View the H-OIEAHC Discussion Logs by month
View the Prior Message in H-OIEAHC's May 2002 logs by: [date] [author] [thread] View the Next Message in H-OIEAHC's May 2002 logs by: [date] [author] [thread]
X-Sender: Hbrewer To: John.Saillant@wmich.edu Reply-to: hbrewer@social.chass.ncsu.edu X-Mailer: NIMS ModWeb Module Original-recipient: rfc822;John.Saillant@wmich.edu As a way of examining the charges against Michael Bellesiles _Arming America_, I did my own experiment with masters' students in two classes I was teaching on historical research, methods, and writing this spring. The experiment had four parts (and lasted four weeks): First, they had to read Bellesiles' book, which we then discussed. Second, they had to look up recent secondary articles and reviews, from professional journals to internet exchanges, and bring them to class to discuss. Third, they had to track down one of his footnotes to the primary source he cited, which they had to xerox, bring to class, and comment on. The particular questions they had to focus on for the footnote project were 1)did he quote and cite accurately? 2)did he interpret it fairly, given its context? In class, we went over his tables, the problems with probate records and methods of quantitative analysis. Fourth and last, they had to write a paper analyzing an issue raised by Bellesiles' book. Although I am a colonial historian, it was not an issue I had heretofore followed closely. After reading Bellesiles' book carefully myself, going over all of the students' findings and papers and the articles in the _William and Mary Quarterly_ (for January 2002), I have formed a considered judgment of the book. Although _Arming America_ has problems, its argument that gun ownership was far from universal among adult males is basically sound. Bellesiles' was sloppy in some places, and sometimes overstated his case (e.g. clearly guns could be accurate sometimes and were fearsome weapons). Guns also did appear in more probate records than Bellesiles' calculations indicated, especially in some areas (as shown, for example, by James Lindgren, Gloria Main). Bellesiles' most egregious quoting problem (here one of my students tracked the George Washington quote that has been emphasized on some internet sites) was to mis-characterize a comment that Washington made about only a few of Virginia militias so that he seemed to be saying it about all (p. 157 of _Arming America_, recent paperback ed.). But generally, most of his quotes were fine. He did tend to lean towards one kind of interpretation, but many scholars do that. I was particularly fascinated by the probate issue, especially in terms of the questions it raised about quantitative methods. Clearly, Bellesiles should not have claimed them to be as representative as he did. Yet his critics (e.g. Lindgren) are equally at fault in focusing on them so much. Quantitative analysis for this period of history is almost by definition deeply problematic. Not only were probates not kept in any consistent manner, but probates reflect only a minority of the population: most of the wealthy and some of the middling sort. They do not capture most poorer and other middling estates. By definition, probates can only provide part of the answer: we have to look to non-quantitative sources to piece together a fuller response to the questions of gun ownership. It is at that point that Bellesiles' study is most convincing (Don Higginbotham, a specialist on military history of this period underscores, in conversation, that he has encountered the same evidence): over and over again, despite statutes that often required those eligible for militia service to own guns, they showed up at muster without one. While we can argue about this evidence as well, it suggests strongly that gun ownership was far from universal among the poorer and middling sort in Colonial America. When this evidence is coupled with the evidence about the price of guns (and more than that, about the skilled labor it took to produce them---more than a month, in most cases), then it becomes clear that these poor showings by militia men occurred, at least in part, because guns were expensive. The broader issue, to me, is one that we historians have been studiously avoiding: it is about reliability in a world where we are expected to produce ever broader studies ever more quickly to move forward in the profession. The quality of editing has gone down, especially with a commercial press such as that Bellesiles' used, but also even with academic presses. To my knowledge (and I have asked many scholars about their recent publishing experiences) no one except the Omohundro Institute of Early American History and Culture, which still produces books slowly and thoroughly--a process I am going through, agonizingly slowly, at present-- actually checks footnotes any more: it is up to the author. The same questions, about coverage and pushed publication, relate directly to Bellesiles' method of collecting his quantitative data on guns in probate records. We have been sidestepping fundamental questions about quantitative methodology: Gloria Main, in her WMQ article, was particularly sharp about Bellesiles' sloppiness with quantitative records; at the same time, she questioned how he could have surveyed 11,000 records. The point is that when an historian examines old records such as eighteenth century probates, she or he can take two different approaches: Main's, which is to examine a small group intensively, entering everything meticulously into databases; or Bellesiles', which is to examine a huge body of records very quickly, making ticks to record incidents. By definition, Bellesiles could not have employed Main's methods and have covered 11,000 probates. We historians want to have our cake and eat it too: we want coverage but also meticulous reliability. Without a fleet of full-time research assistants--and indeed, even with those, Bellesiles could not have done what Main expects for 11,000 probates. It is exactly here, in fact, that the San Francisco probates enter in. Working hurriedly through the probate records for a county near San Francisco, he noted the county as San Francisco itself (he claims that the file was mislabled, but whether it was or not is minor). Then, in his poorly documented table (his commercial press probably urged him to minimize his citations) he wrote San Francisco as one of the many cities for which he had examined some probates. That is the only place (Table I in his Appendix, p. 445) that he referenced those probates. In short, Bellesiles tried to pull off a grand book on a medium budget, with mediocre editing, and with insufficient time. It needed a good, old-fashioned editor, one who would check his footnotes and help him refine his prose and arguments. But in the modern world of short print- runs (partly because of a Reagan-era tax on publishers' inventory), slashed budgets for academic presses, and professional expectations about "quick books" this is where we, as a profession, are being led. The last question--and I have to pose it--is this. Is Bellesiles' being roasted only because of the political implications of his book? Are other books by historians--and non-historians--as sloppy in these (most of them mild) ways? The answer, of course, is yes, and books with slight misquotes and mistakes and misrepresentations have been published for decades, and, indeed, centuries (I remember my surprise when checking some footnotes from the Robert E. and B. Katherine Browns' book on Virginia as a democracy, where they had completely misrepresented the meaning of a primary source). Of course, there are other books that are meticulously checked, by their authors if not their editors, but those take longer to produce, and often (perhaps usually) do not have the grand scope that Bellesiles was aiming for. Holly Brewer Early American History North Carolina State University Holly Brewer Early American History North Carolina State Univ. Raleigh, NC 27695-8108 (919) 832-7638
|