Introduction - Volume 1: From Gilgamesh to the End of the Nineteenth Century

Unutterable Horror - A History of Supernatural Fiction - S. T. Joshi 2014

Volume 1: From Gilgamesh to the End of the Nineteenth Century

The study and analysis of the mode of writing that I call supernatural horror is vexed with a multitude of difficulties and paradoxes. I cannot think of any other genre, with the possible exception of the love story (itself a highly nebulous and imprecise construct, since love plays a role in a number of literary modes, including supernatural horror itself), that is defined by an emotion. The genres of the mystery story, the science fiction story, and even the Western are largely formal designations, chiefly constructed around a certain kind of plot or scenario (a puzzle regarding the perpetrator of a crime; the role of science—or, more precisely, some future extrapolation of science or technology—in human life and society) or setting (the American West). What is more, the genre—if indeed it is a genre—is generally designated by those literary exemplars where the twin elements of the supernatural and of horror (which are logically separable) function exclusively or predominantly. There is horror of a sort in Sophocles’ Oedipus Rex or Shakespeare’s Titus Andronicus, but these works are not customarily regarded as works of supernatural horror. Analogously, the supernatural can function in a narrative without the concomitant sensation of horror, although such instances of what is called benign supernaturalism are relatively rare and have generally not been well received by readers and critics, chiefly because of a certain emotive tameness and, more relevantly, because of their excessive closeness to the fairy tale or folktale.

What this means is that the horror story (whether supernatural or not) somewhat untidily encompasses those works that focus on the emotion of fear, largely to the exclusion or minimisation of other elements, emotions, or motifs—specifically a broad portrayal of character or of those human relations where fear of terror does not play a role. That this exclusive or extensive focus on terror substantially accounts for the prejudice directed against the genre by mainstream critics is a matter I shall discuss presently.

Part of the reason why this genre is so fraught with conceptual difficulties is that it may not have become a definite genre until well along in its literary development. If, by general consensus, the first work of supernatural horror—the work that initiated the “Gothic” school of writing—was Horace Walpole’s The Castle of Otranto (1764), it is at least arguable that supernatural writing did not become a concretised genre until about a century and a half later. The great majority of the nineteenth-century writers who either dabbled or specialised in this mode—notably Edgar Allan Poe and Ambrose Bierce—do not appear to have considered themselves “horror writers” in anything like the contemporary sense of the term. It is not merely that they published much fiction that is very far from horror (humour and satire in both instances), or that the supernatural work they did publish did not appear in venues specifically devoted to this kind of writing (there were none such until the establishment of the pulp magazine Weird Tales in 1923); it is that, so far as can be ascertained from their critical and autobiographical writings, they seem not to have envisioned themselves as writing exclusively or even partly in a given mode of writing radically separable from the mode of writing commonly referred to as “mainstream” fiction. The closest that Bierce came to designating his horror tales (whether supernatural or non-supernatural) by some kind of discrete label is when he referred to them in a letter as “tragic” stories (A Much Misunderstood Man 197). I will maintain that the establishment of Weird Tales itself was the definitive beginning of “supernatural horror” as a distinct genre, just as the establishment of Amazing Stories in 1926 canonically introduced the genre of science fiction.

Further difficulties have been caused by a wide confusion—or, perhaps more charitably, a lack of agreement—in the use of terminology to designate the genre. Freud’s essay “Das Unheimliche” (1919) was translated into English as “The Uncanny”—perhaps not the happiest rendition, since the German appears to refer chiefly to phenomena that are unfamiliar or strange. Matters became further confused when Tzvetan Todorov’s treatise Introduction à la littérature fantastique (1970) was translated into English by Richard Howard as The Fantastic: A Structural Approach to a Literary Genre (1973). Howard rendered le fantastique as “the fantastic,” understandably enough; but when Todorov sought to distinguish what he regarded as neighbouring genres or subgenres by the terms l’étrange (what would now be called non-supernatural horror or psychological suspense) and le merveilleux (supernatural horror), Howard rendered these terms as “uncanny” and “marvelous,” respectively, even though the former is very different from what Freud had in mind. Analogously, Rosemary Jackson’s stimulating treatise Fantasy: The Literature of Subversion (1981) uses “fantasy” as an apparent synonym for supernatural horror, even though the former designation now means something very different among writers, readers, and critics working within the genre of supernatural literature.

It may be best if I were simply to lay down my own terminology—one chiefly drawn from a general consensus (so far as I can ascertain it) within the contemporary horror community—and elucidate it accordingly. In my understanding, the broad paradigm within which supernatural horror functions is “imaginative fiction,” which includes three fairly separate and distinct genres—science fiction, fantasy, and supernatural horror. (Whether mystery or suspense fiction should be regarded as a component of imaginative fiction is a matter of dispute; in my judgment, it should not be, even though this genre has significant relations to supernatural fiction.) There is in my view a certain overlap in all the genres mentioned, and they might be schematised as follows:

Some of these overlapping segments have their own designations, although there is not the strictest agreement either as to the actual terms to be used for them or their precise peramaters. The overlap between supernatural horror and crime/suspense is perhaps the most fruitful, and I see it as encompassing at least three different subgenres: the explained supernatural, where the supernatural is suggested but explained away at the end as the product of error, hallucination, madness, or trickery (its exemplars are Ann Radcliffe and Charles Brockden Brown among the Gothic writers, as well as the crude “weird menace” pulps of the 1930s); psychological suspense or dark fantasy, where the scenario is non-supernatural but the portrayal of madness or hallucination is so intense as to generate a quasi-supernatural emotion of horror (prototypical are Robert Bloch’s Psycho and the early novels of Thomas Tryon); and the ambiguous horror tale, where doubt remains to the end whether the supernatural has come into play or whether the events are the product of madness, hallucination, or error. Todorov privileged this last subgenre as representative of what he called “the fantastic,” evidently regarding it as aesthetically superior to either supernatural or non-supernatural horror. But an examination of the entire range of horror literature suggests that the ambiguous horror tale represents a very thin sliver of the field, and that the “hesitation” that Todorov felt was the dominant trait of fantastic writing is, on the whole, merely a literary technique to maintain the reader’s interest until the tale finally resolves (as almost all do) into either supernaturalism or non-supernaturalism. (I may remark incidentally that Todorov and others have apparently been led to their view by their admiration of what they take to be the most distinguished instance of the ambiguous horror tale, Henry James’s The Turn of the Screw, although I shall argue that it is in fact not “ambiguous” in the sense in which it is most frequently assumed to be.) Todorov’s embarrassing suggestion that the rubric of the fantastic can be extended “by temporarily omitting the end of the narrative” (43) shows how far a critic’s perspective lies from that of a creative artist—and how far a theoretician will go to maintain the spurious integrity of a misguided theory. Although Todorov’s schema has been influential (among critics) and has been largely followed by such a work as Terry Heller’s The Delights of Terror: An Aesthetics of the Tale of Terror (1987)—who adds some interesting reflections on the “pleasures” to be gained from reading horror tales, derived in part from Wolfgang Iser’s reader-response theory—I have not found Todorov’s classification of “the fantastic” to be helpful.

The intersection of science fiction and supernatural horror would seem to be paradoxical, since science fiction (like mystery fiction) is a mode manifestly based upon the use of reason, whereas the essence of supernatural horror is the incursion of the irrational into an objectively real setting. L. P. Hartley emphasised this disjunction between the mystery story and the supernatural horror story in terms that apply just as well to science fiction. In speaking of the “extra thrill” that some people need to maintain an interest in life, Hartley states:

Detective-story writers give this thrill by exploiting the resources of the possible; however improbable the happenings in a detective story, they can and must be explained in terms that satisfy the reason. But in a ghost story, where natural laws are dispensed with, the whole point is that the happenings cannot be so explained. A ghost story that is capable of a rational explanation is as much an anomaly as a detective story that isn’t. The one is in revolt against a materialistic conception of the universe, whereas the other depends on it. (vi—viii)

Those final two sentences are a bit problematic, but the general sense of the passage is sound. Science fiction is commonly misunderstood to be fiction utilising some element of science or technology, but in fact it must be based upon some hypothetical advance or development of contemporary science or technology; this is where the “imaginative” element enters in. Nevertheless, in doing so, it must adhere to the “possible” (or, at least, the conceivable), else it will cease to be science fiction. (Early instances of the mode, such as the novels of Jules Verne, where some of the uses of scientific advance seem nowadays pretty fantastic do not affect the overall thrust of the argument, since Verne appears to have regarded his scientific developments as plausible or conceivable.) The fusion of supernatural horror with science fiction occurs prototypically in the work of H. P. Lovecraft and some of his followers, with perhaps an anticipation in the work of William Hope Hodgson and others. There are clear historical reasons why the mingling of these two modes occurred at this time (the early twentieth century), as I shall explain when I study these writers. In the course of their work, it is possible that supernaturalism gets left behind altogether; but since the manifest purpose of these writings was to create a sense of horror, they can be regarded as hybrids.

The relation of supernatural horror and what I call fantasy is perhaps the most problematical—or, at any rate, the one with the fewest examples that come to mind. Fantasy, as now understood, refers to a scenario where the author creates his/her world either out of whole cloth (as in J. R. R. Tolkien’s Middle-Earth) or, at least, creates the metaphysical “rules” of his/her realm, even if it is not markedly different from the objectively real world; and these rules are fundamentally arbitrary (I do not use this term pejoratively) because there is no reason why the particular “rules” have been utilised as opposed to other rules that might be imagined. Much of Lord Dunsany’s work may fall into this mode: the realm of Pegana that he envisioned in his early work is a kind of fusion of Greek myth, Middle Eastern or Asian legendry, and so forth, but it is fundamentally a never-never-land of his own creation that—even though he made some faint efforts to suggest that it existed in the dim prehistory of the earth—has little connexion with the contemporary world as we know it. It is populated by a multitude of gods, demigods, worshippers, and so forth; but there is no intrinsic reason why these particular gods should occupy Pegana as opposed to others that might be imagined. There is considerable terror in some of his tales, but overall the emphasis is on a kind of otherworldly beauty rather than horror. There are very few other instances of this fusion of supernatural horror and fantasy that come to mind. (Fantasy itself can contain elements of horror, but it is not its main emphasis.)

Whether the core emotion or complex of emotions that are the focus of the genre—fear, terror, horror, dread, and so forth—can be adequately distinguished is a matter of debate. Various critics—beginning with Ann Radcliffe, who, as we shall see in Chapter III, tried to distinguish terror and horror—have attempted to do so, but their conclusions have not been universally accepted. I do not believe that such an attempt is worth the effort, because any such distinctions will be arbitrary and will only be of use for a given theoretical framework or critical analysis. Consider Noël Carroll’s occasionally insightful The Philosophy of Horror (1990). Although, to my mind, Carroll errs seriously in failing to recognise the fundamental distinction between supernatural horror and science fiction, and errs still further in maintaining that the crux of the horror tale is a monster (thereby banishing some of the most significant examples of supernatural and non-supernatural horror to some nebulous ground outside the realm of “horror” as he conceives it), he goes on to define the “tale of terror” (15) as the tale of non-supernatural horror and the “tale of dread” (42) as the tale where weird events come to the fore. This latter distinction, he claims, is necessary because “the emotional response they [tales of dread] elicit seems to be quite different from that engendered by art-horror” (42)—a highly dubious assertion, in my judgment, but one that plays a role in his overall analysis. I suppose it could be maintained that I myself am now devising another brace of technical terms for the purpose of my own analysis, and I do not deny that this is so. I have, for example, long borrowed from Lovecraft the terms “weird tale” and “weird fiction” (see my study, The Weird Tale) as a kind of umbrella term for those facets of imaginative fiction encompassing fantasy, supernatural horror, and the overlaps between supernatural horror and crime/suspense.

As for horror, my general sense—and it is not much more than that—from an examination of the use of this term both in reference to real life and in reference to fiction is that, in addition to (and perhaps above and beyond) its suggestion of a perception of fear (stemming either from personal danger or from danger to another) and a feeling of disgust and revulsion, it carries with it the idea of the contemplation of something appalling and dreadful. This last component may, indeed, allow for the genre of horror to exist at all, since the sentiment goes beyond the immediate apprehension of bodily harm (which is fear) and points toward the witnessing of some phenomenon that the human mind, whether perceiving immediate danger or not, both fails to comprehend and finds somehow wrong in a moral or metaphysical sense.

My major emphasis in this book will be on instances of pure supernatural horror, although the overlapping subgenres discussed above will have to be treated on occasion, if only because of the singular merits of some instances of them or because of their significant influence on the development of supernatural horror itself. Nevertheless, I regard the distinction between supernatural and non-supernatural horror so essential to the understanding of this genre that it may need further elucidation.

In order for there to be a discrete genre called supernatural horror, there must be a general understanding—necessarily in flux, and in accordance with the fluctuations of human knowledge—of what constitutes the “natural.” Non-supernatural horror, or psychological horror, evokes a certain kind of horror, but that emotion is so radically different from the emotion we experience when we witness what Lovecraft somewhat flamboyantly called a “malign and particular suspension or defeat of those fixed laws of Nature which are our only safeguard against the assaults of chaos and the daemons of unplumbed space” (S 23) that the two must be considered radically distinct. There is an undeniable sense of fear in witnessing the depradations of a mass-murderer, or even in sensing that the murderer may come after oneself; there is also a sense of fear in witnessing extreme aberrations of the human mind (something we find, for example, in Ramsey Campbell’s magnificent novel of paranoia, The Face That Must Die); but the fear here evoked is not a metaphysical fear, because there is no sense in which our understanding of the universe is jeopardised. But if we were forced to believe in the actual existence of a vampire or a werewolf, our whole conception of the universe would be seen to be fatally erroneous, and this would occur all apart from any terrors evoked by physical mayhem or even by the vagaries of a diseased mind.

It is, therefore, entirely understandable that supernatural horror only came into existence when, by the eighteenth century, science (and human knowledge as a whole) had advanced to the point where certain objects or events could be stated with fair certainty to be impossible or, at best, highly improbable. The ghost, the witch, the vampire, the werewolf, the haunted house—these and other motifs only gained currency as supernatural fiction once they were banished from the realm of fact. Lovecraft is particularly emphatic on this point. Remarking that “the crux of a weird tale is something which could not possibly happen,” he goes on to say:

If any unexpected advance of physics, chemistry, or biology were to indicate the possibility of any phenomena related to the weird tale, that particular set of phenomena would cease to be weird in the ultimate sense because it would become surrounded by a different set of emotions. It would no longer represent imaginative liberation, because it would no longer indicate a suspension or violation of the natural laws against whose universal dominance our fancies rebel. (Selected Letters 3.434)

This distinction between the natural and the supernatural is vital in understanding why fairy tales, folktales, and the like cannot be regarded as components (although they can be regarded as significant anticipations) of supernatural horror: the events related in them are not rigorously distinguished in terms of their metaphysical status, so that what we would regard as supernatural (if occurring in an objectively real setting) is regarded merely as wondrous or awesome or perhaps even just a little out of the ordinary.

Religion is also an important consideration here. The “events” described in most of the world’s scriptures would nowadays be considered “supernatural”—but they were presumably not so considered by their original believers (and, strictly speaking, by their present-day believers, although many non-fundamentalists would not wish to be pinned down on this point), because these events are putatively regarded by each religion as phases of the actual workings of the cosmos. Accordingly, any literary works that rely in religious presuppositions, or that employ religious allegory, cannot be considered instances of supernatural horror. I am not suggesting that all supernatural horror must be rigidly secular in its outlook; we will, indeed, encounter any number of writers who use weird fiction to convey what they believe to be truths about the universe that are in line with some religious worldview or another. Rather, from an historical perspective, literary work must have segregated itself from religious orthodoxy for the supernatural to evoke its full range of emotive effects. (If it is pointed out that many supernatural tales, right down to the present day, make extensive use of demons or of the Devil himself, I can reply that in nearly every instance these figures are used symbolically or metaphorically—in some cases just on this side of allegory.)

It is my contention that supernatural horror is a distinctively metaphysical mode of writing, because it allows writers—whether they are aware of it or not—to confront directly the very nature of entity. No other mode of writing appears to embody this possibility. The universe can, in essence, be refashioned—at least as regards the specific supernatural phenomenon utilised—in consonance with an author’s philosophical conceptions. Let it pass that many authors may not be entirely cognisant of this procedure; the fact that supernatural horror holds out the prospect of doing so is what is significant.

Indeed, I will utter the seeming paradox that, just as supernatural horror was born at a time when the very idea of the supernatural was being banished from science and human thought in general, so the majority of supernatural writers—and readers—do not in fact “believe” (literally) in their supernatural creations. The insertion of a ghost in a narrative does not automatically mean (although it might) that the author ascribes to philosophical dualism. I do not wish to suggest that all or even most authors of supernatural fiction are atheists or agnostics; rather, these authors have chosen entities or events “which could not possibly happen” as a means both of conveying a sense of dread to their readers and for broader aesthetic ends. The distinction with mimetic fiction comes to the fore here. Sinclair Lewis may not have thought that any single individual named George F. Babbitt existed, but the essence of the novel Babbitt, and the chief reason why it is such a stellar piece of social realism, is that such an individual could have existed and could have done the things that Babbitt does in that novel. Lovecraft, for his part, was well aware that there was no such entity as Cthulhu, but the distinction is that Lovecraft knew further that there could not possibly be any such entity.

Supernatural motifs can be—and usually are—used either for purposes of “imaginative liberation” (as Lovecraft pointed out) and/or for symbolic purposes. The supernatural has been found to be a particularly felicitous and forceful way to convey central human concerns in a way that mimetic fiction cannot do—or, at least, cannot always do so vividly. The vampire, to put it crudely, can be a symbol of a human being’s isolation from society in a way that a depiction of the most radical social “outsider” may not be. I do not wish to assert that supernatural fiction is somehow always superior to mimetic fiction in this use of symbolism; but the history of this literary mode suggests that writers of many different stripes have found the supernatural a valuable means to convey sentiments and effects more potently and plangently than can be done in conventional realism. At the same time, there is a danger that the use of supernatural motifs as symbolism can veer into allegory—a process that effects a negation of the supernatural, since the disjunction between an objectively real world (a world that is, in this precise sense, natural) and the incursion of a given supernatural element would cease to be maintained.

The exact relation between author and reader comes into play at this point. It is widely known that many of the leading motifs of supernatural horror—the ghost, the witch, the haunted house—have their origin in the depths of human history or prehistory. There is every reason to believe that these motifs were believed in as literal realities for millennia, and in some cultures (including our own) a certain proportion of individuals may continue to believe in them. Indeed, it is worth noting that, in the West, many of the standard “monsters” of suprenatural fiction—preeminently the vampire, but also the witch, the sorcerer, and even the werewolf—gained a particular potency because they came to be regarded as violations of the norms of Christianity. But in modern cultures (beginning, say, in the eighteenth century), scientific advance has relegated these motifs to the dustbin of intellectual history, where they can inspire at most a vestigial sentiment of quasi-belief. It is this sentiment that supernatural writers depend on: because belief in ghosts or vampires, for example, can never be wholly eliminated from human thought and emotion, the convincing depiction of them in fiction evokes a dual response—terror at the possible existence of such an entity, and supernatural horror because the conscious mind is aware that these entities “could not possibly” exist. If these two responses seem antipodal or paradoxical, it is by design: the evocation of the strange “reality” of the unreal is the secret to the effectiveness of supernatural horror as a literary mode. Freud seems to have come to this conclusion when he noted that “Our analysis of instances of the uncanny has led us back to the old, animistic conception of the universe” (240). He elaborates the point later in his essay, discussing various conceptions such as the return from the dead:

We—or our primitive forefathers—once believed that these possibilities were realities, and were convinced that they actually happened. Nowadays we no longer believe in them, we have surmounted these modes of thought; but we do not feel quite sure of our new beliefs, and the old ones still exist within us ready to seize upon any confirmation. As soon as something actually happens in our lives which seems to confirm the old, discarded beliefs we get a feeling of the uncanny; it is as though we were making a judgement something like this: “So, after all, it is true that one can kill a person by the mere wish!” or, “So the dead do live on and appear on the scene of their former activities!” and so on. (247—48)

Freud may have been a bit sanguine in adding, “Conversely, anyone who has completely and finally rid himself of animistic beliefs will be insensible to this type of the uncanny” (248), since some of the most distinguished practitioners of weird fiction—notably Poe, Bierce, and Lovecraft—were, by all accounts, pretty hard-headed materialists. But it is exactly the secret of their success that they were momentarily able to frighten themselves by the contemplation of things they knew could not be, and were therefore able to convey that fear to their readers.

The literature of supernatural horror has for many years been the subject of literary and cultural prejudice, and perhaps in some circles it continues to be today. There are a multitude of reasons for this, and I think one of the main ones is a surprising inability on the part of mainstream critics to grasp the rhetoric of supernatural fiction. Consider a remark by a distinguished critic, Graham Hough, on D. H. Lawrence’s ghost stories: “Ghosts should be raised in fiction by people who believe in them or by those whose aim is to produce a shudder of the nerves. Lawrence belongs to neither of these classes, and his ghostly visitants only produce effects that in his more vigorous moods would have been achieved through the conflict of character and circumstance” (quoted in Thornton 138). I have already maintained that very few writers of ghost stories actually believe in the literal reality of ghosts (or, at any rate, in the ghosts they have depicted in their tales), and very few aside from hack writers are interested only in producing a shudder. The supernatural can—and, in Lawrence’s work, does—enhance the “conflict of character and circumstance.” Possibly the relative absence, in weird fiction as a whole, of the kind of interpersonal conflict found in mainstream fiction has had something to do with this critical blindness. Many weird writers’ emphasis is elsewhere. It is not that they are uninterested in the portrayal of character or of interpersonal interaction, but that their focus is largely upon the psychology of fear as it affects individuals and groups. And, of course, the obtrusive presence, on bookstore shelves and movie theatres, of flamboyant works of “horror” that are nothing more than excuses for bloodletting and the display of outlandish monsters has worked its harm—although to judge the whole field of weird fiction by these examples would be as fair as to judge mainstream fiction by the examples of Danielle Steel and Dan Brown.

But there are—or have been—broader cultural forces that have led to a denigration of the literature of terror. The chief of them is that, in the Anglo-Saxon world at any rate, the evocation of such an unpleasant emotion as “horror” is regarded as somehow indelicate or even blasphemous. I am not entirely sure that this accusation can be entirely deflected; even L. P. Hartley noted, “Even the most impassioned devotee of the ghost story would admit that the taste for it is slightly abnormal, a survival, perhaps, from adolescence, a disease of deficiency suffered by those whose lives and imaginations do not react satisfactorily to normal experience and require an extra thrill” (vii). Lovecraft put the matter a bit less pejoratively:

Reality is all right enough so far as it goes … The only trouble is that it doesn’t go far enough for a guy with extreme sensitiveness… . It is perfectly true that mild, conventional, and highly respectable people like the average business or professional man can get enough of a kick out of watching the meaningless routine phaemonena of this pimple on the cosmos to warrant their staying alive—but even with them you can see it wears thin now and then, especially in this latest age of standardisation and decreased variety and adventurousness… . (Selected Letters 3.139)

This may be pejorative in a different way, but it leads to the following conclusion:

The real raison d’être of [weird] art is to give one a temporary illusion of emancipation from the galling and intolerable tyranny of time, space, change, and natural law. If we can give ourselves even for rather a brief moment the illusory sense that some law of the ruthless cosmos has been—or could be—invalidated or defeated, we acquire a certain flush of triumphant emancipation comparable in its comforting power to the opiate dreams of religion. Indeed, religion itself is merely a pompous formalisation of fantastic art. Its disadvantage is that it demands an intellectual belief in the impossible, whereas fantastic art does not. (Selected Letters 4.417—18)

It is customary to deride this “illusion of emancipation” as “escapism,” but Lovecraft’s final remark makes clear what nearly every reader of supernatural fiction will acknowledge: readers are well aware that the whole effect is indeed an illusion—albeit a pleasurable one for them. What is more, the use of the supernatural as symbol underscores the obvious point—one that I trust this treatise will confirm—that significant truths about humanity and its place in the cosmos can be conveyed well outside the context of mimetic realism.

If, then, supernatural horror is written for a small coterie of the “sensitive” who don’t find “reality” quite compelling or satisfying, then we are forced to admit that it is a relatively minor literary mode; but that admission does not carry with it the corollary that great work cannot be done in this mode. The long and rich history of supernatural literature, of which this book can treat no more than the most noteworthy exemplars, is a sufficient argument that many distinguished authors have found it tempting to engage in it when conventional modes of realism prove insufficient, and that those relatively few authors who have specialised in it have in some instances produced a body of work that need not fear competition with any other mode of writing. Supernatural horror enjoyed general popularity in only two periods in literary history—the Gothic novel (roughly from 1780 to 1820) and the “horror boom” of the 1970s and 1980s. For the rest of its history, its appeal was indeed restricted to the few, and there is a good case to be made that its greatest aesthetic successes were directly due to that fact.

If the prejudice against the supernatural in literature has waned—and a signal instance of it may be the recent publication of two large volumes of American Fantastic Tales (2009) by the Library of America, the nation’s guide to the American literary canon—then it may have done so in part for adventitious reasons, such as the general increase in attention that all genre literature has gained in recent decades as academic and other critics go slumming in the bogs of “popular culture” as a relief from the stodgy and overworked standard canon. My own view is that the study of the literature of supernatural horror need not be regarded as merely a sociological exercise, but rather that the genre at its best can offer aesthetic rewards as high as any other mode of writing, and that it does so in a manner that could not be achieved in any other mode. To delineate and analyse how weird fiction achieves its effects is a large part of the critical task I have set for myself.