In lieu of an abstract, here is a brief excerpt of the content:

  • Correspondence
  • Daniel A. Cohen

To the Editor:

In his recent review of my book, Pillars of Salt, Monuments of Grace, and Ronald Zboray’s A Fictive People, Christopher Morris constructs his analysis around a postulated polarity between “social” and “intellectual” strands in the “new cultural history.” While Zboray has the relative good fortune to be assigned to the “social” category, my own work is consigned to the “intellectual” realm which, as construed by Morris, consists primarily of a naive and uncritical focus on texts without sufficient regard to social contexts. In trying to make my book conform to that caricature of intellectual history, Morris repeatedly misrepresents my methods and findings in ways that will be fairly obvious to anyone who reads the book. In this response I will document just a single example of Morris’s habitual inaccuracy and then address a broader methodological issue raised by his review.

In contrasting my book to Zboray’s, Morris flatly claims that “Cohen makes no effort to discern who in fact read what, why, and under what circumstances” (p. 254). Actually, my book deals extensively with issues of readership and reader reception, particularly in the opening overview chapter and occasionally thereafter. As noted in the preface, my claims concerning readership are “based on a variety of indicators, including prices, numbers of editions, sizes of editions, advertisements and choice of advertising vehicles, diary references or owner signatures, and hints or references within the texts themselves” (p. ix). A quick scan of the book indicates discussions of readership or reception on the following pages: ix, 4–6, 10, 18, 35–38, 121, 14647, 162, 177–78, 180–81, 195, 207, 212–13, 231–32, 235, 259–60 notes 17 and 20, 267–68 note 85, 278–70 notes 171, 173, and 174, and 327 note 53. While Morris is certainly entitled to argue that I should have done something more to explore the readership of early crime literature, his unqualified claim that I make “no effort” in that area is simply inaccurate.

Near the end of his review, in summing up his preference for social over intellectual history, Morris declares that he is “more comfortable when thought is discerned from behavior and context, rather than the other way around. Admittedly, that is far from satisfactory, but we can know behavior more certainly than we can read the minds of masses of dead people, so I prefer to start there” (p. 257). The implication is that I — or intellectual historians generally — are guilty of trying to discern “behavior and context” from “thought,” whatever that means, and of trying to “read the minds of [End Page 192] masses of dead people.” Underlying those rather cryptic charges seems to lie some fairly serious epistemological confusion.

Whether trying to understand past “thought” or “behavior,” virtually all historians of early America (with only the partial exception of material-culture specialists) necessarily rely on written texts or documents of one sort or another. That is as true of most social historians as it is of intellectual historians; neither group typically has direct access to past experience unmediated by words, or other signifiers, on paper. Further, such popular crime genres as last speeches (short criminal autobiographies) and trial reports (transcripts or synopses of judicial proceedings) purport to be — and generally are — records of “behavior” as much as embodiments of “thought.” To the degree that I attempt to infer “behavior” and “context” from such crime genres (and I do so quite often), I am doing nothing fundamentally different than Christopher Morris when he attempts to infer the experiences of the residents of Warren or Madison County, Mississippi, from documentary sources like newspapers, census returns, and county records. In either case, the historian must try to understand the conventions governing the type of text or document that s/he is using; determine the personal biases and idiosyncrasies of individual authors or compilers; and, perhaps most important, critically compare as many different types of relevant texts or documents as possible, both in order to determine the relative reliability of individual texts and genres, and to piece together as rich and persuasive an account of past behavior and experience as possible. In establishing the social...

Share