Cathy N. Davidson

New York Oxford



Chapter 9--Scanned, copy-edited, spell-checked and tagged by Scott Eric Atkins, The University of Virginia, 9/18/95, 9/22/95.


Afterword: Texts as Histories

Fiction is a lie that tells the truth.

Jean Cocteau

Friedrich von Schiller, one of the most eminent historians of the late eighteenth century, was carried into the lecture hall in Jena upon the shoulders of enthusiastic admirers who deposited him at the podium where he would answer, once and for all, a question he had raised before: "What Is Universal History and Why Do We Study It?" Universal History, Schiller explained, discovered in the past a vast network of causal relationships that, with relentless logic, produced the present. It was the task of the Universal Historian to chart out those causalities in order that the reader might see the coherent whole that was modern Europe, might understand how previous events had produced a stable Continental society that enjoyed an enduring peace and that promised (history portended as much) to last forever. Schiller's speech on that May day in 1789 was constantly interrupted by cheers from the crowd who could not foresee, any more than the Universal Historian, that in less than two months another crowd would storm the Bastille and launch Europe into its most revolutionary epoch. '

This anecdote is also an admonition, for any attempt to make a Universal History assumes an interpretation of the present and then discovers or even creates a past that validates that interpretation. Perhaps even more dangerously, it also projects the past-present into the future, as if the status quo, as defined by the historian, were expressly designed for eternity. A revolution looks different, too, depending on which side of the ocean or Bastille one views it from, all of which would seem to be an obvious enough assertion. Yet it is unsettling in that the alternative to Schiller's Universal History or to the positivism of nineteenth-century history seems to be cultural relativism, a valueless history with its own attendant political and theoretical nightmares.

Obviously, I have not suggested a relativistic reading of the early American novel nor do I propose, in this brief Afterword, a Universal History of the American novel. In fundamental ways, both relativistic account and totalizing history assume the objectivity of the historian, the ability of the historian to catalogue what is, even though the end and interpretation of that descriptive categorization are quite different in each case. Less obviously but no less thoroughly than Universal History, cultural relativism is, finally, a myth in that no historian ever fully avoids value judgment. The simplest fact that one is a historian choosing to examine a particular period in history and a particular problem already encompasses a complex hierarchy of values, ranging from assumptions about periodization (Why define the Revolutionary War as a beginning, not an ending-or a middle, for that matter?) to categorization (Are these facts and do they bear on the question at hand?) to authority (What institutional or other hegemonies allow any scholar the leisure to pursue this search?).

Oppositional or dialogical history challenges conventional literary history by questioning both the relative value of what is examined and the implicit values of the examiner.2 It sees the very processes and ambitions of historiography as products of much larger forces and it seeks to understand the relationships between those present forces and the hierarchical imperatives of the past. History can be crudely but compositely defined as an assemblage of texts (the labor of historians) addressing a different assemblage of texts (the archive). The historian, first, attempts to engage in an exchange with the past through the mixed and amorphous record of what does survive and through the particular interests and agenda of the individual historian-what she or he is looking for and to what end. Second, that exchange is inevitably filtered through the medium of competing histories and other interpretations. In essence, every history-borrowing here, disputing there, consciously or unconsciously filling in omissions left in different accounts or reshaping the territory that another historian has already mapped-reembodies its predecessors even as it labors to suppress them. As Paine noted, "when once any object has been seen . . . it is impossible to put the mind back to the same condition it was before it saw it.... It has never yet been discovered, how to make man unknow his knowledge, or unthink his thoughts." 3 A dialogical history pretends neither to unknow nor to know absolutely. Quite the contrary, it focuses its attention on those areas where the case seems the most obvious, and gets to work where there is consensus. If we already have the answer, we well may have been asking the wrong question.

Some of the pitfalls of a monological literary history are appropriately illustrated by the contemporary critical reception of Arthur Mervyn. Most commentators who read Charles Brockden Brown approach him backwards through the standard nineteenth-century fictional syllabus of Melville, Hawthorne, and Poe. The governing but unexamined assumption underlying this retrospective genealogy is that a few selected writers somehow constitute a unique "American pantheon" and are, simply (and perniciously), the representative authors deserving of special attention. Criticism has already rendered its crucial verdicts and its remaining business is mostly to buttress further its case. Criticism thus becomes a kind of house committee on un-American fictional activities, with consequences pedagogical (what works are taught), economic (what reprinted), political (who is privileged, who invisible), and, finally, interpretive (what becomes the grid whereby we assess other works). Read through this American tradition in order to be placed at the head of it Arthur Mervyn almost inevitably becomes father to the powers of darkness, not brother to The Power of Sympathy. The title character, Arthur himself can be seen as an early unreliable narrator cunningly misdirecting his own story. The gaps in the text exist to be explained away by the enterprising critic who comprehends far more what is at issue than do the other characters in the fiction. Indeed, Maravegli, Estwick, Mr. Hadwin, Curling, Lodi, the Carltons, the Watsons, Fanny Maurice, and, most of all, Stevens all become so many fools serving simply to be gulled by the likes of Welbeck or Arthur. They become the blank faces on the social mask through which the author covertly strikes to be followed only by the most discerning readers, and the novel becomes a kind of Moby Mervyn, Confidence Man, the predictable, if implausible, product of a predetermined canon.

Read within the context of his contemporaries, however, Mervyn seems rather less dark and diabolical than much recent criticism suggests. He need not be an Ahab, a Chillingworth, or even an Usher; he might be mostly a Reverend Boyer or a Major Sanford or even an odd and inconclusive amalgam of the two. And the novel, too, looks different as a cheap two-volume duodecimo ready for the local lending library than as a Center for Editions of American Authors "Approved Text'' replete with a "Historical Essay," "Textual Essay," "Textual Notes," "List of Emendations," "Historical List of Substantive Variants," and "Record of Collations and Copies Consulted."4 The doubles in the novel can seem less Freudian aberrations and portentous symbols and can be seen as plot devices conventional and even commonplace in the novels of the time.5 Various gaps in the text are no longer clear evidence of sinister repressions that the "careful reader" must discern but might be merely the failures of a hasty writer (or an ambivalent man) to tie up his story's loose ends, loose ends of the type that plague (from a modernist perspective) many early American novels. A common feature in these works is the scramble, in the penultimate chapter, to dispose of various subsidiary characters and events in order that the final chapter can focus on the protagonists and their ending. In short, Arthur Mervyn as a work of fiction and as a fictional character looks different depending on the historical context in which we place it/him. Our problem as readers is a version of Dr. Stevens's, and our solution can be no more final than his. Or, to return to the theoretical issue of canon that I have raised before in this study, what we read shapes how we read-a reversal of the usual critical presupposition.

We have a choice of Mervyns. Dialogical history gives us a choice of pasts, too. But that very choice or pluralism is subversive since it implies that a literary tradition is not simply inherited but constructed, and constructed according to the critical categories we devise. For example, as Annette Kolodny has demonstrated, when Moses Coit Tyler set out in 1878 to discover the origins of American literature, he found beginnings that affirmed what he was looking for: a "common national accent" in the American literary past; a tradition "single in its commanding ideas and in its national destinies"; an American literature characterized by "uniformity," like the nation itself. Yet, as critics as diverse as Kolodny and Harold Bloom have argued, an imperative toward canonization, toward the creation of a univocal history (literary, social, political--the three are always intertwined) requires the exclusion of what does not fit into the a priori definition of precisely what is to be defined.7 Moreover, a uniform American literature defines a uniform American and, by extension and implication, excludes those who do not fit the definition by reasons of gender, race, ethnicity, religion, class, education, wealth, region, place of origin (as with immigrants or, conversely, Native Americans), or language.

Nor are the consequences of a reductive theory of American identity purely literary. If high school and college textbooks, standardized tests (typically the most conservative measure of an educational system), doctoral reading lists, or scholarly monographs regularly present a homogeneous America, then the literary or pedagogical image of a nation is at odds with (or simply irrelevant to) the increasingly heterogeneous American population receiving this message. If, after a brief period of educational revisionism, there is again a call for the "basics," then once more pedagogy narrowly defines what is and what is not essential or basic to America. The spirit and educational philosophy of Noah Webster prevail, and the humanities classroom takes on the rather dubious role of perpetrating the status quo by providing a restrictive, exclusionary answer to the question posed earlier by Crevecoeur: "What is an American?"8 It is an answer maintained through a rigorous (if unconscious) process of selection which perpetrates itself generation after generation, as new instructors teach what they were taught, as publishers reprint (with new introductions) what is still being taught, and as scholars analyze the same old works according to whatever current critical interests prevail, the whole cycle spinning merrily on.9

With notable (but relatively few) exceptions, literary history is a history of the most available texts. The conditioning of our perceptions of what literary history is operates on even the most elementary level. An example might be useful here. A number of times in this study I have referred to the "What is an American?" section of Crevecoeur's Letters from an American Farmer. Scholars of American literature certainly know that Crevecoeur did not conclude his farmer's letters with the same unqualified sense of American egalitarianism he early exhibited, nor did the author himself retain his original opinion of the land to which he had immigrated but which he later left.10 When, near the end of the volume, Crevecoeur's narrator/double witnesses the punishment inflicted upon slaves in South Carolina, we have a very different vision of America than that with which he began-and a very different answer to the question, "What is an American?" From the earlier paean Letters from an American Farmer devolves to disillusionment with a system that seemed to promise so much and, at least implicitly, requires the reader to think about reform, about improving a nation which does not fully live up to its promise and its possibilities. Predictably, however, the later sections of Crevecoeur's classic are anthologized far less often than the exultant (if unrealistic) Letter 3. For most readers of Crevecoeur (who typically encounter his work in anthologies, if at all), the important analysis of American racism set forth in the latter portions of the book simply does not exist. It is not part of our literary inheritance.

The issue here is not that literature provides an inaccurate reflection of history but that no documents (from Crevecoeur's Letters to signing evidence on marriage records) can simply be "read" as if they were objective, scientific data produced or preserved as some pure product of a people and the abiding record of their times. The record always suppresses more than it tells. Why, we must ask, are certain records kept in the first place? Why are they saved? The whole process of historiography, the archive itself, must be subjected to rigorous analysis. Who is keeping the records and for what purpose? Who is writing, to whom, and why?

Another example from the early national period can again illuminate the politics of literary history. The Connecticut Rising Glory Poets are often described as representative of the time and place in which they wrote. But what do they represent? The glory of American society or the views of a somewhat hypocritical gentry determined that the Revolution should proceed no further than their own self-interest required; a reasoned opposition to the "irrational" (such as the Shaysites) or a calculated attempt to keep "unruly" peasants in their proper, lowly station? As Kenneth Silverman points out, when Joel Barlow, John Trumbull, David Humphreys, Lemuel Shaw, and other poets began, in late 1786, their "counter-mobocracy" literary campaign in the form of "The Anarchiad," this serial of satires, speeches, and poems (first published in the New- Haven Gazette and then reprinted throughout the East) spoke to fears in much of the populace, for the series sold out so fast that "Humphreys was unable to find extra copies to send to Washington." 11 "The Anarchiad" stressed the need for a strong, centralized government, praised the wise authority of that government, and celebrated a pastoral ideal quite contrary to the realities of country life as perceived by Shays's outraged agrarians. Inverting an earlier language of Whig sentimentalism, the Rising Glory Poets ominously equated Shays's "anarchists" (i.e., anti-Americans) with evil seducers out to despoil the serene dignity of Columbia. In short, the poets attempted to counter incipient social unrest by extolling the vision of an abiding, enduring Federalist Republic grounded in the enlightened interest of entitled men. When accepted as representative Americans, their bias becomes the given, the glory of the early Republic, while the Shaysites are encoded in literature as the would-be destroyers of a nascent democracy.

Yet, if we are going to rely on literature to present us with the data of social history, with a portrait of the American self, we cannot look only at the literature written for privileged readers, which advocated the interests of those readers. By looking at a different genre, the novel, a genre derided by authority figures even while it was increasingly read by the lower orders (as well as by some of those official spokespersons who more publicly condemned it), one almost inevitably encounters a different representative view of America. Although the American novel would soon become respectable (and perhaps lose some of its oppositional edge with age and respectability), its original, lowly status virtually assured that it would early speak for those also marginalized by American society as a whole. Many of America's first novels, as we have seen, emphasized the class, gender, and racial inequities in the new land and even explicitly advocated an end to these inequities. In a novel such as Martha Meredith Read's Monima, or the Beggar Girl for example, we have a strikingly different version of America than the one provided by the Connecticut poets, yet Monima was widely read too-first published in 1802, reprinted in both New York and Philadelphia editions the following year, and pirated, plagiarized, and paraphrased, in condensed forms, in various American magazines and newspapers until the middle of the nineteenth century.

Read's representative American was not some established member of the gentry, nor even a poor boy in the process of making good, but a most disenfranchised member of society, a "beggar girl":

Every day during the week, the almost heart-broken Monima was compelled to beg; it would be an endless task to recount the insolence, repeated insults, the cold contempt, the mortifying strictures on idleness, the affected pity of the seemingly feeling, and the ostentatious charity of the naturally contracted and cringing hypocrites, with their bitter effects on her tender heart. Another week of such excruciating harassment, would have made her an inmate for the grave.

Yes, the less diction is sentimental. But Read's portrait of America is far less sentimental than that presented by the Rising Glory Poets or in Crevecoeur's Letter 3. Read's novelistic vision of America gives us a New World as corrupt as the Old. The Revolution did not perform its office for all Americans equally; wealthy Americans as convinced of their own privilege as any European aristocrats rob Monima of her money in order to force her first into financial dependence and then into sexual submission, graphic metaphors for oppressions by class and gender.

Nor does Martha Read limit her critique to a ruling class. The assistance provided to Monima by the bourgeoisie is equally debilitating. They offer her demeaning employment at starvation wages: "Four months elapsed thus with a degree of comparative happiness; but tho' Monima was indefatigable in her industry, still she could not earn a cent more than merely enough to provide necessary diet and fuel." An early indictment of commodity capitalism, this sentimental novel shows the way in which middle-class Americans carefully calculate how much they must pay the worker to keep her alive (and thus able to work) without giving her the means to rise above her social class-a cycle of economic enslavement. The novel also includes a stinging critique of well-to-do philanthropists who validate their own achievements through pious precepts about America as a "land of opportunity," precepts singularly inapplicable to Monima's distress. Even the proletariat mostly jeer the poor girl when, assaulted and robbed of her "small store of cash" by the villainous De Noix, she shouts, "Murder! Murder!"--fully aware that this calamitous theft could well spell the death of herself and her infirm father: "A mob quickly gathered about her. Some judged her to be intoxicated, whilst others were for examining her throat to see if it were cut. A considerable time the unfortunate girl was the jest, and butt of the hardened unfeeling populace. Her wailing, her lamentation, her tears were of no avail, they rather increased the sport and derision of the mob."12

If Monima's plight strikes the contemporary reader as ludicrous or overdone, whereas "The Anarchiad" represents the "real" America, perhaps it is because scholarship has long had a vested interest in documenting one genealogy and not the other. 13 The problem lies not in using literature as a source for social history but in thinking that literature, any more than history, exists except as humans compile, preserve, and canonize it; or in forgetting the ideological purposes served by the study of literature and history within a society. Moreover, what is officially saved exists not merely in its own right but according to the formulations whereby it is preserved. Thus, if we analyze, in scholarly publications and classroom lectures, the earlier Godwinian novels of Charles Brockden Brown primarily in terms of the same author's purposed conversion to a middle-aged conservatism, we in effect perpetrate a parable of impetuous, improvident revolutionary youth maturing to sober, sensible, right-minded adulthood (and, one must add, to the end of art also). Or, conversely, with Barlow we reverse the procedure and teach early Federalist works such as "The Anarchiad" or The Vision of Columbus (1787) but decently cover up his decline by ignoring later radical texts such as Advice to the Unprivileged Orders (1792-94). In short, we carefully select our authors, we select from those authors, we select our genres and our readers all to confirm the history our literature "proves."

Ideology, of course, is present in any version of American literary history (and certainly the present study is no exception), but the predominant ideology postulated to date has been much more the ideology of the Rising Glory Poets than of Martha Read. Like history, literature selects its data, its canon, in order to confirm not the past but the present, as Raymond Williams movingly notes in his personal account of how his father, a rural laborer, set to memorizing a work of literature, Herrick's patronizing paean to the rural poor, A Thanksgiving (1647):

As it happens, I first read this poem as a child, under a roof and porch probably lower than Herrick's, and I could then neither get the lines out of my mind nor feel other than angry about them. My father had brought it home, in a book called Hours with English Authors, which was a set-book for an evening class he was attending in the village. He had been asked (it is how values are taught) to learn it by heart he asked me to see if he could. I remember looking and wondering who the poor were, and . . . if the poet's condition was indeed so low. I understand that better now. 14

Literature is one way in which values are taught in a society, and canonization is the mechanism for that larger enterprise. By focusing on fiction, I have attempted to re-place the standard literary syllabus of the new Republic by adding another voice to the history of America's postrevolutionary epoch. But making history is only one reason to read early novels, and, borrowing here from New Criticism, I hasten to emphasize that fiction is not history and that literature can never be simply "reduced" to history. We cannot turn the early American novel into some Procrustean bed exactly accommodating the complex history of an era, nor, conversely, can we insist that postrevolutionary history is a bed in which the early American novel comfortably lies. To subject novels to a reductive content analysis in order to extract their historical "meaning" misses the meaning of the genre and deprives a major form of republican discourse of its power and effect. For the early American novel entered fully into its own culture not simply as a commentator on the times but as an active participant in the interpretation of events-which is to say, as a creator of events. The historian who would return to these novels for some Universal History of the age necessarily meets with frustration, in that novels are not only susceptible to varying interpretations but require them, not only resist rational reduction to an ostensible message but are inimical to the whole prioritizing of the rational over the "wisdom of the human heart."

Novels are not history. "Literature becomes redundant," as Dominick La Capra warns, "when it tells us what can be gleaned from other documentary sources." 15 That caveat is crucial. If we read literature simply for its historical content, for its data, we are throwing out the fruit and settling simply for the chaff-chaff, moreover, that can better be gleaned from less contradictory and less ambiguous sources. More to the point, such historicizing of literature renders each a kind of puppet parody of the other. In a process of dizzying circularity, the meaning of history is reduced to the meaning of a text whose meaning is proven by its historical context. Neither history nor fiction speaks in a unitary voice, nor with enough authority to drown out the other. As Jorge Luis Borges has insisted in a somewhat different context, "the concept of a definitive text" or, he might have added, a definitive reading "belongs to religion or fatigue." 16

The early American novel happened at those places in its society where issues were unresolved, at the interstices between public rhetoric and private expression. The very formlessness of the new form made it resistant to univocal readings and served as a catalyst to enveloping explications, tentative trials, and forays into alternative possibilities of meaning where readers might not willingly venture on their own. The new novel genre welcomed the participation of its readers-even those marginally educated new readers who had no place, except a passive and subservient one, in the classical rhetorical tradition.

In the intensely personal, secluded world of the imagination even the most "commoditized" best-seller can assume a special, even intimate, possibly subversive shape. The "mass privacy" of the novel both is and is not the "real" world and may be-- depending on the interaction between an individual reader and an individual text--either more or less powerful than other events occurring outside that interaction.17 A novel never simply mirrors reality. It is its own artificially framed world, an organized structure with its own rules and interpretations. At the same time, it is part of a larger structure (or a different one) and is determined by those economic, institutional, and ideological forces that govern its composition, its publication, its circulation, its reading, and the end-canonization or obscurity-to which it is read.

[Here would be an image that has not been included in the electronic version:


Fig. 16. Although published in an 1811 edition of Charlotte Temple, this frontispiece suggests that readers already associated Charlotte with pathetic death and dying, all of which became almost a cult by the mid-nineteenth century. Courtesy of the American Antiquarian Society.]

A novelist, it must also be remembered, chooses to write fiction precisely for what the form allows and for what it disallows. In America during the early national period that choice necessarily meant doing something suspect. For all the protestations of being "FOUNDED IN TRUTH," a novel was still a novel, and at least part of what made it interesting was that it was written in defiance of an established social and literary tradition and for readers not recognized by many other literary forms or acknowledged only as minor participants in the whole process of making literature. As I have suggested earlier, the novel, in certain fundamental ways, is its readings and its readers. And those readers selected novels over tracts or histories precisely because novels were not tracts or histories. You could empathize with the characters in novels in ways you could not with the generals and kings of the history books. You could believe in them in ways in which you did not believe in the generals and kings. They spoke your language, and a language, as novelist Margaret Atwood has observed, is everything you do, even the grammar of a seduction and the declaration made by a subsequent death.18

I have been arguing, unfashionably, for the mimetic properties of the novel because the reader of the early national period read mimetically, so much so that for many readers fiction came to have a paramount reality of its own. The grave of Charlotte Temple over which so many lovers wept provides an apt symbol of the relationship between early American readers and their books, and it is with that multivalent and symbolic scene of pilgrimage to a fictional character's real tomb that I wish to conclude this study. For, unlike a Universal History which totalizes the past and presumes to predict the future, novels such as Charlotte Temple, The Coquette, or Arthur Mervyn are very much rooted in a perpetual present, the interactions between a text and a reader--any reader--which gives enduring life to the artifact of a book and enlivens words printed on a page into a novel, a finally magical creation whose origins, inspirations, and aspirations no historian or critic can presume to master.

Chapter 9

1. For somewhat different versions of this anecdote, see Louis O. Mink, "Narrative Form as a Cognitive Instrument," pp. 135-36, and Lionel Grossman, "History and Literature: Reproduction or Signification," pp. 18-19, both in The Writing of History: Literary Form and Historical Understanding, ed. Robert H. Canary and Henry Kozicki (Madison: Univ. of Wisconsin Press, 1978).

2. For important discussions of oppositional criticism, see Frank Lentricchia, Criticism and Social Change (Chicago: Univ. of Chicago Press, 1983), esp. p. 15; and Raymond Williams, Marxism and Literature (New York: Oxford Univ. Press, 1977), esp. pp. 112-13. My discussion has also been influenced by: Nicola Chiaromonte, The Paradox of History: Stendahl, Tolstoy, Pasternak and Others (London: Wiedenfeld and Nicolson, 1970); Hayden White, Metahistory: The Historical Imagination in Nineteenth-Century Europe (Baltimore: Johns Hopkins Univ. Press, 1973); and White, "The Historical Text as Literary Artifact," in The Writing of History, pp. 41-62. On a more practical level, the eminent historian William H. McNeill, in Mythistory and Other Essays (Chicago: Univ. of Chicago Press, 1985), has recently argued that world survival in part depends upon the recognition that one historian's truth is another's myth.

3. Thomas Paine, The Rights of Man (New York: Pelican Classics, 1969), pp. 140-41.

4. Charles Brockden Brown, Arthur Mervyn or Memoirs of the Year 1793 (Kent. Ohio: Kent State Univ. Press, 1980).

5. The eighteenth century's fascination with doubleness has been discussed by Sydney J. Krause, "Historical Essay," in Charles Brockden Brown, Edgar Huntly or Memoirs of a Sleep-Walker (Kent, Ohio: Kent State Univ. Press, 1984), p. 296.

6. Moses Coit Tyler, A History of American Literature, 1607- 1765, 2 vols. (1878; repr. Williamston, Mass.: Corner House Publishers, 1973), I:v. These passages are quoted and discussed in Annette Kolodny's important essay "The Integrity of Memory: Creating a New Literary History of the United States," American Literature, 57 (May 1985), 291-307.

7. Harold Bloom, "Criticism, Canon-Formation, and Prophecy: The Sorrows of Facticity," Raritan, 3 (1984), 1; and Kolodny, "Integrity of Memory," pp. 291-307.

8. The relationship between pedagogy and literary theory is one of the most important issues now facing the profession. For an excellent discussion, see William E. Cain, The Crisis in Criticism: Theory, Literature and Reform in English Studies (Baltimore: Johns Hopkins Univ. Press,1984), esp. pp. 67-121 and 247-77. The theoretical journal boundary 2 is also planning two double issues (l986-l987) devoted primarily to the topic of ideology and the politics of humanities pedagogy. See also Nancy Hoffman, "White Women, Black Women: Inventing an Adequate Pedagogy,'' Women's Studies Newsletter, 5 (Spring 1977), 21-4; and the essays in The New Feminist Criticism: Essays on Women, Literature, and Theory, ed. Elaine Showalter (New York: Pantheon, 1985), esp. the essays by Barbara Smith, "Toward a Black Feminist Criticism" (pp. 168-85), and Deborah E. McDowell, "New Directions for Black Feminist Criticism" (pp. 186- 99). For provocative discussions of the socializing role played by the traditional humanities, see Carol Gilligan, In a Different Voice: Psychological Theory and Women's Development (Cambridge: Harvard Univ. Press, 1982); and Richard Ohmann, English in America: A Radical View of the Profession (New York: Oxford Univ. Press, 1976). For one of the starkest assessments of the function of the traditional humanities--"There has never been a document of culture which was not at one and the same time a document of barbarism"-see Walter Benjamin, "Theses on the Philosophy of History," in Illuminations, trans. H. Zohn (New York: Schocken, 1969), p.253.

9. Two recent studies of anthologies--what is included, what excluded--arrive at different, but not incompatible, conclusions about the politics of canonization. See the chapter "'But Is It Any Good?': The Institutionalization of Literary Value," in Jane Tompkins, Sensational Designs: The Cultural Work of American Fiction, 1790-1860 (New York: Oxford Univ. Press, 1985), pp. 186-201, for an analysis of the ways in which the canon changes according to the prevailing economic, literary, and political interests of the generation (although each generation insists that it knows what constitutes a masterpiece). Susan Koppelman has isolated a different pattern in her survey of over a century of short story anthologies. She documents an almost unchanging percentage of both women writers (of all races) and black writers (of both sexes) in the anthologies, but she notes that each generation of anthologists tends to select new representatives for women and blacks. The result is a sense of historical amnesia, as if there were no tradition of women or minority writing in America. See the introduction to her anthology Old Maids: Short Stories by Nineteenth-Century U.S. Women Writers (New York: Pandora Press, 1984).

10. See Bernard Chevignard, "St. John de Crevecoeur in the Looking Glass: Letters From an American Farmer and the Making of a Man of Letters," Early American Literature, 19 (Fall 1984), 173-90; Everett Emerson, "Hector St. John de Crevecoeur and the Promise of America," in Forms and Functions of History in American Literature: Essays in Honor of Ursula Brumm, ed. Winfried Flunk et al. (Berlin: Erich Schmidt Verlag, 1981), pp. 44-55; and Albert E. Stone,Jr., "Crevecoeur's Letters and the Beginnings of an American Literature," Emory University Quarterly, 18 (Winter 1962), 197-213.

11. Kenneth Silvemman, A Cultural History of the American Revolution (New York: Thomas Y. Crowell, 1976), p. 513.

12. Martha Meredith Read, Monima, or the Beggar Girl (New York: T. B. Jansen, 1803),pp. 434, 436, 440-41.

13. The ways in which purportedly aesthetic judgments are actually founded upon ideological (and often tautological) arguments has been perceptively analyzed by Nina Baym in "Melodramas dramas of Beset Manhood: How Theories of American Fiction Exclude Women Authors," American Quarterly, 33 (Summer 1981), 123-39.

14. Raymond Williams. The Country and the City (New York: Oxford Univ. Press. 1973) p. 72.

15. Dominick La Capra, History and Criticism (Ithaca: Cornell Univ. Press, 1985), p. 126. For additional comments on the relationship between literature and history, see also La Capra's Rethinking Intellectual History: Texts, Contexts, Language (Ithaca: Comell Univ. Press, 1983), esp. pp. 13-22. For an analysis of the politics of literature by a social historian, see Steven Watts, The Republic Reborn: War and the Making of Liberal America, 1790-1820 (Baltimore: Johns Hopkins Univ. Press, forthcoming). Two journals concerned with literary theory have recently addressed themselves to the relationships between literature and history. Poetics, 14 (1985), is devoted to the related issues of the empirical sociology of cultural production and the politics of writing the history of literature; and New Literary History, 16, 3 (1985), has similarly explored the politics of literary history.

16. Borges quoted in Walter L. Reed, An Exemplary History of the Novel: The Quixotic Versus the Picaresque (Chicago: Univ. of Chicago Press, 1981), p. 265.

17. Reed, ibid., p. 264.

18. Margaret Atwood, Surfacing (Don Mills, Ontario: Paperjacks. 1972). D. 129