For a part of the field that has only been active for a decade or so, composition history has already generated a considerable list of controversies and disputed issues. The disputes tend to fall into methodological and epistemological areas, although most refuse to fit neatly into one class or another.

An example of a dispute that spans categories is the issue of whether composition historians should write histories of theory or histories of practice. One strong tacit tradition in rhetorical history has been to write theoretical histories, narratives that show how one set of ideas was propounded, criticized, adopted, revised. "Influence studies" is the term often given this sort of scholarship in literature. Some of the most impressive works in rhetorical history have been written to trace and categorize these theoretical influences, most notably Wilbur Samuel Howell's two magisterial volumes on English logic and rhetoric from 1500 through 1800, which cover every rhetorical treatise written during those years but make little attempt to situate rhetoric culturally. The other strong tradition has been to describe rhetorical praxis and education in the context of their times and cultures, paying more attention to the meaning and uses of the discipline than to the content of its theory. George Kennedy's books on classical rhetoric in Greece and Rome and Brian Vickers' recent Defense of Rhetoric exemplify this tradition.

The immense influence of Albert Kitzhaber moved composition history in both directions, but ultimately his work tended to head more toward theoretical than toward cultural histories of composition. Coverage and theoretical analysis of early composition textbooks were Kitzhaber's great strength; like Porter Perrin and Glenn Hess before him, Kitzhaber worked primarily from the artifacts he had gathered and mastered, which were textbooks, and, to a lesser degree, journal articles. Since textbooks remain the largest mine of evidence concerning nineteenth-century composition theory and teaching, and since they remain more easily (although randomly) available to researchers than most other data, and since Kitzhaber had made their use respectable, the early members of the "second generation" of composition historians-primarily Stewart, Berlin, Connors, and Johnson--relied on textbooks to a great extent. As a result, the early work in composition history tends, with the exception of Wallace Douglas' Marxist readings of Harvard records, to be about the theories found in early composition texts and about how those theories evolved.

Even at their most theoretical, however, composition historians have never approached the almost complete lack of interest in culture and practice seen in some theoretical histories of rhetoric. They have remained involved. The main reason for this practical and cultural focus in composition history has been the fact that most composition historians are also writing teachers. They are immediately implicated in their subject. Speech historian W. S. Howell could afford a certain distance toward his discovery that Adam Smith in 1749 approached rhetoric with more acumen than John Holmes did in 1755. The fact may have been interesting, but the work of neither of them affected Howell and his daily world of professional reality very much. From Kitzhaber on, however, composition historians have never had the luxury of scholarly distance. They exist, as composition specialists have for a century, in a world of complex social and institutional problems whose solution is writing teachers' charge, and thus even the theoretical and "textbook" histories of the field have always been implicitly polemical. We followed Kitzhaber not only to his sources, but also to the sometimes savage indignation about current conditions that often characterized his work.

So Kitzhaber's methodological legacy has been two-pronged; early composition history had a tendency to look at the past in terms of its theory and textbooks, and it had a distinct tendency to view the past through the sometimes narrow lens of how it seemed to affect the troubled present. We found the works of our heroes--Fred Newton Scott, Gertrude Buck, J oseph Denney, Franz Theremin, Henry Day, C. S. Baldwin, Sterling Leonard, Porter Perrin--and celebrated them. We traced the elements in composition teaching that we thought were questionable back to their lairs in the works of our pantheon of villains and dupes--Samuel Newman, Richard Whately, Alexander Bain, Adam S.

Hill, Barrett Wendell, John Genung, Edwin Woolley, John Warriner, John Hodges. This early work was clearly sided history.

The perspective on American composition taken in most of these analyses is easy to understand in retrospect: historians writing in a troubled present constructed a narrative of its genesis based on sources at hand. Thus the metanarrative of most work before 1984 or so might have the subtitle "Decline and Fall." It was a tragic tale of bad theory driving out good, of the loss of the liberal tradition in rhetoric, of calculating, hegemonic Harvard taking over the rhetorical world, of a noble Fred Newton Scott fighting a hopeless rear-guard action against encroaching barbarisms like "grammar" and "workbooks." It ended with the ugly triumph of Bain's formalism over Emerson's and Theremin's idealism and with the onset of our current Iron Age, where until recently the lamp of rhetorical humanism guttered low.

This was a rattling good story, and in certain ways it is even an accurate one. But it was not the complete story, and work in composition history since 1985 has been struggling to add some depth to the alltoo-simple tale of Decline and Fall. The essential problems with the old narrative are, first, that it ignores or discounts too much information we now have, and, second, that it does not look deeply enough into the social, cultural, and ideological contexts of rhetoric and composition as they developed in their own eras. For instance, it was a natural and necessary step to trace paragraph theory back to Alexander Bain in 1866 and then to show how his theory is not accurate or useful according to our current knowledge. The harder task confronting historians now is to draw the analysis out in deeper and stronger ways. What led Bain to this theory? How does it relate to changing ideas of English prose style? Why were teachers attracted to it? What do formal theories suggest about pedagogic attitudes?

Perhaps we see the most important sign of maturation in historical research in the crumbling of the simple heroes-and-villains narrative. While no one would deny Fred Scott his eminence, we no longer see his work as the touchstone of all that is True and Good. And after more than thirty years as everyone's bête noire, Harvard's A. S. Hill is being seen as the much more complex thinker and actor that he was. We are looking at a broader range of sources and learning that our metanarrative has been too simple. While I would hope that composition history never completely loses touch with the dissatisfaction that fueled its earliest works with fervor and gave meaning and passion to its narratives, our work is richer now. Historians' growing awareness of the causal complexities and sociocultural motivations that are as important as any theoretical history to the development of our field can only make sharper our awareness of current conditions and make more realistic our hopes for solving contemporary problems through understanding them.

Another set of current issues in the writing of composition history has to do with the development of research and with the presentation of data after they've been found. It's our version of the old Platonic/Aristotelian debate about deduction and induction. Some historians tend to see, to research, and to present their findings within overt and carefully created frameworks of meaning, and others eschew this approach. Of contemporary historians, James Berlin is probably the best-known "framework" researcher. In each of his books, Berlin creates a taxonomic structure and follows its implications by fitting various figures and works into it; in Writing Instruction in Nineteenth-Century American Colleges his taxonomy breaks nineteenth-century rhetoric into classical, psychological-epistemological, and romantic types, while in Rhetoric and Reality he classifies twentieth-century movements into objective, subjective, and transactional classes. Wallace Douglas' Marxist analyses, which use an existing class-structure perspective, are also examples of "framework" research.

Other historians opt for a more inductively derived historical narration, one that takes up a problem at the "beginning"--the first place their research can discover it--and follows it through to contemporary times, or that traces the work and influence of one figure. Paul Rodgers on the Bainian organic paragraph, or David Russell on writing across the curriculum, or my work on the development of handbooks are all examples of problem- or figure-based history.

Both the overt framework histories and the problem- or figure-based works are, of course, subject to the criticism that they present narratives based on a priori viewpoints that control and constrain the research beneath them. Berlin chooses his classes, then seeks evidence that reifies them; I start with an a priori definition of handbooks and only look at books that fulfill that definition. I cannot think of any work of composition history that cannot to some degree be accused of this sort of a priori subjectivism; historians disagree mainly on how much of it exists. In a review of Writing Instruction in Nineteenth-Century American Colleges, I once critiqued James Berlin for filtering his research effort through powerful terministic screens and for failing to take important alternate perspectives into consideration. Berlin's reply was that such screens are inevitable in any research project, and that since objectivity is impossible, any historical research project is automatically interpretive and thus radically subjective; all a historian can do is try to be aware of the terministic screens that exist for him or her.

This whole issue of how and when a researcher reaches closure--of where the data she is discovering begin to assemble themselves in her mind into a structure that will form a thesis claim and will then inevitably guide and constrain subsequent research--leads to the great historiographic question of facts versus interpretations. Some of the field's major historians met in an "octolog" at the 1987 CCCC to discuss these complex issues, and in spite of some real methodological disagreements there was surprising agreement from most of the participants. Of the eight panelists discussing "The Politics of Historiography," seven were working historians, and one, Victor Vitanza, was a historiographic theorist. When the dust settled, it was clear that: (1) No one on the panel believed that any "objective" or definitive history was possible or even desirable; all believed that multiple histories are possible and desirable; (2) The seven working historians all believed, tacitly or explicitly, that "evidence" or "data" or "sources" or "historical materials" were essential stuff of their day-to-day researches. Of the eight panelists, only Vitanza valorized language and its infinitely regressive possibilities as the central component of history. ("Is there any evidence for evidence?" he asks.)

That there was so much tacit agreement in a panel specifically convened to air disagreements is less surprising when we consider that the participants were rhetoricians as well as historians. They all expressed an essential rhetorical position: that assent could be based both on inartistic proofs--evidence--and on artistic proofs--the perspective, the method of presentation, the language. The working historians all accepted the concept that evidence must be searched for and weighed, that prejudices must be taken into consideration, and that induction and deduction were both necessary parts of historical research and writing.

The entire question of historiographic theory animating this 1987 octolog has been brought to the fore in the last five years largely through the efforts of Victor Vitanza, who has used his journal Pre/Text to advance various radical critiques of current historical works and practices. Vitanza has attracted a brilliant group of younger scholars to the journal, and one of their primary interests has been in historiographic issues. Such scholars as Susan Jarratt, John Schilb, Jan Swearingen, and James Berlin have weighed in with historiographic articles in Pre/Text. Although only Schilb (and, to a lesser extent, Jarratt) might be said to hold many ideas in common with Vitanza, the general effect of Vitanza's efforts has been to valorize historiographic questions. Since 1986, the theoretical and epistemological issues surrounding the writing of history have been much discussed.

The proponents of "revisionist historiography," as it has come to be called, fall roughly into two camps: those who seek to promote a specific program or perspective, and those who point out the incompleteness, potential for totalization, or naivete of any specific program or perspective. Into the former camp might fall Sharon Crowley and Jan Swearingen, who have both been engaged in recovering heretofore marginalized figures in rhetorical history: the sophists and women. Here, too, we find James Berlin and Wallace Douglas, who take the perspective of neo-Marxists and argue for dialectical history and classbased historical analyses. The implicit program of this group is actionbased; they make the claim that traditional histories are biased, or incomplete, or controlled by sexist or racist or class purposes.

The other group of revisionist historiographers are the epistemological radicals, primarily Victor Vitanza, John Schilb, and Susan Jarratt. (It is interesting to note that with the exception of Jarratt, who is the least radical and most obviously "political" of the group, the epistemological radicals have themselves primarily written critiques of historical writing rather than history itself.) The critiques coming from this end of the table mainly descend from the interpretive issues popularly argued over in literary criticism during the past fifteen years: the undecidability of meaning in texts; the aporias that riddle every text and source; the interplay of social conditioning and understanding or ordering of meaning; the hegemony of linearity and "clearness" as criteria for worth in historical writing; the utter lack of support for any concept of objectivity. In radically undercutting meaning systems, Vitanza is probably the most extreme; like his eidolons Deleuze and Guattari, his own prose style is deliberately playful and hallucinatory, and his suspicion of all proffered meaning-systems as totalizing and potentially fascistic makes him the great epistemological anarchist in the field.

Thus far the working historians have tended to react to the Vitanza position with a mixture of humor, discomfort, and distracted annoyance. We cannot supply proofs for proof, in Vitanza's terms, but thus far no historian has been willing to allow the theoretical uncertainty underlying his or her making of meaning to close down the enterprise. We may argue about the relative power of facts versus interpretations, but finally the community of working historians feels constrained by and dependent on both. All we can do is continue to be aware of the necessary balance between induction and deduction in any research enterprise, trying to avoid totalizing perspectives that force us to closure too early in the research process. No historian sets out deliberately to twist the truth, but Vitanza's critique remains a salutary reminder that our natural prejudices constantly create terministic screens that control what we see--and can control what we look for as researchers.

The writing of the history of composition is still at a very early stage. Much remains to be done. We need to continue looking closely at the connections between rhetoric and writing instruction--indeed, at all the issues surrounding the relation of orality and literacy. We need much more work on the period 1790 to 1850, which still remains the subject of only a few articles and dissertations. We still know very little about the teaching and learning of writing outside the United States and Canada; the other English-speaking countries are only beginning to be examined. We need to articulate our knowledge, to connect college issues with the increasingly detailed historical work being done on elementary and secondary schooling. We also must put our research ever more strongly in context by making ourselves aware of the larger issues of class, gender, race, and franchise that have always been the "silent" realities behind college education. The larger issues of literacy and power, which have begun to appear in historical works of the last few years, will be inescapable for historians in the future. Though we may wax nostalgic for the simple days when the Decline and Fall narrative provided continuity, when textbooks clumsily mated and bred without sociocultural influence, and when neat taxonomies made everything understandable, that is not how we think about things anymore.

Historians of composition in the future will need to be both peripatetic and widely read. The primary sources are out there, and finding them--especially the ephemeral pedagogical materials and the almostas-ephemeral student papers--will be a challenge. We will need to evolve serious collections and depositories of composition materials. In addition to being scholar-gypsies, composition historians of the future will need to immerse themselves in collateral reading about their subjects and periods, as good rhetorical scholars always have. We cannot understand the teaching of writing in 1870 without understanding the causes of the Civil War; we cannot understand the "American English" movement of the 1940s without understanding the McCarthy era. We are ineluctably tied to the movements of our cultures, and as rhetoricians we have to watch the signals. Only then will we write histories truly informed by all the good evidence needed to gain a hearing from an increasingly skeptical discourse community.