View the H-OIEAHC Discussion Logs by month
View the Prior Message in H-OIEAHC's January 2003 logs by: [date] [author] [thread]
View the Next Message in H-OIEAHC's January 2003 logs by: [date] [author] [thread]
To: H-OIEAHC <H-OIEAHC@H-NET.MSU.EDU> Reply-to: email@example.com X-Accept-Language: en-us, en Delivered-to: H-OIEAHC@H-NET.MSU.EDU User-Agent: Mozilla/5.0 (Windows; U; Win98; en-US; rv:1.0.2) Gecko/20021120 Netscape/7.01 X-Authentication-Info: Submitted using SMTP AUTH at out003.verizon.net from [188.8.131.52] at Thu, 23 Jan 2003 10:52:29 -0600 Original-recipient: rfc822;firstname.lastname@example.org from Don Williams , email@example.com On December 14 2002, the New York Times announced that Columbia University had rescinded the Bancroft award given to Michael Bellesiles. In the previous weeks, even Bellesiles' former allies --prominent historians who had cited Bellesiles' work to support their gun control arguments in the Chicago Kent Law Review, the Constitutional Commentary, and the US vs Emerson Yassky brief -- had either seemed silent or had thrown in the towel. A Nov 20 Chicago Tribune article quoted Jack Rakove of Stanford as follows: "It's clear now that his [Bellesiles] scholarship is less than acceptable," Rakove said. "There are cautionary lessons for historians here." (I blinked when I read this, given that page 583 of Arming America has the following acknowledgment: "Jack Rakove kindly went through the second draft with a keen eye and improved every page he read.") In several past H-OIEAHC posts, I have criticized Bellesiles' "Arming America" for what I see as false and misleading depictions of the early militias. Yet I find myself in the ironical position of wondering whether this much-vaunted "academic process" is being unfair to Bellesiles. On the one hand, I think that the Emory Investigation ignored much that is questionable within Arming America -- Clayton Cramer's extensive list of questionable citations, for example. On the other hand, the Investigation did not identify where Bellesiles is probably right, partially right, or merely mistaken. Even the area addressed by the Investigation --the probate studies -- was handled in what I consider a superficial manner. Professor Lindgren's criticisms are certainly valid but the Investigating Committee failed to explore whether there were reasons why Bellesiles might have obtained results different from those of Alice Hanson Jones. A more extensive check of probate records in the coastal counties for the 1776-1790 timeframe might have shed more light on this issue. For example, in a Feb 14 2002 H-OIEAHC post , I put forth the (unproved) hypothesis that Bellesiles' probate study might have an unrecognized bias. Briefly, a number of counties sampled by Bellesiles (Arming America, page 445, Table 1) were in high population coastal cities occupied by the British during the Revolutionary War. Probate records for the time during and after the Revolution may have shown low firearms ownership because private firearms in those indefensible areas were either seized by the British or were moved out of the area/impressed for the Continental Army to avoid seizure by the British. In either event, those coastal areas may have been "deserts" for private firearms ownership for decades afterward (post-war depression,etc.) Inland areas , by contrast, may have had a high percentage of firearms ownership. However, the high-population of the coastal "desert" areas could have exerted great weight on Bellesiles' overall regional percentage figure. The wartime effect I speak of, if it existed, would have hit records from 1776 to 1790. The prepondence of records Bellesiles checked would probably have been in that timeframe, both because of greater population in 1780s vice 1760s and because of a greater number of deaths due to the war, wartime smallpox epidemic, etc. By contrast, the probate studies which Mr Lindgren used in his William and Mary Law Review Article on Arming America were as follows: A) Alice Hanson Jones /1774 period /919 inventories B) Providence, Rhode Island / 1679-1726 period / 149 inventories C) Gunston Hall sample of Virginia-Maryland /1740-1810 period /325 inventories D) Essex , Mass / 1636-1650 period / 59 inventories E) Gloria Main'study of Maryland Estates /1657-1719 period / 604 inventories F) Anna Hawley's study of Surry County VA estates/ 1690-1715 /221 inventories G) Male inventories from Vermont / 1773-1790 /289 inventories H) Judith McGaw's study of New Jersey/Pennsylvania /1714-1789 /250 estates Of the above probate samples, note that A,B, D,E,F would NOT have shown the wartime effect I describe. Sample G would not have shown it because it was not in a coastal area. Samples C and H may have shown the wartime effect to a SLIGHT degree (because a small percentage of sample period falls within 1776-1790 timeframe and because part of the samples fall within coastal areas.) The difference between pre-1776 conditions and conditions in 1776-1790 might explain how Bellesiles got results different from those of other researchers. Thomas Jefferson, in his "Notes on the State of Virginia" described conditions in Virginia circa 1781-1782: "The law requires every militia-man to provide himself with the arms usual in the regular service. But this injunction was always indifferently complied with, and the arms they had have been so frequently called for to arm the regulars, that in the lower parts of the country they are entirely disarmed. In the middle country a fourth or fifth part of them may have such firelocks as they had provided to destroy the noxious animals which infest their farms; and on the western side of the Blue ridge they are generally armed with rifles. " (Source: http://www.yale.edu/lawweb/avalon/jeffvir.htm , search down for "Military" ) Jefferson's comment is important because in 1790 Virginia was by far the largest State, both in total population and in number of white men of militia age (16+ years). Virginia was much larger than most other states. (see http://fisher.lib.virginia.edu/cgi-local/censusbin/census/cen.pl?year=790 and make selections ). On page 445 of Arming America, Bellesiles lists the Virginian counties which he sampled. If one looks at the county map of Virginia (see http://quickfacts.census.gov/qfd/maps/virginia_map.html ) , several things are noticeable: The chosen counties are all in the "lower country" -- eastern to southcentral part of the state -- plus 5 out of 8 are clustered together. It seems to me that this choice would have given Bellesiles biased results indicating a rate of firearms ownership far lower than what was really the case for the time period -- because the areas where Bellesiles sampled were areas in the Tidewater-Eastern Piedmont region where , according to Jefferson, private firearms had been taken for use by the Continental Army . These firearms were probably never returned to their original owners but were instead probably stored at West Point after the war with the other Continental army weapons (French donated muskets,etc). Hence, probates taken in the southeastern part of the state would have shown below normal levels of firearm ownership for decades after the Revolutionary War. The guns existed at the start of the Revolution -but they had been moved elsewhere. By contrast, samples taken in the Virginia counties in the western Piedmont and west of the Blue Ridge Mountains would probably have shown much higher rates of gun ownership. (Around this period, there was a chain of frontier forts in southwestern Virginia (Tazewell County ) to deal with Shawnee raids coming across the extremely rugged Appalachian mountains. Militia records for this time speak of pioneer families repeatedly fighting off Indian raids with private firearms. Men from this region were part of the King's Mountain raid that destroyed Cornwallis' left wing in 1780.) The reason why guns were moved out of Virginia's eastern Tidewater region during the Revolution is that the area was impossible to defend. Several large rivers penetrate the area from the coast to about 80 miles inland --which would have allowed easy egress to British warships. Any Continental units deployed on the Tidewater's penisulas would have been constantly in danger of entrapment between British landing parties in their rear and British forces on the coast. (Note what happened to Cornwallis at Yorktown. ) If Bellesiles' sampling in the Tidewater "desert area" yielded below normal rates of firearms ownership, then this could have skewed his overall results because of Virginia's relatively high population -- hard to tell without further definition of sample composition. The same bias is probably found in some of Bellesiles' other samples : a) The sample county chosen for South Carolina, Charleston, largely consists of Charleston city. It's private firearms would have been confiscated by the British when they seized the city from General Lincoln circa 1780. (Southern component of Continental Army was trapped and forced to surrender as prisoners.) b) Chatham county in Georgia would have been disarmed when the British seized Savannah. c) Philadelphia county would have been disarmed when the British occupied Philadelphia. Also, the high level of population in Philadelphia would have meant a high number of probate records relative to rural counties like Westmoreland. d) Suffolk County in Massachusetts would have been disarmed by the British occupation of Boston (again, either because private firearms were seized by British or because they were taken by the Continental Congress and moved elsewhere to prevent seizure by the British.) All of these "desert areas" were areas of high population -- which meant they probably made up a significant portion of Bellesiles overall sample set. By locating his limited sampling in those high population "desert areas", Bellesiles probably got unrealistically low readings for the statess as a whole. Note that this bias would not necessarily have been the result of conscious design by Bellesiles -- lacking proof, we are obliged to assume that he was unaware of it. I myself wondered initially if he might have "cooked the data" --but I have since concluded that he probably was unaware of the bias. The above explanation would have been a better defense than whatever explanation he provided to the Emory Committee --if he was aware of the wartime effect he would probably have cited it to the Committee. Only further study of the probate records can show if I'm right or wrong ---and I'm not a professional historian. The primary criticism to my hypothesis when first posted here was JL Bell's persistent scepticism that the British actually carried off private firearms (confiscated by General Gage) when they evacuated Boston. Since our discussion in February, I have found a report written by Washington's quartermaster Mifflin, giving the inventory/status of military stores in Boston after the British evacuation. The report has numerous reports of British attempts to sabotage military equipment left behind --trunnions knocked off of cannon, cannon and balls dumped in Boston harbor, ships scuttled, barrels of flour spilled on the ground, etc. There is no mention of the private firearms confiscated by General Gage being found.