"The Discourse on the Strange," in Historian of the Strange: Pu Songling and the Chinese Classical Tale, Stanford University Press, 1993, pp. 15-42.

The Master did not speak of prodigies, feats of strength, disorder, and gods.

—The Analects of Confucius, 7.21

"Here is that crazy scholar who didn't believe in ghosts and spirits and who presecuted our minions when he was alive." The King of the Ghosts glared irately at the prisoner: "You possess five sound limbs and inborn intelligence—haven't you heard the line 'Abundant are the virtues of ghosts and spirits'? Confucius was a sage, but still he said: 'Revere them but keep your distance from them!' . . . What kind of man are you that you alone say we don't exist?"

—Qu You, New Tales Under the Lamplight

"A literary work is not an object that stands by itself and that offers the same view to each reader in each period." Hans Robert Jauss's now almost-commonplace pronouncement [in Toward an Aesthetic of Reception, 1982] is given strikingly new visual force in the standard edition of Liaozhai's Records of the Strange, Zhang Youhe's collated and annotated version, which amalgamates editions and the writings that circulated with them before the twentieth century. Embedded in a welter of prefaces, colophons, dedicatory verses, interlinear glosses, and interpretive commentaries, and crowned with a new foreword and appendix, this edition encompasses a virtual, though incomplete, history of Liaozhai's interpretation.

This format derives directly from traditional Chinese critical discourse, which was not simply interpretive but interactive as well. There was a snowballing effect: as a book or manuscript circulated, readers recorded their reactions all over its pages, even between the lines. New readers might even treat the comments of their predecessors as part of the book and comment on them accordingly. In this way, the text became the site of an ongoing dialogue not only between the author and his readers but also between generations of readers. A later reader thus finds it increasingly difficult to ignore this organic process of interpretation, to screen out comment from text in reading.

Although the collation of editions, a mainstay of Chinese scholarly activity past and present, has resulted in redactions in which the amount of commentary far exceeds the amount of original text, the unusual volume of writings in Zhang's edition of Liaozhai is unprecedented for a collection of classical tales. It consists of three full-length commentaries, two extensive glossaries, and a mass of prefaces, colophons, and poems. These waves of literary activity attest both to Liaozhai's great popularity and to the continuous printing of new editions. But these writings also reveal a strong underlying need to interpret the work.

This need to interpret Liaozhai is bound up with the problem of the strange posed by the tales. An understanding of what the strange represents and of the importance or value of the strange within Liaozhai is thus tightly intertwined with the history of the book's overall interpretation. This history began even before the collection had been completed. Pu Songling's literary friends wrote two prefaces, several poems, and scattered comments for the manuscript well before it reached its final form in the early 1700's. After the author's death in 1715, additional prefaces and colophons were written as the collection circulated in manuscript for fifty years. The first printed edition was published in 1766 and, not surprisingly, contributed its own influential preface and foreword. The ambitious full-length commentaries written in the first half of the nineteenth century mark another watershed.

The traditional critical discourse on Liaozhai, like that surrounding vernacular fiction and pornography, is on the whole apologetic and defensive; each contribution must justify anew the value of the work to a sometimes implicit, sometimes explicit, hostile interlocutor. An attentive ear thus enables us to detect elements of the negative reception of Liaozhai as well, even if we allow that for rhetorical purposes the defenders of the book might have altered or exaggerated their opponents' arguments.

An examination of the traditional writings surrounding Liaozhai uncovers three major interpretive strategies: (1) legitimating the practice of recording the strange; (2) understanding the work as an allegorical vehicle for serious self-expression; and (3) acknowledging the work as a model of stylistic brilliance and as a great work of fiction. A fourth approach, a conventional moral didacticism, drones softly through the discourse on Liaozhai, but with one or two exceptions, notably in funerary writings about Pu Songling, this argument seems to have been taken for granted as the most obvious line of defense and is rarely elaborated with much vigor. These approaches, all of which appeared well before the twentieth century, have profoundly shaped modern readings of the work.

In providing this selective interpretive survey, I necessarily simplify and impose order on many often contradictory and sketchy arguments. Since previous arguments are often repeated perfunctorily in later writings, I try to trace changes in emphasis rather than note mere inclusion. Finally, I have concentrated on prefaces and colophons rather than on dedicatory verses because prose writings by necessity entail exposition and argument. Dedicatory verses, in contrast, tend to be written in an altogether lighter and more bantering vein, caring more for a witty turn of phrase than for advancing an argument.

The First Wave: Legitimating the Strange

In 1679, Gao Heng (1612-97), an eminent, retired scholar-official from a prominent gentry family in Pu Songling's hometown of Zichuan and a man of eclectic interests in literature and religion, composed the first preface for Liaozhai. Three years later in 1682, Tang Menglai (1627-98), another retired high official, a leading member of the local Zichuan gentry, and a writer of some renown, completed a second preface for the manuscript. The social and literary prestige of these two men ranked among the highest in the community and extended well beyond Shandong provincial circles. As personal friends of the author, who were also featured as informants or even as protagonists in several tales in the collection, their prefaces offer valuable insight into the immediate circle of readers for whom Liaozhai was written and the social and intellectual climate from which the book emerged.

Gao and Tang's prefaces share a similar orientation: both redefine an interest in the strange in morally and intellectually acceptable terms with the aid of precedents from the Confucian classics. A corollary of their effort was to widen the boundaries of the mainstream literary and philosophical tradition to incorporate the more marginal tradition of recording the strange. To this end, they rehearse many arguments that had become almost standard by the seventeenth century in prefaces to collections of strange accounts.

Tang begins by scrutinizing the concept of the strange. He argues that we cannot base our understanding of the strange on our own empirical experience because the latter is far too limited and individual powers of perception vary too greatly. What is commonly deemed strange is based on convention rather than on any identifiable qualities inherent in strangeness; conversely, familiarity blinds us to the potential strangeness that lies before us.

Now, people consider that what they see with their eyes exists, and that what they don't see, doesn't exist. They say, "This is normal," and what suddenly appears and suddenly vanishes amazes them. As for the flourishing and fading of plants, the metamorphoses of insects, which suddenly appear and suddenly vanish, this does not amaze them; only divine dragons amaze them. But the whistling of the wind, which sounds without stimulus, the currents of rivers, which move without agitation—aren't these amazing? But we are accustomed to these and are at peace with them. We are amazed only at wraiths and fox-spirits; we are not amazed at humankind.

Tang's contention that strangeness is a subjective rather than an objective category echoes a late third-century inquiry into the strange, Guo Pu's (276-324) influential neo-Daoist preface to the mysterious Classic of Mountains and Seas (Shanhai jing), an ancient book of geographic marvels. As Guo Pu argued: "We know not why what the world calls strange is strange; we know not why what the world does not call strange is not strange. How is this? Things are not strange in and of themselves—they must wait for me before they can be strange. Thus the strange lies within me—it is not that things are strange" [Chen Hsiao-chie, Shan hai ching, 1985].

Casting the strange as an epistemological problem to refute skeptics had its seeds in the Daoist parables about great and petty understanding in Zhuangzi. Guo Pu declares, in fact, that he took as his point of departure Zhuangzi's dictum: "What human beings know is far less than what they don't know." It is worth recalling here part of the famous dialogue that Zhuangzi uses to illustrate this point. The North Sea lectures the Yellow River:

"You can't discuss the ocean with a well frog—he's limited by the space he lives in. You can't discuss ice with a summer insect—he's bound to a single season. You can't discuss the Way with a cramped scholar—he's shackled by his doctrines. Now you have come out beyond your banks and borders and have seen the great sea—so you realize your own pettiness. From now on it will be possible to talk to you about the Great Principle."

Although elsewhere Zhuangzi draws upon marvels to illustrate these epistemological points, Guo Pu was probably the first Chinese thinker to ask what the strange is and to ponder what makes something strange. His radical conclusion, reached through an elaborate series of double negatives, is that the strange exists only in the perceiver's mind, not in any objective reality, and that therefore "nothing is impossible."

Guo Pu's and Tang Menglai's arguments will seem oddly familiar to a reader who has encountered Montaigne's celebrated essay, "Of Custom, and Not Easily Changing an Accepted Law": "These examples from strange lands are not strange if we consider what we regularly experience: how much habit stupefies our senses." Like Montaigne, who developed this stance after confronting ethnographic accounts of the New World, Guo Pu was responding to the depiction of exotic lands. Not so Tang Menglai, in whose preface the conventional image of the strange is represented by the otherworldly beings in our midst rather than by the inhabitants of distant barbarian lands: "We are amazed only at wraiths and fox-spirits; we are not amazed at humankind."

But Tang's preface to Liaozhai represents another turn in understanding the strange. Although he borrows Guo Pu's neo-Daoist arguments, profound differences exist. Guo Pu was ultimately arguing for the veracity of the places and creatures depicted in the Classic of Mountains and Seas and for its practical use as an omen book and as an encyclopedia of knowledge. Tang is neither confirmin nor denying the factuality of books like Liaozhai; rather, he is contending that unless we allow a greater tolerance for the discussion of things that lie beyond empirical experience and ordinary discourse, "the beginnings and endings of the Way" are in danger of being "obscured to the world." If our curiosity is entirely suppressed, then ignorance will triumph and "what we see becomes less and less and what amazes us becomes greater and greater."

Tang's preface shares some of the concerns voiced in sixteenth- and seventeenth-century prefaces to both strange tales and vernacular fiction. For instance, Jiang Yingk's (1553-1605) comic preface to Tales of Hearsay (Ertan), a collection of strange anecdotes, also admonishes the reader to reconsider what is really strange. Jiang mischievously selects the ear of the title as something that is not amazing because it is too commonplace: "Now an ear measures only one inch in width, twice that in length, and about three inches inside—that's just a couple of inches. And yet it can receive anywhere from a single syllable to millions of words, far too many to count. Now isn't that exceedingly odd? But no one considers it odd."

Similarly, Ling Mengchu's (1580-1644) preface to his first collection of vernacular stories, Slapping the Table in Amazement (Pai'an jingqi) (dated 1628), closely resembles Tang's preface. Both begin with different halves of the same proverb ("To see a camel and call it a hump-backed horse"—Tang; "To the man of little experience, everything is strange"—Ling), and both demonstrate that ordinary experience is far more extraordinary than is commonly recognized. The two men draw different inferences, however; Tang justifies recording otherworldly beings ("wraiths, fox-spirits, and prodigies"), whereas Ling advocates depicting "the wonders before our very eyes," by which he seems to mean the curiosities to be found in daily life.

Tang insists that accounts of the strange should not be dismissed as untrue or subversive. Strange tales are valuable because they can break down the limitations of petty understanding and reason, just as Daoist parables do. His arguments, penned to an obscure manuscript with no immediate hope of publication, were presumably aimed at a small hypothetical audience of Neo-Confucian skeptics. Ling Mengchu, on the other hand, is arguing that stories of daily life can compete in interest and novelty with more fantastical and exotic accounts. This is clearly an appeal to a broad, existing reading public, one that Ling was trying to wean from what he perceived to be a considerable appetite for supernatural tales. Ling is thus distinguishing the intriguing and novel sense of strange from the supernatural and exotic sense; the former he tries to capture in his fiction, the latter he vehemently rejects, at least in principle.

"The Rakshas and the Sea Market" ("Luosha haishi"), one of the few Liaozhai tales about a voyage to a foreign country, vividly plays out the argument that strangeness and normality lie in the eyes of the beholder. A young Chinese merchant is blown ashore on a strange island populated by a race of hideously deformed people, who are in turn appalled by his monstrosity. A slightly more human-looking inhabitant finally plucks up his courage and explains the native point of view: "I once heard my grandfather say that 26,000 miles to the west lies the land of China whose inhabitants are all of a weird physical appearance. But this was hearsay; only today do I believe it." Pu Songling is here mocking those proverbial cramped scholars who refuse to believe anything that they have not seen with their own eyes. On this isolated island, the ordinary appearance of the Chinese merchant becomes truly extraordinary. However, the merchant quickly becomes habituated to the sight of these monstrous natives, and he is no longer frightened by them; indeed, he quickly learns how to profit by frightening them. In the world of Liaozhai, the extraordinary is made to seem ordinary, but the ordinary is also made to seem extraordinary.

In the first half of his preface, Tang argues that the strange is a subjective and relative concept. In the second half, in a radical shift, he attacks the common understanding of strangeness as anomaly and its subsequent equation with monstrosity and evil. In his hands, the strange is redefined exclusively in human ethical terms.

I consider that regardless of whether something is normal or abnormal, only things that are harmful to human beings are monstrous. Thus [evil omens like] eclipses and meteorites, "fishhawks in flight and mynah birds nesting," rocks that can speak and the battles of dragons, cannot be considered strange. Only military and civil conscription out of season or rebellious sons and ministers are monstrous and strange.

By relocating the strange to the human world and moving the marginal to the center, Tang has diffused any potential threat that anomaly poses to the moral order. For Tang, strangeness in the sense of evil can exist only in the realm of human events, especially in the political arena. In this regard, he sets the stage for the satiric demystification of the strange often found within Liaozhai itself. At the end of the tale "Guo An," for example, it is announced that this court case is amazing not because a servant saw a ghost but because of the utter stupidity of the presiding magistrate and his miscarriage of justice.

In the other seventeenth-century preface, Gao Heng also argues that the strange is primarily a moral category with canonical roots. He begins by defining the term "strange" to explain its inclusion in the title of the book: "To say that something recorded is 'strange' clearly means that it differs from the norm." This definition is presented as the common understanding of strange, and indeed, judging from other examples, it seems to be so. Like Tang, however, Gao seeks to demonstrate the inadequacy and even the inappropriateness of such a simple definition: by juggling a quotation from the Book of Changes and an audacious pun, he glosses yi (strangeness, difference) as yi (righteousness), one of the cardinal Confucian virtues. This is possible, he declares, because "the principles of Heaven, Earth, and Man, the writings of the Six Classics, and the meanings of the sages, can be 'bound together with a single thread'." Thus this strangeness, this difference, is not external to the proper workings of the universe and moral concerns but is incorporated within them. The potential threat that irregularity poses to order, as deviation or heterodoxy, is neutralized. The strange is no longer unfathomable, but coherent and intelligible.

Both Gao and Tang are clearly operating within what Charlotte Furth has described [in Late Imperial China, Vol. 9, No. 2, December 1988] as "a long-standing Chinese view of cosmological pattern that sought to incorporate anomaly rather than reject the irregular as inconsistent with the harmony of natural pattern." In this tradition of correlative thinking, anomalies were taken as omens manifesting Heaven's will and played a powerful role in political discourse, especially during the Han dynasty. But if we accept the argument that by the late Ming people were "beginning to question the tradition of correlative thinking which assumed that natural moral and cosmological phenomena were rendered intelligible by an underlying pattern of affinities," [Furth] we can perhaps understand Gao's sophistry and Tang's brashness as efforts to reassert the old moral and political implications of anomalies in the face of the age's increasing dissatisfaction with correlative thinking.

Aware that his rhetorical conflation of strangeness and righteousness is shaky, Gao goes on to upbraid would-be critics for construing the great cultural tradition too narrowly. To this end, he refutes the staunchest attack against an interest in the strange, the statement in The Analects of Confucius that "the Master did not speak of prodigies, feats of strength, disorder, and gods." Like many other apologists for recording the strange, Gao argues that Confucius was also the author of the canonical Spring and Autumn Annals, a repository of the very subjects that the Master supposedly avoided speaking of:

The narrow-minded scholars of later generations, whose pupils are as tiny as peas . . . explain away everything they haven't seen with their own eyes with the phrase "the Master didn't speak of it." Don't they know whose pen recorded [the omens of] 'fishhawks in flight and meteors falling"? To blame Master Zuo [commentator on the Spring and Autumn Annals] for such errors is no different from covering one's ears and loudly declaring there is no thunder.

Gao also exploits other loopholes opened by contradictory remarks within The Analects itself to justify such Buddhist-influenced preserves of the strange as a belief in hell and the workings of karma and retribution. Modern critics may explain such textual contradictions as stemming from different strata of scriptural transmission, but for scholars such as Gao and Tang the Classics were a unified whole; any apparent contradiction arose from an inadequate understanding of the lines rather than from a problem inherent in the text. This attitude still prevailed in the seventeenth century, despite the new advances in philological studies (kaozheng), which were subjecting the Classics to increasingly rigorous modes of scholarship.

Instead Gao and Tang prefer to resolve such contradictions in the canon by appealing to the role of the listener or reader. Gao in particular emphasizes the power of the interpretive act to activate the moral potential of a written text:

For the intelligent men of this world, even "what the Master didn't speak of can help in the places that conventional teachings don't reach. [The strange accounts of] The Librarian's Miscellany [Youyang zazu] and Records of the Listener [Yijian zhi] can thus accomplish the same as the Six Classics. But for other types of men, even daily recitations of what Confucius always spoke of can be used to abet evil.

Thus, the good reader can glean enlightenment from any text; the bad reader can find a justification for evil in the most canonical of texts. What is striking in this formulation is not that esoteric or subversive texts require discerning and enlightened readers—appealing to the superior and understanding reader is a conventional move—but that bad readers can pervert a sacred text. Although Gao grounds this point in historical precedent, he is chipping away at the privileged authority of the Classics over other texts: moral authority is contingent not on a superior text but on a superior reader.

We thus find a merging of two seemingly unrelated and even contradictory arguments: since strangeness is a subjective perception, the morality of strange accounts ultimately depends on the reader and his interpretation of the text. This is a particularly powerful method of challenging the canon to include non-canonical texts and non-canonical traditions. But this concern with the superior reader is also symptomatic of an anxiety that Liaozhai will be misread. And for a book to be in danger of being misread, there must be a marked disjunction between the content and the underlying meaning that the inferior reader would miss.

Although Tang and Gao argue along similar lines, Gao's final discussion of the relationship between the strange and the fictive imagination is unique. The last in Gao's series of skeptical interlocutors reluctantly allows that strange things do occasionally occur in this world and that one can chat about them, but he bristles at taking imaginative license with them. "To allow the imagination to gallop beyond the heavens and to realize illusions in the human sphere, isn't this modeled on Qixie['s legendary book of marvels]?" Gao's first defense is rather predictable: he cites textual precedents for indulging the imagination in Sima Qian's biographies of court jesters and the fanciful parables of Zhuangzi. But his next defense is more startling, for he openly calls into question the veracity of the official histories: "And is every record in the twenty-four histories solid [shi]?" Once this point has been granted, he can logically argue that since we tolerate fictions in the histories, we ought also to tolerate fictions in other works.

Gao begs allowance for authorial inspiration and invention, "for the swift literary mind whose pen supplements the process of creation, not only by embellishing the surface but even by smelting the material." The allusion is to the myth of the goddess Nü Wa repairing the toppling sky with molten rock. Thus "supplement" is meant in the sense of "filling in the holes"—of placing new material where it belongs within a pre-existing structure, of mixing small doses of fiction with history. In this metaphor, literary invention bolsters and reinforces order rather than distorting and subverting it. This is not the Western image of the writer who freely imitates the Creation but rather a view of the writer as an assistant to the natural process of creation who selectively fills in gaps as needed. This image of the fictional imagination as a "rock filling in holes" culminates in the opening of the eighteenth-century novel Story of the Stone: the novel itself originates as a rock rejected from the celestial repair process, who becomes both the protagonist of the story and the surface upon which the story is inscribed.

But for Gao, the hard-earned license of literary invention is not to be squandered; it must be well spent in refining human beings. The polarity between exceptional and non-canonical (qi) and orthodox and canonical (zheng) that permeates the discourse on the strange is hereby introduced. Literary invention is qi, refining human beings is zheng; they are two sides of the same coin, not incompatible extremes.

The Second Wave: Self-expression and Allegory

The earliest discourse on Liaozhai primarily defends the tradition of recording anomalies: Liaozhai itself is hailed as a superior but typical example of the zhiguai genre. To this end, an attempt is made to redefine the notion of the strange and to widen the margins of mainstream literature. But the next group of writers, particularly those seeking to publish the manuscript in the fifty years following Pu Songling's death, advanced a radically different approach. These new champions of Liaozhai sought to distance it or even to remove it altogether from the anomalies tradition, claiming that the book was not really about the strange at all.

This tendency was reflected in the first published edition of Liaozhai. The prefect Zhao Qigao (d. 1766), who sponsored the publication, mentions in his foreword that he had excised forty-eight of the shorter, more insipid, and more commonplace items. Although much attention has been paid to Zhao's censorship of a group of supposedly anti-Manchu tales, these number only a handful; the rest are standard records of anomalies in style and content: unembellished, factual reports of strange events such as "A Freak Melon" ("Gua yi"), "A Passion for Snakes" ("She pi"), or "The Clam" ("Ge"). Moreover, Zhao tells us he had originally planned to publish only the tales he considered the best, but he eventually decided to append the ones left over after his initial selection to the end of the book. According to Allan Barr [in "The Textual Transmission of Liaozhai zhiyi," HJAS, Vol. 44, No. 2, 1984], "The tales which were later incorporated . . . are by no means lacking in interest, but have much more in common with the short anecdotes recorded by other seventeenth and eighteenth century writers, and as such, are rather unexceptional." In other words, these last tales, which tend to accentuate Liaozhai's similarity to conventional collections, are relegated to the most inconspicuous place in the book and grouped together as an afterthought.

Although Zhao does acknowledge Liaozhai's affiliation with the anomalies tradition, the aim implied in his selection of tales is to distinguish Liaozhai from a stereotypical image of strange accounts. Zhao's secretary and the collator of the edition, the painter and poet YuJi (1739-1823), explicitly states this idea in his preface: "Comparing it to Qixie's book of marvels or saying that it differs little from collections of rare phenomena or strange tales is a very shallow view and one that greatly contradicts the author's intent."

Pu Lide (1683-1751), Pu Songling's grandson and a keen advocate of Liaozhai's publication, makes this point even more forcefully in a colophon to an edition that he never succeeded in publishing:

Since this book has the word "strange" in the title, someone who doesn't know the work will assume that it must be like The Magician's Records [Yu Chu zhi] or Seeking the Spirits [Soushen ji], or else that it's something like Su Shi's ghost stories, randomly selected and casually prolonged, which are told simply as material for conversation, otherwise they'll say the title is unfair. Someone who knows the work, on the other hand, will say that it uses the supernatural to demonstrate rewards and punishments. But none of these understand this book.

In Pu Lide's scheme, the ideal reader of Liaozhai is not one who reads the stories for pleasure or one who understands the work as a didactic tract, but one who realizes that the book is an act of serious self-expression.

Earlier, Gao Heng had introduced the dichotomy between Liaozhai's surface content and its underlying meaning, appealing to a superior reader capable of discerning this meaning, but he was still interested in the subject matter of the tales and the implications of the strange. In contrast, the second wave of writings on Liaozhai vehemently denies the importance of its content. The bizarre subject matter of the tales is dismissed as a smoke screen, one that veils not so much a concrete meaning as the presence and intention of the author.

Writers adopting this new approach interpreted the strange almost exclusively as a vehicle for the author's self-expression. Liaozhai is lifted into the highest reaches of the literary tradition, not by challenging the conventional boundaries of that tradition but by assimilating strange tales to the autobiographical reading conventions of the major literary genres, especially poetry. The ancient definition of poetry, "that it speaks of what is intently on the mind," had long been extended to other literary genres and other arts; by the late Ming and early Qing, this theory of self-expression could be applied to virtually any field of human endeavor, no matter how trivial or eccentric.

In this mode of interpretation, recording the strange was merely the means through which Pu Songling articulated "what was intently on his mind"; the very outlandishness of the material alerted the reader to the personal distress behind the work. For the reader who styled himself a knowing reader, a zhiyin (literally, "one who understands the tone"), the primary question was no longer "What is the strange?" or "What can we learn from the strange?" Rather, it was "Why would a man channel such extraordinary talent into a work of such a dubious genre?" Read against the background of Pu Songling's lifelong failure to realize his political and social ambitions, the strange content of Liaozhai was familiarized and excused. As the collator Yu Ji lamented [in Wusheng xi], "He entrusted to this book all the extraordinary qì [energy] that otherwise had no outlet in his life. And so in the end he did not care that his accounts often involve things so weird and unorthodox that the world is shocked by them." (This mechanistic view of qì, which here seems to mean something like creative energy, may remind the twentieth-century reader of the Freudian model of libido: if denied access to a proper outlet, it will involuntarily force its way out through some other channel.)

The promotion of Liaozhai as the author's self-expression probably began to take shape toward the end of Pu Songling's life. By this time, it had become clear that Pu Songling would never achieve conventional success and that Liaozhai, which had expanded in size and scope over the years, would be his lifework. The first written evidence of this view appears in a grave inscription commissioned by Pu Songling's family: since the normal channels were insufficient for Pu to unleash his pent-up sorrow, he "sought out the strange and composed his Records of the Strange. Although things in it involve the fantastic, his judgments are sober and serve to warn the people." This eulogist, Zhang Yuan (1672-1756), bore a strong resemblance to the man he was eulogizing. Like Pu, he was a first-degree holder who spent most of his life failing higher examinations, the only avenue to success for intellectuals of limited means; like Pu, he was a man of literary talent forced to support himself as a tutor in a wealthy household, separated from his own family. Both Pu and Zhang, then, epitomized the frustrated, public-minded literary man unable to realize his ambitions in the political, social, or literary system. This resemblance reveals not so much an uncanny correspondence between the two men as the typicality of Pu Songling's career during the Qing. Yuan Shishuo's painstaking study of Pu Songling's friends and family demonstrates that this pattern applies by and large not only to Pu's childhood friends and his pupils, but even to his sons and his favorite grandson, Pu Lide. The literary work of such frustrated scholars, especially if it betrayed any originality or impropriety, was invariably interpreted to fit the ancient paradigm of the worthy man who meets unjustly with failure and so vents his sorrow and disaffection in literature. Zhang Yuan's eulogy introduces the self-expression theory not only because he felt sympathy and admiration for his subject, but because it was by then virtually required to confer literary value on an unusual work and to explain its emotional power.

The pervasiveness of this reading tradition ensured that a brief biography of Pu Songling would be inserted in the first published edition of Liaozhai and appear in the many subsequent reprints. Later readers thus began their reading of the book with a strong impression of the author's personal failure. One such reader, the late Qing scholar Fang Junyi (1815-89), professes wonder that a writer of such exceptional talent chose to squander it on fantastic tales rather than employing it more fruitfully in poetry and prose essays. But the question already contained within it the answer: the choice of form and subject matter was given meaning as a desperate act. Thus Fang concludes: "This work must certainly have been written by a great man who met with failure in his time. I ache on his behalf." Here we see a two-way process at work: an image of the author's life gleaned from his writing is reinforced by his biography, and this knowledge is then read back into his work.

What caused this shift in interpretation? We cannot explain it as a result of historical differences between the intellectual climate of the seventeenth century and that of the eighteenth. Seventeenth-century readers were just as prone as eighteenth-century ones to interpret problematic works as acts of self-expression. Seventeenth-century readings of the macabre and visionary poetry of Li He (791-817), for instance, reveal exactly the same impulse to locate stereotypical political motivations behind a difficult work (although in the case of Li He's poetry these explanations seem much more forced). Alternatively, a famous eighteenth-century recorder of the strange, the prolific and successful Yuan Mei (1716-98), specifically forestalls such an interpretation of his work by telling the reader that the contents of his collection were gathered purely for fun, "not because I was moved by something."

A better explanation for this shift may be found in the aging of Liaozhai. Often the process of interpretation follows its own pattern, one that may have less to do with a specific historical period than with the passage of time and how this alters subsequent views of a work. In this light, the reinterpretation of Liaozhai as a vehicle for self-expression, that is, as a plaint of personal failure and a diatribe against the failings of the age, is a highly predictable move. It is predictable not only because it was an ancient way to reclaim works that otherwise threatened the tradition, but also because the work itself had become gilded with the patina of age. To Pu Songling's senior contemporaries Gao Heng and Tang Menglai, he may have been a man of talent, but he was an insignificant figure. Moreover, when they wrote their prefaces, it was still not too late to hope for an improvement in Pu Songling's career, and the collection was much more modest in scope. For those writing later (and for critics of our times who may identify with Pu Songling's plight), the author's personal failure, which seemed merely pathetic in its own time, lent a tragic glamor and profundity to Liaozhai. One treats the work of a dead author differently from that of a living writer.

As the emphasis shifted from the content of Liaozhai to its author's intention, a general allegorical reading of the tales perhaps became inevitable. In this reading of Liao-zhai, the evil demons and ghosts in the stories are transparent symbols of human wickedness, the bureaucratic hells of the underworld satires on corrupt human officialdom. Pu Songling was certainly cognizant of the metaphorical possibilities of the strange, a tradition that preceded the zhiguai genre and could be traced back as far as Zhuangzi and Liezi, works he particularly loved. In many tales he calls attention to an allegorical reading, usually in the evaluative comments following a story, under his sobriquet Historian of the Strange. For example, in "The Painted Skin" ("Hua pi"), a man who has been dallying with a beautiful woman peeps through the window one day and discovers a hideous demon using a paintbrush to touch up a human skin spread out on the couch. She lifts up the skin, and "as though shaking out a garment," drapes it over her body, transforming herself back into a beautiful woman. When he seeks to exorcise her through a Daoist charm, she flies into a rage and tears out his heart. The Historian of the Strange underlines the obvious moral allegory in his final comments to the story: "How stupid are the people of this world! Someone is obviously a demon, but people consider her beautiful." This exact point, that beautiful appearances can conceal souls blacker than any demon's, is in fact offered in collator Yu Ji's preface as an example of how to read the strange in Liaozhai.

Gao Heng, the author of the first seventeenth-century preface to Liaozhai, had already hinted that the more fanciful subject matter of the tales could be explained as yuyan—literally as "loaded words," a common, all-purpose Chinese figure variously translated as "allegory," "metaphor," and "parable." In its broadest sense, yuyan designates fiction as opposed to fact. A comment to the seventeenth-century novel The Carnal Prayer Mat (Rou putuan) makes this usage quite clear: "Fiction is 'loaded words.' To say that words are loaded means they're not fact." Since the term yuyan is employed so broadly in this period, it may be best to think of it simply as "figurative language"—something not meant to be taken as literally true that points to a larger truth.

Although Gao Heng introduced the figurative possibilities of Liaozhai, he was still willing to tolerate the coexistence of several levels of meaning, and he enjoyed playing with the intellectual paradoxes posed by the concept of strangeness. The lapses in logic and wide leaps in his preface reveal a refreshing lack of dogmatism. The next generation of readers, men like Yu Ji and Pu Lide, however, are rigorous allegorists: they reject the literal sense altogether and retain only the figurative moral sense. By reducing a story to only one possible meaning, they eliminate Liaozhai's strangeness; they try to homogenize the collection, both in terms of itself (all the stories are alike) and in terms of other works (all great literature is alike).

One of the most original discussions of the interpretive problems posed by Liaozhai appears in a preface that has only recently come to light. Written by the philologist and official Kong Jihan (1739-89), a member of the illustrious Kong clan that traced its origins back to Confucius, this preface was preserved in Kong's collected works. This preface may be seen to some extent as a bridge between the first wave of interpretation and the second, or as a compromise between the two.

For Kong, the central problem raised by Liaozhai is still its strangeness (yi), which he explores in terms of its related meaning, difference. He begins by setting out the common understanding of the strange: "People always consider that what runs counter to the norm and counter to nature is strange." But he immediately wonders what happens when so many tales about the strange are read collectively: "When you put together all these many piled-up stories and compare them, it's like fishing in a dried-up marsh—though every fish head is strange or different, they no longer seem strange or different." When so many tales are assembled, the impression of strangeness disappears, for the tales resemble one another more than they differ.

This point repeats an earlier objection leveled at the truly voluminous twelfth-century zhiguai collection, Records of the Listener. "Now it is only because things that run counter to the norm and counter to nature are rare that we say they're amazing. If, however, they're too numerous to record, then we can no longer find them strange." But Kong refuses to conclude that Liaozhai transcends strangeness or that strangeness is only relative. He continues: "Then why did the author put the word 'strange' in the title? Because it can be considered strange."

Kong's fish-head analogy exposes the paradox that in quantity unusual things seem to lose their singularity. This leads him to introduce the opposite paradox: when ordinary things we take for granted become rare, they suddenly become strange. Kong's examples are the biographies of "singular conduct" (duxing zhuan) in the dynastic histories. "Transmitting biographies of 'singular conduct' in the histories began with [the historian] Fan Ye. He placed them in a separate category because they differed from ordinary biographies, that is, because of their strangeness. But all the biographies of singular conduct that he transmitted display loyalty, filial piety, and virtuous principles. These are qualities present in everybody's heart; so how could they be considered strange or different?" Kong resolves this contradiction, one that has profound implications for Liaozhai, by suggesting that in Fan Ye's time morals were so odious and rebellions so frequent that ordinary behavior deserved to be singled out. Then why, he objects, when the ethical climate had presumably improved, did later histories continue the practice of singling out ordinary morality as extraordinary?

Kong is exploring the possibilities of how something can simultaneously be both strange and commonplace. For him, this paradox is the key to Liaozhai's bipartite structure of meaning: "All of what Records of the Strange relates here are things that are seldom seen or heard; so of course people will say they're strange. But nine out of ten are allegories [yuyan], and if we generalize [tong] their meaning, then none of them are about things that people would say are strange."

Kong locates the structure of allegory in the term yuyan itself, where yu is the figurative meaning and yan the literal meaning. The things the stories describe are strange because they are unusual, but the ethical values they convey are commonplace. Unlike his contemporaries Pu Lide and Yu Ji, Kong is not entirely willing to dismiss the literal content of the tales and concede that Liaozhai is not strange. Nor is he quite willing to follow his predecessor Tang Menglai and dismiss strangeness as purely subjective perception. For Kong is quite frank about the pleasure that people (including himself) take in reading about the strange, a pleasure that is not necessarily diminished by grasping the underlying moral significance. "If people don't find strange the meaning of the allegory but find strange only the words as written, it is because of people's fondness for the strange. But if this fondness for the strange is pushed to the opposite extreme [i.e., completely negating it?], then I don't know what happens to the notion of strangeness!"

Kong posits two levels of reading and three kinds of readers for Liaozhai: the frivolous reader who sees only the obvious allure of the strange; the dogmatic reader who sees only the hidden moral or satirical meaning; and the hybrid reader who sees the surface and underlying meanings and is affected by both. The third reader in his scheme is naturally the best. Thus Kong resembles other interpreters of Liaozhai who attempt to prescribe an ideal reader. He concludes, however, with another paradox:

We can't know whether future readers of Records of the Strange will be startled at its strangeness and take delight in it. We can't know whether some will despise the allegory and grow furious or enraged at it. And likewise, we can't know whether some will comprehend the strangeness of both the allegory and the words and sigh passionately, shedding tears over it. For we see that people call strange what they find strange and don't call strange what they don't find strange. Someone might even deny that Records of the Strange is strange and argue instead that it is only reading it that is strange, thus arguing that there's nothing really strange about the work at all—but how could this be?

Kong's scheme presents strangeness as an elusive concept, in constant danger of disappearing into relativity, subjectivity, or allegory. He is ambivalent about whether as a concept strangeness exists independently in the abstract or whether it must be grounded in concrete readers and reading. In the end, he seems to propose a two-tiered reading method, in which the strange is accepted as both a subjective and an objective phenomenon, and in which the surface allure of strangeness and the internal moral balance each other out.

Wang Jinfan, a contemporary of Kong's who published a rather drastically altered edition of Liaozhai in 1767, also explores some paradoxical implications of the strange. Like Kong, Wang distinguishes between the content of the tales, which is admittedly strange, and their underlying morality, which is decidedly ordinary: "There are certainly strange events in this world whose underlying principle is ordinary, and extraordinary language whose intent is orthodox." Nevertheless, Wang is more interested than Kong in the didactic potential of Liaozhai. Thus, rather than propose an ideal reader who perceives the author's true intentions, Wang posits two inferior extremes who are manipulated by the author, the uneducated reader and the overly sophisticated reader: the former is aroused by the satiric moral of the tales, the latter finds new delight in conventional morality. From a stock appeal to the vastness of the universe to support the claim that strange things really do exist, Wang shifts to another important topic related to the discourse on the strange: fiction making. If the principle behind an event is true, he claims, then it does not matter if the event occurred or not. In the end, he attempts to collapse the disjunction between story and message by appealing to the ancient principle that opposites become each other at their extreme: "Thus, there is nothing that is not figurative and nothing that is not real."

The Third Wave: Style and the Analogy to Vernacular Fiction

The approaches introduced above may differ over the meaning and import of the strange in Liaozhai, but they basically agree that the content of the book is at stake. But the third wave, the authors of detailed, full-length commentaries on published editions of Liaozhai, circumvents this debate almost entirely. The strange is no longer a charged issue for them. What is most valuable in Liaozhai is no longer insight into the workings of the universe that it contains, the intellectual paradoxes that it poses, or the allegorical self-expression that it conceals. Instead nineteenth-century interpreters defend Liaozhai largely on grounds of literary style and narrative technique, and these concerns shape their entire commentary project.

To summarize, because of their subjective and relativistic understanding of the strange, previous defenders of Liao-zhai had in some way to situate the strange in the reader. For them, the strange was not an absolute value or independent quality but was realized only in the reading process, for it required interpretation and mediation. In the nineteenth-century discourse on Liaozhai, the concern with the strange per se evaporates; what remains is essentially an interest in the reading process itself. Liaozhai is now defended because its mastery of language and allusion can teach one to read other more important texts, such as the Classics and histories. Commentator Dan Minglun (1795-1853) exemplifies this new approach in his 1842 preface:

I remember that when I was losing my baby teeth, I'd come home from school and read Liaozhai's Records of the Strange. I couldn't bear to put it down. My father used to scold me: "How can a boy whose knowledge is still unformed like to read about ghosts, fox-spirits, and freaks!" A friend of my father's once happened to be sitting there, and he asked me why I loved this book. "Well," I replied, "all I know is I enjoy how in some places it's allusive like The Classic of Documents, valuable like the Zhou Rites, or vigorous like the Ritual Canon, and how in others the narrative is profound like the Zuo Commentary, The Conversations of the States, or Intrigues of the Warring States. From Liaozhai, I also gain insight into literary methods." When my father heard this, his wrath turned to laughter.

The child Dan has precociously demonstrated himself to be a "better" reader than his father by divorcing Liaozhai's problematic content from its brilliant literary style. Once again an obvious pleasure in the strange has been deflected onto another, subtler level of reading. To borrow Formalist terms, we may say that our child-commentator has distinguished discourse ("the world of the author-reader") from story ("the world of the characters"). This favoring of discourse over story characterizes the great Jin Shengtan's (1610-61) influential approach as commentator and reader of fiction and drama. Jin's annotated and amended editions of The Water Margin (Shuihu zhuan) and The Western Wing (Xixiang ji) were so successful that they virtually drove previous editions of these famous vernacular works off the market until the twentieth century. Scores of readers, writers, and commentators were trained in Jin's method of literary analysis, in which every word and every sentence were considered deliberate and meaningful within the structure of the work as a whole.

Dan's own commentary clearly reveals that he was well schooled in Jin Shengtan's reading methods. We even begin to suspect that Dan may have been a bit less precocious than he pretends since the gist of his schoolboy eloquence comes directly from Jin's "Reading Instructions for the Fifth Book of Genius" (i.e., The Water Margin). Jin's edition of this novel was specifically addressed to his young son.

In the past when children read The Water Margin, all they learned were some trivial episodes. Now when they read this edition, they'll learn some literary methods; and they won't learn literary methods only in The Water Margin, they'll also be able to detect them in books like Intrigues of the States and Records of the Historian. In the past when children read books like Intrigues and Records of the Historian, all they saw were some trivial episodes—how absolutely ridiculous! . . . Once children gain some sense of literary methods, they'll be unable to tear themselves away from such books. The Water Margin can do quite a lot for children.

In fact, even Dan's recollection of his boyish love for Liaozhai echoes Jin's own account of his childhood passion for The Water Margin, which "he clasped to his bosom day and night."

These commentators are by no means the only devotees of fiction to ground their strong attachment to a particular work in childhood reading experience. Wu Cheng'en, the supposed author of the fantastic novel Journey to the West (Xiyou ji), wrote in the preface to his zhiguai collection of the youthful delight that he took in such books.

In my childhood, I loved marvelous accounts. As a pupil at the boy's academy, whenever I sneaked off to buy unofficial histories and fiction, I was always afraid that my father or teacher would bawl me out and confiscate them; so I'd read them in secret. But as I grew up, my passion became ever more intense, the accounts ever more marvelous. By the time I was an adult, I sought them in every way until I had accumulated a vast store.

All these writers may have been influenced by the ideology of childhood in the philosophy of Li Zhi (1527-1602). In his famous essay "On the Childlike Heart" ("Tongxin shuo"), Li Zhi argued that all great literature derived from an author's "childlike heart," that is, from a mind that had not lost its original authenticity and spontaneity.

The nineteenth-century commentators' debt to Jin Shengtan is essentially threefold. First, their prefaces to Liaozhai borrow wholesale his defense of vernacular literature—that if properly read, it can teach children, and by extension adults, the literary methods necessary to read beneath the surface of canonical texts, especially the histories. Second, their commentaries adopt the literary methods and criteria Jin and his followers had developed for vernacular literature. Finally, Jin's example showed them that commentary could be as important and taxing as authorship itself.

Feng Zhenluan, an important nineteenth-century commentator on Liaozhai, explicitly modeled himself on Jin, to whose literary prowess he attributed the very survival of the masterpieces of vernacular literature. As he wrote in his 1819 "Random Remarks on Reading Liaozhai" ("Du Liaozhai zashuo"): "Jin Shengtan's commentaries on The Water Margin and The Western Wing are so insightful and cleverly worded that they constantly open the eyes and minds of later readers. This is why these works [belonging to the lowly genres] of the novel and drama have not been discarded in our own day."

Jin Shengtan's favoring of discourse over story provides the cornerstone for Feng's understanding of Liaozhai. From the beginning of his "Random Remarks," Feng emphasizes that Liaozhai's aim is "to create literature" (zuowen), not merely "to record events" (jishi). "Anyone who reads Liaozhai only as stories and not as a literary work is a blockhead!" he warns. The eighteenth-century distinction between literal and figurative readings of Liaozhai has given way to a distinction between literal and literary readings. This new literary reading is not synonymous with a purely formal reading; rather, an attention to formal features alerts the reader to the moral nuances of a text. Although this approach ultimately derived from the traditional method of combing the Spring and Autumn Annals' laconic text for its "subtle meaning" (weizhi), Feng concentrates on the stylistic techniques through which the moral nuances are uncovered rather than on the moral nuances themselves. This emphasis becomes obvious when he castigates the vogue for Liaozhai imitations: "Lacking Liaozhai's ability, these are just stories of wraiths and fox-spirits, exaggerated accounts of strange phenomena. Since their literary style is negligible, their purport is unintelligible"

Feng makes a halfhearted attempt to defend Liaozhai's strange content by echoing an old seventeenth-century argument: numerous accounts of ghosts and prodigies are also included in the histories; Liaozhai cannot be blamed for doing likewise. But Feng's solution is more daring: he suggests that the reader simply "take the writing itself." It does not matter whether the strange events in a story are true or not if the writing is good. Feng has arrived at a full-fledged defense of Liaozhai as creative fiction.

In Feng's "Random Remarks" we encounter for the first time an explicit comparison between Liaozhai and the masterpieces of vernacular fiction and drama. Feng likens Liaozhai to The Water Margin and The Western Wing because all three works have "large structures, finely wrought ideas, extraordinary writing, and orthodox meanings." Unlike eighteenth-century literary claims for Liaozhai that assimilated the work into the autobiographical reading tradition, the nineteenth-century arguments for Liaozhai's literary merit derive from analogies drawn between vernacular fiction and historical narrative. This is a great change, one that attests to the improved status of vernacular literature. In this new environment, Liaozhai is understood as an offspring of a genuine fictional tradition.

By the early nineteenth century, Liaozhai had become so identified with fiction that Feng was compelled to point out that the book records many historical events and personalities. Compare this with original publisher Zhao Qigao's caveat that although Liaozhai contains some verifiable accounts, it is difficult to take most of it as "reliable history." The emphasis has unmistakably shifted. With the passage of more than a century and with the expansion of the readership outside Pu Songling's native Shandong province, the historical nature of many events and characters in the tales would inevitably fade and be forgotten; the fictional impression of the tales would be correspondingly enhanced. Indeed, a major task of the nineteenth-century annotators was to signal which characters and events had a basis in history and to provide necessary facts about them for the common reader. The impression of Liaozhai's fictionality has accelerated with the immensely greater distance separating the modern reader from Pu Songling's world.

Feng's reading of Liaozhai as literature in which writing takes precedence over event obliges him both to uphold the practice of writing fiction and to defend Liaozhai against the charge of being bad history. The disparaging of the fictional imagination has deep roots in the Chinese tradition. Even fiction's chief defender, Jin Shengtan, argued that it is easier to write fiction than history, for in fiction the author can give free reign to his imagination, whereas in history the author is constrained by the facts. Jin's insight recalls the ancient philosopher Han Fei's famous remark on representation in painting: it is easier to paint a phantom or a demon than a horse or a dog; since no one knows what a phantom looks like, the artist need not worry about painting a recognizable likeness as he would in painting familiar creatures. Although this valuing of mimetic representation in painting was eclipsed quite early in China, vernacular fiction writers frequently used Han Fei's remark to attack the supernatural orientation of popular literature and to defend the focus on daily life in their own work. Feng refutes this charge of reckless imagination by arguing that even when writing about phantoms, Pu Songling always conforms to the logic of the human world; he makes the incredible detailed and vivid enough to seem credible.

Pu Songling's use of fictional detail and dialogue lies at the heart of Ji Yun's (1724-1805) well-known complaints against Liaozhai. Ji, a leading scholar-official who wrote the late eighteenth-century's finest collection of strange accounts, objected to Pu's inclusion of both "short anecdotes" (xiaoshuo) and "narratives in the biographical style" (zhuanji) in a single work. In light of this complaint and the abbreviated style of his own stories, it is clear that Ji Yun's real objection to Liaozhai was primarily epistemo-logical. He maintained that as varieties of historical narrative, both short anecdotes and narratives in the biographical style had to be based on plausible sources—autobiographical experience or eyewitness testimony—and not freely invented by the author, "like plot elements in a play." All stories need not be true, but they must at least persuade the reader that they might have been seen or heard by an actual source. Thus Ji Yun complains: "Now P'u Sung-ling [Pu Songling] gives a vivid picture of the smallest details down to amorous gestures and the secrets whispered before lovers. It would be unreasonable to assume that the writer experienced these things himself; but if he was describing what happened to others, how could he have known so much?"

Pu Songling's stories are too detailed and too vividly dramatized for Ji Yun to accept as based on something heard or experienced by the author himself. For Ji Yun, verisimilitude decreases the impression of a narrative's realness, since he understands realness as "the claim to historicity," that is, as the claim that the events in a narrative really happened. It is not the strangeness of Liaozhai that bothers Ji Yun; rather, Pu Songling's narrative techniques too obviously betray authorial fabrication.

Feng Zhenluan defends Liaozhai against these charges by applying these rigid epistemological standards to the histories. Are the histories always true accounts of events? Are their sources impeccable? Or does their narrative technique also betray traces of overt fabrication? As an example of fictionalizing in the histories, Feng singles out a famous speech in the Zuo Commentary delivered by the assassin-retainer Chu Ni just before he smashed his head against a tree and killed himself. "Who heard the words of Chu Ni beneath the locust tree? How could Master Zuo have known them?" Feng's solution to this and the related problem of discrepancies between different historical accounts of the same event is once again to distinguish between discourse and story: the mode of telling a story may vary without harming the essence of the story. This example in turn helps justify Feng's assertion that he reads the Zuo Commentary as fiction and Liaozhai as the Zuo Commentary.

This argument for fictional license in narrative did not originate with Feng. A letter nearly two centuries earlier from a seventeenth-century collection had cited the identical incident from the Zuo Commentary for the identical purpose: "As far as Ch'u Ni's [Chu Ni's] utterance is concerned, there was no one else to know what he had said, so how did Tso Ch'iu [Master Zuo] know about it?" The letter's bold conclusion is to hail Master Zuo as "the progenitor of a whole line of literary lies." Feng accepts the definition of fiction as literary lies, but argues that "even lies must be told fully," that is, fleshed out with sufficient skill and logic to convince the listener. Feng's defense of lies is thus essentially the same as his defense of painting phantoms. But what is a lie? A lie is an utterance that the speaker knows is untrue. In understanding the Liaozhai tales as literary lies, Feng reflects another nineteenth-century view, that the ghosts and fox-spirits in Liaozhai are nothing but a game, a trick played by the author on the naive reader. Once again, a two-tiered level of meaning is posited; an appeal is made to a superior reader aware of the discrepancy between content and intent who does not let himself be hoodwinked by the author's literary lies. In this last formulation, the strange in Liaozhai has finally become a purely fictional and ironic construct, one predicated on the author's and reader's mutual suspension of disbelief.

To conclude, we may also understand the development of these three interpretive approaches in terms of the circumstances behind their adoption and the context in which they were written. Pu Songling's personal friends wrote the first prefaces and dedicatory verses when his manuscript was still unfinished. Their efforts were inherently social in nature. They wrote to help introduce his work into society, that is, to a limited circle of like-minded readers. These established literary figures and statesmen lent their authoritative voices to an obscure and potentially suspect manuscript, supplying it with a pedigree and moral approbation. To this end, they tried to carve a niche for records of the strange within the dominant literary and intellectual tradition.

The advocates of Liaozhai's publication primarily constituted the second wave. They were arguing to a new class of readers, the general reading public, why people ought to read an unknown author's work. For this reason, they sought to distinguish Liaozhai from the plentiful collections of strange tales on the market, attempting to convince the public that something special about this book warranted purchase and persual. These writers strove to elevate a mere collection of strange tales by reclassifying it as an allegorical work of self-expression, a high literary value not ordinarily associated with works of this kind. At the same time, social networks also shaped this second generation of interpreters. Pu Lide composed one of his postfaces to enlist the help of the sons of his grandfather's friend Zhu Xiang, who had expressed interest in helping him get Liaozhai published. Yu Ji, the collator of the first published edition, wrote his preface at the behest of the publisher, the prefect Zhao Qigao, who was also his friend and employer. Yu Ji's dedicatory verse for the edition is essentially a eulogy to Zhao, who died before the book came out.

The nineteenth-century commentators who constituted the third wave were associating themselves with an already famous book. By elaborating the book's literary methods, by "scratching the author's itch," as Feng Zhenluan put it, they hoped to win literary fame for themselves. (And to some extent, they have succeeded. We remember these men today solely as commentators on Liaozhai.) Because a fictional tradition had been firmly established by this period, the third wave was able to transcend the problem of the book's strange content by "simply taking the writing itself."