Ballantine, Betty and Ian, 3, 202

234

INDEX

Ballantine Adult Fantasy Series, 1–2, 3–7, 10, 12, 13, 16, 19, 23, 43, 48, 49, 53, 54, 87, 92, 104, 108, 111, 112, 114, 115, 117, 121, 122, 123, 124, 127, 131, 135, 138, 139, 143, 146, 148, 149, 151, 152, 153, 154, 155, 168, 170, 174, 177, 178, 179, 183, 184–85, 186, 187, 191, 194, 195, 196, 198, 201, 202, 203, 204, 205, 207, 213, 214, 217, 223, 225, 226, 227, 228, 229, 231, 232; BAFS canon, ix, x, 4–5, 32, 33, 43, 53, 74, 81, 91, 99, 104, 111, 114, 127, 129, 160, 167, 185, 191, 198, 204; BAFS template, 4, 12, 13, 14, 15, 16, 19, 28, 34, 85, 87, 91, 113, 115, 117, 123, 130, 131, 134, 135, 137, 138, 139, 140, 142, 152, 153, 154, 155, 158, 169, 170, 171, 174, 175, 179, 180, 181, 186, 191, 198, 203, 205, 224, 228 Barfield, Owen: The Silver Trumpet, 225 Barrie, James: Peter Pan, 129–30 Barringer, Leslie, 227 Basile, Giambattista, 208, 221 Baudelaire, 171, 229 Baum, L. Frank, 129, 225; Oz, 110, 222, 225 Baynes, Pauline, 225 Beagle, Peter S.: The Last Unicorn, 3, 6, 192–93 beast fable, 21, 22, 130, 225 Beattie, James, 216: The Minstrel, 55–56, 78 Beckford, William, 77, 113, 171, 185, 213, 229; Episodes of Vathek, 53–54, 113, 213; Vathek, 32, 53–54, 75, 76, 104, 113, 114, 122, 171, 209, 223 Beddoes, Thomas Lovell, 81; “Pygmalion,” 216 Belloc, Hilaire, 147 Beowulf, 23, 44, 96, 97, 101, 121, 122, 133, 157, 168, 210, 214

Beyond, 181 Bhagavad Gita / Bhagvat Geeta, 65, 94 Bible, 30, 31, 33, 208. See also King James Bible Bierce, Ambrose, 229 Blackmore, Richard: Arthurian epics, 57 Blackwood, Algernon, 169 Blake, William, 65–66, 80, 116, 121, 212; Poetical Sketches, 66; Prophetic Books, 48, 65, 66, 67, 116 Blaylock, James, 207, 212 Blish, James, 230 Boccaccio, Giovanni, 80 Boiardo, Matteo, 42; Orlando Innamorato Berni’s revision Rose translation, 215 Bok, Hannes, 179, 205, 229 Boyer, Robert H., and Kenneth J. Zahorski, 11–12, 143, 203, 204, 208, 217, 227 Bradbury, Ray, 177; Timeless Stories for Today and Tomorrow (ed.), 10 Bradley, Marion Zimmer, 44, 212, 230 Bramah (Smith), Ernest, 77, 132, 146–48; Kai Lung tales, 12, 32, 114, 146–48, 185, 226 Brooks, Terry, 146, 196–98, 232; The Sword of Shannara, 196–98, 203 Brothers Hildebrandt, 6, 196 Browne, Fanny, 107 Browne, Maggie, 111 Browne, Thomas, 79, 80 Browning, Robert, 207; “Childe Roland to the Dark Tower Came,” 219 Bryant, Jacob, 65, 219 Bulfinch, Thomas, 35, 176, 219 Bulwer-Lytton, Edward, 37, 217; King Arthur: An Epic, 219; Last Days of Pompeii, 173; short fiction, 84, 85 Bunyan, John, 78; Pilgrim’s Progress, 42 Burgess, Anthony, 8

INDEX

Burroughs, Edgar Rice, 167, 174, 175, 205; Barsoom/John Carter, 174, 175, 229; Tarzan, 174, 175 Burton, Richard, 171, 209 Byron, Lady, 223 Byron, Lord, 69, 78; Don Juan, 76, 78 Cabell, James Branch, 1, 4, 5, 13, 27, 32, 33, 36, 41, 52, 63, 77, 84, 113, 121, 132, 133, 134, 143–45, 148, 151, 153, 155, 159, 160, 161, 163, 168, 175, 178, 185, 187, 202, 203, 205, 209, 210, 220, 226, 227, 228, 232; Biography of Manuel, 41, 143–45, 209, 226; Cream of the Jest, 143, 226; Domnei, 27, 161, 209, 226; Figures of Earth, 27, 33, 144–45, 161, 167, 209, 226, 227–28; High Place, 226; Jurgen, 27, 145, 153, 161, 168, 209; Lineage of Lichfield, 148, 226; Music from behind the Moon, 145, 161, 226; Silver Stallion, 16, 33, 226; Something about Eve, 226; Way of Ecben, 161, 226 Calvino, Italo, 210 Camelot, 154 Campbell, John F., 94 Campbell, John W., 178–79, 205, 229 Campbell, Joseph, 65 Carlyle, Thomas, 217; German Romance (tr.), 86 Carroll, Lewis (Charles Dodgson), 5, 222; Alice Books, 111, 115, 222; Sylvie and Bruno, 111 Carter, Angela, 212 Carter, Lin, ix, 3–5, 8, 9, 10, 11, 12, 15, 23, 24, 25, 28, 29, 32, 34, 112, 115–16, 121, 127, 135, 138, 139, 143, 149, 154, 155, 160, 183, 186, 187, 192, 197, 201, 202, 207, 223, 224, 225, 227, 228, 229, 230; Imaginary Worlds, 7, 16, 117, 222, 229

235

Cavendish, Margaret, 211 Caxton, William, 59, 121, 129, 138, 211–12, 221 Celtic Fantasy, 67, 99, 132, 139, 193 Celtic myth (legend, tradition, etc.), 31, 32, 33, 57, 58, 208; Gaelic/Irish translations, 93–94, 129; lack of currency in the eighteenth century, 60–61; 67; Romantic editions/ translations, 70; translations of Welsh poetry, 60, 93, 217. See also Ferguson, Samuel; Gregory, Lady; Guest, Charlotte; Joyce, P. W.; Mabinogion; MacPherson, James; Morris, Kenneth; Peacock, Thomas Love; Stephens, James; Tain bo Cualgne; Walton, Evangeline Celtic Revival, 134ff. Cervantes, Miguel de, 34, 78; Don Quixote, 136, 211 Chanson de Geste, 27 Chant, Joy, 6, 195, 231; Red Moon and Black Mountain, 194, 195, 201 Chapman, George, 80, 216 Chatterton, Thomas, 64–65, 214 Chaucer, Geoffrey, 57–58, 100, 102, 221 Cheever, John, 204 Chester, Thomas, 219 Chesterton, G. K., 8 Child, Francis James, 92, 214 “children’s fantasy,” 129–31, 202, 231 Chretien de Troyes116, 144, 211 Clute, John, and John Grant, 5, 8, 84, 139, 161, 209, 216, 217, 222, 226 Coates, Robert M., 204 Coe, R. H., 221 Coleridge, John Duke, 86, 217 Coleridge, Samuel Taylor, 22, 69, 71, 77, 85, 115, 121, 122; “Christabel,” 72, 73, 75, 105; “Kubla Khan,” 75; “Rime of the Ancient Mariner,” 38, 71, 72, 81

236

INDEX

Coleridge, Sara, 198; Phantasmion, 37, 69, 70, 86–87, 91, 224 Collins, William, 55, 62, 75, 80 Congreve, William, 54 contes de fees, 25, 43, 85, 107, 108, 110, 129, 208, 219 Cooper, James Fenimore, 173 Cooper, Susan, 194, 231 Corley, Donald, 32, 163, 227 Cott, Jonathan, 218 Craik, Dinah Mulock, 110, 114; Alice Learmont, 72, 110 Crawford, F. Marion: Khaled, 32, 114 Crawford, John Martin (translator of The Kalevala), 218 Croker, T. Crofton, 213 Cruikshank, George, 221 Daley, Brian, 196, 232 Dante, 22, 80, 116 dark fantasy (Lovecraft and Smith), 169–72 Darley, George, 81; Nepenthe, 116 Datlow, Ellen, and Terri Windling, 25 D’Aulnoy, Madame, 107 Dean, Pamela: Tam Lin, 72 de Camp, L. Sprague, ix, 3, 4, 5, 7, 9, 11, 12, 162, 167, 169, 179, 181, 187, 192, 201, 202, 205, 209, 229; Conan “collaborations” and spin-offs, 181, 183, 192; Literary Swordsmen and Sorcerers, 8, 222. See also Nyberg, Bjorn; Pratt, Fletcher, and L. Sprague de Camp Defoe, Daniel, 47, 212 de France, Marie, 219 deGatno, Paul J., 63 de la Mare, Walter: The Three MullaMulgars, 225 del Rey, Judy-Lynn, 6, 196 del Rey, Lester, 6, 196 de Morgan, Mary, 107 Derleth, August, 181, 229 de Worde, Wynkyn, 59, 211, 212 Dickens, Charles, 47, 48, 154, 223

Dickinson, Emily, 105 Dickson, Gordon R., 196, 232 Disraeli, Benjamin, 37, 83–84, 85, 217 Donaldson, Stephen R., 196–97, 198, 203, 232; First Chronicles of Thomas Covenant, 196–97, 203 Doyle, Arthur Conan, 174 Drake, David, 187 Dryden, John, 83, 213; “To Oedipus,” 49–50, 51, 80 Dumas, Alexander, 173 Dunbar, William, 150 Dungeons and Dragons, 10 Dunn, Joseph, 225 Dunsany, Lord, 1, 4, 5, 7, 8, 13, 14, 16, 17, 23, 32, 33, 66, 67, 77, 81, 113, 121, 127, 132, 133, 134–39, 140, 141, 142, 143, 144, 145, 148, 151, 153, 155, 156, 159, 160, 163, 168, 169, 187, 198, 202, 203, 205, 208, 210, 225, 226, 227, 228, 229, 230; Blessing of Pan, 137; Book of Wonder, 16, 136, 228; Charwoman’s Shadow, 17, 24, 136, 137, 139, 203; Curse of the Wise Woman, 134, 137; Don Rodriguez, 12, 136, 137, 147, 167; A Dreamer’s Tales, 135, 136; Fifty-One Tales, 136; Gods of Pegana, 16, 135, 161; influence on literary writers, 160–62; influence on Lovecraft, 170; influence on Smith, 171; King of Elfland’s Daughter, 11, 136–37, 139, 151, 203; Last Book of Wonder, 136; Sword of Welleran and Other Stories, 135; Tales of Three Hemispheres, 135, 136, 225; Time and the Gods, 16, 135 Early English Text Society (Oxford University Press), 218 Eddison, E. R., 1, 3, 4, 5, 7, 8, 9, 13, 17, 18, 23, 25, 26, 31, 32, 33, 34, 51, 65, 66, 77, 81, 113, 121, 124, 127,

INDEX

132, 133, 143, 148–51, 152, 154, 155, 158, 159, 160, 161, 168, 176, 178, 181, 182, 187, 198, 202, 203, 205, 210, 227, 228, 232; Egil’s Saga (tr.), 132, 150, 210; A Fish Dinner in Memison, 18, 150; Mezentian Gate, 18, 24, 36, 150, 227; Mistress of Mistresses, 150; Styrbiorn the Strong, 149–50, 226, 228; Worm Ouroboros, 16, 17, 26, 27, 104, 114, 117, 148–49, 150–51, 156, 162–63, 168, 176, 203, 226, 229; Zimiamvia series, 16, 18, 145, 149, 150–51, 155, 163, 164, 182, 203 Eliade, Mircea, 65 Eliot, George, 113, 210 Eliot, T. S., 128, 224; The Waste Land, 23. See also modernism epic fantasy, 12, 204 Erdman, David, 48 Evans, Evan, 61 Ewing, Juliana Horatia, 221 Faletra, Michael, 224 Faulkner, Peter, 120 Faulkner, William, 48, 144 Ferdowsi. See Shah-Nameh Ferguson, Samuel, 99, 100, 101, 104, 220; Lays of the Red Branch, 38 Fielding, Henry, 47; Tragedy of Tragedies, 57 Fischer, Otto, 177 Flaubert, 171 Fletcher, Giles, 211 Fletcher, Phinneus, 211 Flieger, Verlyn, 157 Fouqué, de la Motte, 84, 108, 115, 221, 223; Magic Ring, 86, 123; Undine, 86 Four Sonnes of Aymon, 59, 211 France, Anatole, 207, 210 Frazer, James, 65, 219 Frye, Northrup, 48, 65 Fussell, Paul, 121

237

Gaelic traditional literature. See Celtic myth (legend, tradition, etc.) Galland, Antoine, 32, 51, 213 Gallardo, Gervasio, 196 Ganguli, K. M., 94 Garner, Alan, 194, 204, 231 Garnett, David, 8 Garnett, Richard, 84; Twilight of the Gods, 114–15, 146, 217, 223 Geoffrey of Monmouth, 27, 57, 157, 205, 212, 227; Historia Regum Britannia, 157, 218 German Romantic fairy tales, 43, 70, 85–86, 108 Giles, J. A., 218 Gilgamesh, 23, 28, 176 Gillies, Robert Pease, 86 Gnome Press, 178, 180, 181, 183 Goethe, Johann Wolfgang, 63, 84, 85, 86 Gold, H. L., 179, 204 Golding, William, 8 Gothic novel/fiction, 34, 35, 48, 53, 54, 72, 74, 169, 205, 222 Graham, Kenneth: author, 130; editor, 213 Grant, John. See Clute, John, and John Grant Gray, Thomas, 55, 58, 61, 66, 70, 75, 80; “The Bard,” 61–62; “Progress of Poesy,” 50–51, 57, 61 Greene, Robert, 212 Gregory, Lady, 134, 225 Grimm Brothers, 25, 26, 27, 85, 87, 95, 103, 107, 110, 129, 219 Guerber, H. A., 132, 176, 225 Guest, Charlotte, 93, 141 Haggard, H. Rider, 14, 149, 173–74, 178, 197, 205, 209, 229; Allan Quatermain, 175, 229; Ayesha: The Return of She, 175; Cleopatra, 174; King Solomon’s Mines, 173, 229; Saga of Eric Brighteyes, 174, 226, 229; She, 173, 175; Wisdom’s Daughter, 173, 174

238

INDEX

Hapgood, Isabel Florence, 218 Harper’s, 13 Harry Potter. See Rowling, J. K. Hawkesworth, John, 54; Almoran and Hamet, 52–53 Hawthorne, Nathaniel, 9; House of Seven Gables, 27; New England gothic, 170, 171 Haywood, Eliza, 77, 198; Adventures of Eovaai, 51–52, 67, 69, 217, 224 Hearn, Michael Patrick, 218 Heimskringla, 149, 162 Heinlein, Robert, 178 Heliodorus, 42 Hemingway, Ernest, 1 Henley, Samuel, 53, 76 heroic fantasy, 2, 7, 12, 186, 201 high fantasy, 12, 204 Hodgson, William Hope, 146, 169, 228 Hoffmann, E. T. A., 25, 34, 84, 85, 86, 108, 115, 204, 209, 217, 221, 223 Hogg, James, 69, 71, 74–75, 77, 84, 139, 215, 216–17; Mountain Bard, 215; Pilgrims of the Sun, 75; Queen Hynde, 75; Queen’s Wake, 74–75, 102, 105; short fiction, 83; Three Perils of Man, 37, 82–83, 84, 216–17 Holdstock, Robert, 204 Homer, 28, 31, 50, 76, 79, 96, 150, 176, 214; Odyssey, 23, 122, 209 Hood, Tom, 109 Hoole, John, 214 Hooper, Walter, 7 Horace, 49 horror fiction, 34, 48, 74, 168, 169, 171, 172, 186, 198, 205 Housman, Clemence, 146, 220, 226 Howard, Robert E., 4, 5, 7, 13, 14, 41, 127, 167, 169, 172–77, 178, 180, 181, 187, 202, 203, 205, 228; Conan/Hyborian Age, 35, 39, 172–73, 181, 183, 187 Hull, Eleanor, 225

Huon of Bordeaux, 59, 92, 122, 211 Hyne, C. J. Cutliffe, 174, 175, 209 Icelandic sagas. See Scandinavian/ Germanic traditional literature Ingelow, Jean, 111 Irish Renaissance, 134ff. Irish traditional literature. See Celtic myth (legend, tradition, etc.) Irwin, W. R., 8–9, 10; Game of the Impossible, 204 Jackson, Rosemary, 9, 10, 204 Jesus Christ Superstar, 208 Johnson, Samuel, 63, 212 Jones, William, 65 Jordan, Robert, 16, 197 Joyce, James, 23 Joyce, P. W., 94 Kafka, Franz, 9 Kalevala, 23, 29, 94, 101, 157, 162, 181, 208, 210, 218, 227 Kalidasa, 65 Kangalaski, Jaan, 196 Kay, Guy Gavriel, 16, 232 Kazantzakis, Nikos, 44 Keats, John, 38, 69, 72, 77, 79–81, 83, 84, 86, 91, 97, 102, 121, 122, 133, 169, 216, 220; Endymion, 79, 81; “Fall of Hyperion,” 79, 80; “On First Looking into Chapman’s Homer,” 79; “Hyperion,” 79, 80; “La Belle Dame Sans Merci,” 72, 80; “Lamia,” 38, 79, 81, 105 Keightley, Thomas, 207 Kelmscott Press, 121, 122, 123, 212 Kendall, Carol, 131, 193 Kennedy, Patrick, 94 King Arthur. See Arthurian literature (legend, tradition, etc.) King James Bible, 14, 33, 133, 135, 138, 170, 171, 208 Kingsley, Charles, 8; The Heroes, 219; Water Babies, 18, 110, 203, 222

INDEX

Kipling, Rudyard, 130 Kirby, W. F. (translator of The Kalevala), 218 Kirchhoff, Frederick, 224 Kurtz, Katherine, 16; Deryni books, 6, 187, 195, 201, 231 Kuttner, Henry, 177 Lancer Books, 3, 6, 177, 183, 186, 201, 202 Landor, Walter Savage, 75–76 Lang, Andrew, 26, 109, 209, 219, 222, 229; Blue Fairy Book, 22, 25, 219; Gold of Fairnilee, 110; Pantouflia Chronicles (Prince Prigio; Prince Ricardo; Tales of a Fairy Court), 21, 109, 112, 203; Red Fairy Book, 113, 134 Langland, William, 116 Leahy, A. H., 225 Lee, Tanith, 195 Le Guin, Ursula K., 131, 146, 160–61, 195, 231; Earthsea books, 16, 158, 193, 194, 202, 207, 222, 230; “From Elfland to Poughkeepsie,” 143, 187 Leiber, Fritz, 4, 5, 13, 14, 32, 127, 167, 169, 177–78, 181, 183, 185, 187, 192, 201, 205, 229, 230; Fafhrd and the Gray Mouser stories, 177–78, 181, 187, 229 Lewis, C. S., 5, 8, 17, 25, 28, 53, 113, 128, 133, 159, 160, 202, 224, 225, 228; Dymer, 128, 159; The Lion, the Witch, and the Wardrobe, 224; Narnia series, 1, 109, 110, 130, 131, 151, 159, 193, 202, 204, 206, 222, 225; narrative poems, 128, 159, 224; “On Science Fiction,” 7, 104, 159; Out of the Silent Planet, 159, 203; Perelandra, 159, 203; Pilgrim’s Regress, 27, 159; That Hideous Strength, 24, 159, 207; Till We Have Faces, 159; Voyage of the Dawn Treader, 104

239

Lewis, Matthew, 48 Lewis, Sinclair, 145 Life of Brian, 208 “Lilliputian story,” 21, 25. See also Swift, Jonathan Lindsay, David, 3, 66; A Voyage to Arcturus, 3, 117, 152, 159 Linklater, Eric, 154, 155, 227 literary fantasy (canonical), 35–36, 37, 39, 127–64, 192, 205 Lockhardt, J. G., 214 Lodge, Thomas, 216 Lofting, Hugh, 130 Longfellow, Henry Wadsworth, 30, 100–101, 104, 106, 220, 221; Song of Hiawatha, 38, 100–101, 221; Tales of a Wayside Inn, 100, 102 Longus, 42 Lönnrot, Elias. See Kalevala Lovecraft, H. P., 13, 14, 16, 53, 167, 168, 169–70, 171–72, 173, 177, 179, 181, 203, 204, 205, 228; Cthulhu Mythos, 171, 205, 228; Dream-Quest of Unknown Kadath, 169, 229; “Dunsanian”/ Dreamlands fiction, 169–70, 181, 205 Lowell, James Russell, 100; “Vision of Launfal,” 219–20 low fantasy, 12, 204 Lydgate, John, 221 Lyly, John, 42, 52 Lyrical Ballads, 47, 48, 71, 72 Mabinogion, 12, 28, 29, 30, 44, 63, 93, 96, 141, 143, 168, 210, 217, 218. See also Guest, Charlotte; Morris, Kenneth; Walton, Evangeline MacDonald, George, 5, 8, 25, 26, 66, 74, 84, 87, 110, 112, 115–17, 131, 152, 202, 222, 223; At the Back of the North Wind, 110; Curdie books (The Princess and the Goblin; The Princess and Curdie), 109, 110, 117, 202, 222;

240

INDEX

MacDonald, George, (continued) “Fantastic Imagination,” 24, 206, 222; “Giant’s Heart,” 203; “Golden Key,” 85, 109, 203, 230; Lilith, 27, 109, 116–17, 152, 202, 203, 204, 208, 223; Phantastes, 18, 27, 37, 72, 85, 86, 87, 109, 115–16, 123, 202, 204, 206–7, 216, 217, 223; short fairy tales, 108, 222 Machen, Arthur, 169, 228 Mack, Robert, 54 MacPherson, James, 38, 42, 62–64, 67, 139; Ossian, 74, 99, 122, 169, 214 Madden, Frederic, 70, 92 Magazine of Fantasy and Science Fiction, 180–81, 184, 204, 232 Magnusson, Eirikr, 93, 96, 103, 122 Mahabharata, 23, 94 Malory, Thomas, 30, 34, 57, 83, 92, 97, 106, 116, 129, 133, 146, 153, 168, 208, 212, 220; Le Morte d’Arthur, 39, 59, 70, 97, 210, 211, 212; Winchester manuscript, 129 Manabozho, 101, 106, 220 Manguel, Alberto, 11 Manley, Delariviere, 52 Manlove, C. N., 15, 22, 28, 206–7, 222; Impulse of Fantasy Literature, 8, 25–26; Modern Fantasy: Five Studies, 8, 18–19, 20, 162 Mann, Thomas, 44 märchen. See German Romantic fairy tales Masefield, John, 128 Mathews, Richard: Fantasy: The Liberation of Imagination, ix, 8, 23, 58, 222 Maturin, Charles, 9; Melmoth the Wanderer, 82 McKillip, Patricia, 195, 231; The Forgotten Beasts of Eld, 194, 202 McKinley, Robin, 25, 212 medieval romance, 27, 33, 40, 55, 58, 121, 122, 133, 138; lack of

currency in eighteenth century, 59–60 Melling, O. R., 44, 212 Mencken, H. L., 145 Meredith, George, 53, 223; Farina, 113; Shaving of Shagpat, 32, 54, 113–14, 147, 223 Merriman, James Douglas, 74 Merritt, A., 5, 167, 174, 175, 202; Ship of Ishtar, 131, 174, 179 Merwin, Sam, 185 Milne, A. A., 131, 225 Milton, John, 57, 80, 97, 208, 210 Mirrlees, Hope, 205; Lud-in-the-Mist, 151–52, 167, 203, 219 Mitchell, Margaret: Gone with the Wind, 27, 28 Mitchison, Naomi, 154, 155 Mobley, Jane, 11 modernism, 49, 128, 210, 224, 228 Montalvo, Garci Rodriguez de, 40, 210, 211. See also Amadis of Gaul Moorcock, Michael, 9, 210, 230; Elric of Melniboe stories, 192; Wizardry and Wild Romance, 8, 39–40, 222 Moore, Catherine L., 177 Moore, Thomas, 77; Lalla Rookh, 38, 76, 77, 101 Morris, Kenneth, 29, 30, 31, 32, 33, 38, 63, 64, 81, 84, 113, 115, 132, 134, 141–43, 144, 148, 152, 161, 163, 187, 204, 207, 210, 226, 228; Book of the Three Dragons, 1, 36, 63, 141–42, 143; Chalchiuhite Dragon, 142, 143; Fates of the Princes of Dyfed, 12, 27, 36, 141–42, 143, 155; Secret Mountain and Other Tales, 142, 161 Morris, May, 27 Morris, William, 5, 7, 8, 9, 13, 15, 17, 19, 23, 27, 31, 32, 34, 35, 36, 37, 38, 39, 41, 43, 49, 51, 63, 65, 66, 72, 73, 74, 77, 81, 82, 84, 86, 87, 91, 93, 95, 96, 101–4, 113, 114, 117–24, 127, 132, 133, 134, 135,

INDEX

138, 140, 142, 146, 148, 152, 154, 158, 160, 162, 163, 169, 176, 181, 182, 187, 197, 198, 202, 203, 205, 207, 208, 209, 212, 215, 221, 223, 224, 228, 229, 232; Child Christopher and Goldilind the Fair, 119–20, 123, 176, 206, 210; Defense of Guenevere and Other Poems, 101–2, 221; early short tales, 122–23; Earthly Paradise, 102–3, 104, 105, 106, 123, 162, 210, 221; Glittering Plain, 118–19, 123; House of the Wolfings, 104, 117–18, 122, 123, 209; Life and Death of Jason, 102, 104, 105, 123, 210; Roots of the Mountains, 117–18, 123, 209; Story of Sigurd the Volsung and the Fall of the Niblungs, 103, 123, 134, 210; Sundering Flood, 16, 36, 120, 209, 224; translations, 93, 103, 122, 133, 134; Water of the Wondrous Isles, 86, 120, 122, 123, 210; Well at the World’s End, 86, 117, 119–20, 121, 122, 148, 162, 182, 203, 208, 209; Wood Beyond the World, 4, 11, 17, 86, 87, 119–20, 210 Muller, Max, 65, 219 Mundy, Talbot, 174, 178 Munn, H. Warner, 6, 177, 192, 194, 203, 209, 230, 232 Musäus, Johann Karl August, 85, 86 Narayan, R. K., 44 Nesbit, E., 130, 131, 194, 204, 225 Newbolt, Sir Henry, 164; Aladore, 72, 146, 162 Newcastle Forgotten Fantasy Library, 5, 143, 146, 152, 174, 231 Nibelungenlied, 93, 149, 157, 218 Norse myth and legend. See Scandinavian/Germanic traditional literature

241

Novalis, 25, 85, 108, 115, 223 Nyberg, Bjorn, 181, 183 Odyssey. See Homer Orlando Furioso. See Ariosto, Ludovico Ossian. See MacPherson, James Ovid: Metamorphoses, 30 Palmerin of England, 39, 40, 41, 59, 210, 211; Munday’s translation, 211 Parnell, Thomas: “Fairy Tale,” 75, 213 Peacock, Thomas Love, 69, 113, 217; Maid Marian, 83; Misfortunes of Elphin, 37, 83, 84 Peake, Mervyn: Gormenghast books, 3, 4, 8, 26, 154–55, 204, 206, 207; Gormenghast, 154; Titus Alone, 154; Titus Groan, 18, 154 Pepper, Bob, 196 Percy, Thomas, 70, 73, 214, 215; Northern Antiquities, 61; Reliques of Ancient English Poetry, 59, 60, 61 Perrault, Charles, 107 Pindar, 49, 51 Poe, Edgar Allan, 34, 35, 169, 170, 171, 185, 207, 209, 229 Poetic Edda, 26, 27, 96, 157, 162, 182, 210, 218 poetic fantasy (eighteenth and nineteenth centuries), 37–38, 71–81, 91, 96–106, 163, 210 Pope, Alexander, 47, 57, 79, 212, 215; “Essay on Criticism,” 50 popular fantasy (canonical), 34–35, 167–87, 205 Powys, John Cowper, 154 Pratt, Fletcher, 4, 13, 17, 124, 167, 169, 181–83, 202, 203, 205, 228, 230; Blue Star, 17, 18, 182–83, 185, 205; Well of the Unicorn, 16, 18, 33, 181–82, 183, 185, 203, 209, 228, 230. See also Pratt, Fletcher, and L. Sprague de Camp

242

INDEX

Pratt, Fletcher, and L. Sprague de Camp, 5, 13, 169, 180, 181, 202, 203, 205, 214, 229; Carnelian Cube, 180, 181; Harold Shea tales, 17, 29, 180, 181, 187, 203, 214, 229; Land of Unreason, 180, 229, 230 pre-Raphaelite medievalism, 82, 115 Pringle, David, 139 Prose Edda, 93, 157, 162, 209, 221 Pyle, Howard, 107, 132, 225; Garden behind the Moon, 111 Pynchon, Thomas, 9 Pyramid Books, 3, 6, 7, 186, 201, 202 quasi-(or pseudo-)Oriental tale, 32, 33, 43, 48, 51–54, 69, 78, 135, 146, 148, 171, 209, 226; in verse, 75–76 Qur’an, 65 Radcliffe, Anne, 49, 72 Raeper, William, 111 Ralston, W. R. S., 218 Ramayana, 23, 44, 176 Reeve, Clara, 213 Renault, Mary, 152 Richardson, Samuel, 47, 106, 212 Ridley, James; Tales of the Genii, 53, 113, 214 Ritchie, Anne Thackeray, 221 Ritson, Joseph, 70, 74, 92, 215 Robinson, Edward Arlington, 127–28 Rose, William Stewart, 215 Rossetti, Christina: Goblin Market, 219 Rowling, J. K., 231; Harry Potter series, 16, 130, 194, 204, 207 Ruskin, John: King of the Golden River, 107, 217, 221 Scandinavian/Germanic traditional literature, 31, 32, 33, 37, 38, 57, 58; lack of currency in the eighteenth century, 60–61; Romantic and Victorian editions/ translations, 70, 92–93; William

Morris translations, 93, 103, 122, 138. See also Anderson, Poul; Arnold, Matthew; Beowulf; Eddison, E. R.; Heimskringla; Morris, William; Nibelungenlied; Percy, Thomas; Poetic Edda; Prose Edda; Thidrek’s Saga; Thorpe, Benjamin; Tolkien, J. R. R.; Volsunga Saga Schiller, Friedrich von, 63, 71 Schoolcraft, Henry, 100, 220 Science Fantasy Magazine, 192 science fiction, 2, 4, 7, 34, 168, 180, 181, 183, 186, 198, 201, 207 Scott, Walter, 69, 71, 72–74, 77, 81, 82, 84, 85, 91, 92, 97, 99, 100, 117–18, 122, 133, 139, 173, 214, 215, 216, 217; Bridal of Triermain, 73, 216; Ivanhoe, 82, 173; Lady of the Lake, 73; Lay of the Last Minstrel, 72–73, 74, 81, 122; Minstrelsy of the Scottish Border, 70, 71, 72, 74, 214, 215 Shah-Nameh, 23, 122, 221 Shakespeare, William, 150; A Midsummer Night’s Dream, 106; Venus and Adonis, 212, 216 Shelley, Mary, 35, 207; Frankenstein, 213 Shelley, Percy Bysshe, 78–79, 84, 115, 121, 122, 216; Alastor, 38, 78, 79, 81, 116 Shenstone, William, 55 Sheridan, Frances, 54; History of Nourjahad, 53 Shippey, Tom, 9–10, 28, 157, 204 Shwartz, Susan, 54 Sidney, Philip, 42; Arcadia, 41, 52, 150, 211 Sidney-Fryer, Donald, 171 Silko, Leslie Marmon, 43–44, 215 Silver, Carol, 224 Sir Gawain and the Green Knight, 26, 27, 28, 208, 209; first modern edition, 70

INDEX

Smith, Clark Ashton, 4, 5, 13, 14, 53, 127, 167, 168, 169, 170–72, 173, 175, 177, 180, 181, 185, 187, 196, 198, 203, 205, 207, 228, 229; Averoigne, 170; Hyperborea, 170, 229; Poseidonis, 170, 229; Xiccarph, 170, 229; Zothique, 32, 170, 171, 172, 185, 229 Socrates, 49 Song of Roland, 27, 39 Sophocles, 49 Southey, Robert, 38, 41, 69, 71, 77, 78, 91, 214, 215; Curse of Kehama, 76, 211; Metrical Tales, 215, 216; Thalaba the Destroyer, 76, 81, 101, 211 Spence, Lewis, 132, 225 Spenser, Edmund, 34, 58, 80, 81, 122, 210, 212, 214; Faerie Queene, 23, 29, 39, 41, 42, 66, 79, 86, 106, 180, 208, 209, 211, 212, 217; influence of Faerie Queene in the eighteenth century, 55–57 Springer, Nancy, 195 Stableford, Brian, 84, 217 Startling Stories, 229 Stephens, James, 29, 31, 33, 38, 84, 132, 134, 139–41, 142, 144, 145, 148, 149, 152, 153, 160, 161, 163, 204, 210, 226, 228; Crock of Gold, 31, 139–40, 41; Deirdre, 140; In the Land of Youth, 140; Irish Fairy Tales, 24, 140 Sterling, George, 171 Sterling, John, 37, 84, 85, 217 Sterne, Laurence, 154 Stevenson, Robert Louis, 73, 229 Stockton, Frank R., 108, 112, 222 Stoker, Bram, 108 Straparola, Giovan Francesco, 208, 221 Stukeley, William, 65 Sturlusson, Snorri. See Heimskringla; Prose Edda Sweet, Darrell, 6, 196

243

Swift, Jonathan, 25, 83, 219; Famous Prediction of Merlin, 57; Gulliver’s Travels, 43, 52 Swinburne, Algernon Charles, 38, 73, 81, 98–99, 104, 121, 122, 128, 220, 221; Tale of Balen, 220; Tristram of Lyonesse, 98–99, 101, 220 Swinfen, Ann, 20–22, 207 Sword and Sorcery, 1, 2, 3, 5, 6, 9, 12, 13, 172–78, 180, 186, 187, 191, 201, 206, 209, 230 syncretism, 33, 77, 81, 94–95, 122, 145, 150, 163 Tain bo Cualgne, 35, 44, 140, 176, 181, 210 “Tamlane” (Tam Lin), 70, 72 Tasso, Torquato, 214 Taylor, Edgar, 85, 107, 219 Tennyson, Alfred Lord, 38, 73, 81, 101, 102, 104, 121, 128, 133, 220; Idylls of the King, 97–99, 101, 104–5 Thackeray, William Makepeace: Rose and the Ring, 109, 112 Thelwell, Michael, 44 Thidrek’s Saga, 157 “Thomas Rhymer,” 70, 72, 146 Thompson, E. P., 224 Thomson, James, 55, 80, 213; Castle of Indolence, 55–56; Liberty, 50 Thorpe, Benjamin, 93 Thousand and One Nights, 32, 33, 41, 44, 48, 51, 65, 75, 76, 78, 94, 95, 113, 147, 209, 213, 223, 225; Dalziel’s Illustrated Arabian Nights Entertainments, 219. See also Burton, Richard; Galland, Antoine; quasi-(or pseudo-) Oriental tale Thrilling Wonder Stories, 185 Tieck, Ludwig, 25, 86 Time Machine. See Wells, H. G. Tolkien, Christopher, 156, 227 Tolkien, J. R. R., ix, 1, 3, 4, 5, 8, 9, 13, 14, 16, 20–22, 27, 31, 32, 33, 36,

244

INDEX

Tolkien, J. R. R., (continued) 37, 38, 39, 41, 43, 51, 52, 63–64, 65, 66, 67, 77, 93, 95, 104, 112, 113, 124, 127, 128, 132, 133, 134, 139, 141, 143, 146, 150, 152, 155–59, 160, 176–77, 181, 182, 186, 187, 191, 192, 193, 194, 196, 197, 198, 202, 203, 205, 207, 208, 214, 215, 219, 220, 221, 224, 225, 227, 229, 230, 232; Book of Lost Tales, 104, 156, 158, 162, 163, 227; Children of Hurin, 227; children’s fiction, 130–31; “Fall of Arthur,” 128, 214; Farmer Giles of Ham, 3, 21, 131, 155, 156, 225; History of Middle-earth, 156–57; Hobbit, 3, 16, 17, 24, 109, 130–31, 155, 159, 193, 202, 225; influence of literary writers on, 161–63; “Lay of the Children of Hurin” and “Lay of Leithian,” 38, 128, 221; Lays of Beleriand, 156, 224; “Leaf by Niggle,” 155, 156; “Legend of Sigurd,” 128; Lord of the Rings, 1, 2, 3, 7, 10, 11, 16, 17, 18, 19, 24, 28, 39, 63, 72, 117, 121, 145, 148, 151, 155, 156, 157, 158, 159, 162–63, 164, 176, 182, 187, 192, 193, 196, 197, 205, 206, 209, 225, 227, 228, 231, 232; Lost Road, 156; “On Fairy-stories,” 7, 20–22, 24–25, 26, 208, 222; Peoples of Middle-earth, 227; Shaping of Middle-earth, 156, 227; Silmarillion, 154, 156; Smith of Wootton Major, 3, 155, 192, 230; Tree and Leaf, 155; Unfinished Tales, 156, 227 traditional narrative forms, 27–28, 31, 36 Tutuola, Amos, 44, 215 Twain, Mark, 145; A Connecticut Yankee in King Arthur’s Court, 220 Tyrwitt, Thomas, 213, 214

Unknown (Worlds), 14, 17, 18, 138, 167, 168, 175, 177–80, 181, 182, 184, 185, 186, 192, 193, 201, 205, 206, 228, 229, 232. See also Campbell, John W. Vance, Jack, 4, 5, 13, 32, 167, 169, 185, 187, 192, 201, 202, 204, 205; “Dying Earth” tales, 172, 185, 192, 201 van Doren Stern, Philip, 10 van Vogt, A. E., 178; Book of Ptath, 179–80 Verne, Jules, 35 Virgil, 75, 76, 96, 97; Aeneid, 30, 122 Vivian, E. Charles, 174 Volsunga Saga, 26, 27, 35, 93, 96, 97, 101, 103, 106, 122, 133, 157, 168, 203, 210, 218; Byock versus Morris/Magnusson translations, 227 Voltaire, 52, 63, 207 Vulgate Cycle (Arthurian), 210 Wagner, Richard, 103 Walpole, Horace, 213, 214; Castle of Otranto, 27, 48, 213, 214 Walton, Evangeline, 29, 33, 132, 141, 152–53, 154, 203, 205; Mabinogion books, 6, 24, 152–53, 201, 203; The Sword Is Forged, 152, 208–9 Warton, Joseph, 58, 75 Warton, Thomas, 42, 55, 57–58, 60, 70, 75, 212, 214, 215; “Grave of King Arthur,” 59 Webster, John, 150 Weird Tales, 13, 14, 167, 168, 169–77, 178, 179, 181, 186, 192, 201, 228, 229, 230. See also Wright, Farnsworth Wells, H. G., 34, 180; Time Machine, 17, 27, 28, 180, 182, 223 Welsh traditional literature. See Celtic myth (legend, tradition, etc.)

INDEX

Wharton, Edith, 1 White, T. H., 29, 30, 33, 34, 73, 128, 132, 140, 141, 146, 152, 153–54, 155, 159, 193, 208, 220, 227; Book of Merlyn, 153, 154; Ill-Made Knight, 153; Once and Future King, 12, 153–54, 155, 164; Sword in the Stone, 153; Witch in the Wood, 153 Whitman, Walt, 105 Whittier, John Greenleaf, 100 Wieland, Christoph Martin, 85 Wilde, Oscar, 108 Wildside Press, 5 Williams, Charles, 26, 128, 207 Windling, Terri, and Mark Alan Arnold, 212

245

Wollheim, Donald, 2, 187 Wordsworth, William, 69, 71, 97; “Egyptian Maid,” 216 Wright, Farnsworth, 167, 201, 228, 229 Wroth, Mary, 211 Wyke-Smith, E. A., 130 Yeats, William Butler, 99, 134, 141, 225 Yolen, Jane, 194, 195, 207, 231; Magic Three of Solatia, 194; Sister Light Sister Dark, 72 Young, Ella, 141 Zahorski. Kenneth J. See Boyer, Robert H., and Kenneth J. Zahorski Zelazny, Roger, 230 Zipes, Jack, 218, 219, 221