Bellesiles has dispersed the darkness that covered the gun’s early

history in America. He provides overwhelming evidence that our

view of the gun is as deep a superstition as any that affected Native

Americans in the 17th century.

—Garry Wills, New York Times1

 

 

Before there was a scandal, there was a book—Michael A. Bellesiles’s

Arming America: The Origins of a National Gun Culture. In this Review, I

not only discuss the book, benefiting from some of the substantial published

and unpublished literature on it, but review a little of the controversy—at

least the controversy as I understand it at the beginning of 2002.

 

Let me state my biases up front: I dislike guns; I have never owned a

gun; I have not touched one since the age of nine. Yet I don’t understand

the passion that people bring to the issue of their regulation. My own prior

writing on guns has been on the pro-gun-control side of the dispute, and

some of it is so free from passion as to be soporific.2

 

Arming America is a well-written and compelling story of how early

Americans were largely unfamiliar with guns until the approach of the Civil

War. It tells a wide-ranging, detailed, but relatively unnuanced story of

gunlessness in early America. Bellesiles writes: “[T]he vast majority of

those living in British North American colonies had no use for firearms,

which were costly, difficult to locate and maintain, and expensive to use.” 3

 

According to Bellesiles, in seventeenth-, eighteenth-, and early

nineteenth-century America there were very few guns.4 Privately owned

guns were mostly in poor working condition.5 By law, guns were not kept

in the home but rather stored in central armories,6 and guns were too

expensive for widespread private ownership.7 He even claims that men

generally were unfamiliar with guns and that they did not want guns8—

preferring axes and knives instead, in part because guns were so inaccurate

that they were of little use. He argues that few settlers hunted,9 and implies

that axes made very good weapons in hunting.10 According to Arming

America, in battle “the ax [was often considered] the equal of a gun.” 11

 

Bellesiles claims that states enacted laws that restricted gun ownership

to white Protestants who owned property.12 White-on-white homicide was

rare in colonial America, according to Bellesiles, and guns were rarely the

weapon used in homicides.13 Guns were not culturally important, either:

Travel narratives do not show that guns were part of everyday life,14 even

on the frontier, and few people even wanted to own guns.15 At least in

probate records, women did not own guns.16 Since there were few guns, the

laws passed in the early nineteenth century restricting the right to carry

concealed weapons were directed at knives,17 not guns. He further claims

that the background of the Second Amendment shows that the Anti-

Federalists had no problem with restricting militia membership to those

above the lower social classes.18 Last, with a few exceptions, the militia

were extremely ineffective.19

 

Two meta-arguments by Bellesiles might have direct public policy

applications (though, as a work of history, Arming America does not

directly advocate any gun policies). One is that guns and violence go

together. In early America, he claims, we had very low gun ownership and

low homicide rates; after the Civil War, we had lots of guns and high

homicide rates.20 The second is that if guns were not widely owned, then it

is unlikely that gun owning was understood as an individual right in the

Second Amendment.

 

Since the book’s publication, scholars who have checked the book’s

claims against its sources have uncovered an almost unprecedented number

of discrepancies, errors, and omissions. When these are taken into account,

a markedly different picture of colonial America emerges: Household gun

ownership in early America was more widespread than today (in a much

poorer world).

 

Arming America is changing the way that some historians think about

their own profession and how some scholars in fields allied to history

regard historical research and publishing. Understanding this book and the

scandal it generated is important for scholars and teachers across the social

sciences, humanities, and law. Any graduate or professional student who

aspires to be an academic might profit by exploring the twists and turns of

the Bellesiles scandal.

 

 

I. BEFORE THE BOOK

 

In 1996 a well-regarded, but relatively obscure, historian at Emory

University, Michael A. Bellesiles, published an article in the Journal of

American History (JAH).21 It urged a mostly novel thesis about early

America—that there were few guns and that there was no gun culture until

the approach of the Civil War. His primary evidence was low counts of

guns in probate records, gun censuses, militia muster records, and homicide

accounts.

 

The data fit together almost too neatly. In particular, if anyone had

looked closely at the probate data, they would have seen that it did not look

right. The regional differences were suspiciously slight; the increases over

time were extremely regular; the study did not indicate which counties were

in which categories; and in most unconventional fashion, the probate data

were published with no sample or cell sizes. The results were directly

contrary to the existing literature counting guns in probate records,22

including one source Bellesiles cited but did not discuss,23 all of which had

found substantial numbers of guns.

 

Last, the 1765-1790 data were mathematically impossible if there were

more than about 200 cases in his sixteen Southern counties over the twenty-six-

year period,24 which any scholar familiar with probate records would

have known had to be true many times over. If the JAH had insisted on cell

counts (which would have been conventional), the impossibility of the

1765-1790 data would have been fairly obvious.25 This entire scandal might

have been avoided in 1996 with more conventional editing at the JAH.

 

The response by historians to the 1996 JAH article was varied. At a

meeting of the Crime and Justice Network of the Social Science History

Association, historians discussed how such a piece of work could get

through peer review. The consensus was that probably none of the experts

in the room (many of whom were quantitative historians) had been asked to

review it. The Organization of American Historians, on the other hand, had

a different response: They awarded the article the prize for the best article

published in the JAH that year.26 This bipolar response to Michael

Bellesiles’s work on guns continued until recently—those who are most

expert on the subject of guns in early America or tend to understand

numbers best were most skeptical about Bellesiles’s work, while those who

know less about guns or less about numbers were most enamored of it.

 

Bellesiles’s surprising thesis had a few detractors online, mostly among

pro-gun activists and scholars unaffiliated with universities,27 but most

historians were impressed. Alfred A. Knopf, perhaps the top nonacademic

publisher of serious books of history, agreed to publish a much-expanded

version of the article. The educated public first learned of the forthcoming

book in a long, positive article in the Economist in the summer of 1999,28

over a year before the book came out. The Economist article was followed

by a similarly positive article in the New York Times in the spring of 2000,

still five months before the book’s publication.29

 

The response to the Economist article was overwhelming. The president

of the National Rifle Association, Charlton Heston, criticized Bellesiles and

his forthcoming book, saying, among other things, that Bellesiles had “too

much time on his hands.” 30 The tone of anti-intellectualism in the NRA

response was patent—and made an easy target for Bellesiles and his

colleagues. Substantively, Heston criticized Bellesiles’s reliance on probate

records, because of their incompleteness.31

 

In what was to become a pattern, Bellesiles responded in two very

different ways—a political response and a response claiming expertise and

care in his work. First, he obtained (or at least received) a public declaration

of support from other professors. A group of forty-seven law professors and

historians signed a public letter to the NRA expressing a moderately pro-

gun-control view.32 Second, Bellesiles made his own statements supporting

his methods. Defending the use of probate records against criticisms of

incompleteness, Bellesiles made some unusual claims. He said that probate

inventories recorded absolutely everything in an estate, even property given

away during life, and that wills recorded gifts given away up until the time

the will was written.33 These statements conflict not only with common

sense, but with what is written by every probate scholar that I have read or

that Bellesiles cites in Arming America.34 When initially pressed about

problems with the probate records, Bellesiles’s response was to defend his

reliance on them more vigorously with claims that plugged potential holes

in his argument. These claims, however, were not only unsupported but

ultimately proved to be false.

 

When Arming America was published in September 2000, it was treated

to some rave reviews. First, it was welcomed to the front page of the New

York Times Book Review with an uncritical review by Garry Wills.35 Then

Edmund Morgan wrote an enthusiastic review in the New York Review of

Books.36 Other positive reviews followed.

 

The only early negative reviews were in conservative, libertarian, or

gun aficionado magazines or websites, most prominently the National

Review37 and Reason.38 By January 2001, an extraordinary number of errors

had been identified in the book and were being discussed on history and

constitutional law discussion lists, including Bellesiles’s claim to have

examined records that did not exist and his use of data that were

mathematically impossible.

 

Nonetheless, apparently without looking into any of these claims, in

April 2001 Columbia University awarded the Bancroft Prize for history to

Arming America, along with two other books. It was not until a year after

the book’s release that the academic journals began publishing some

devastating critiques—by Robert Churchill in Reviews in American

History,39 Joyce Malcolm in the Texas Law Review,40 Randolph Roth, Ira

Gruber, and Gloria Main in the William and Mary Quarterly,41 and Justin

Heather and me in the William and Mary Law Review.42

Click here for the next page.

NOTES:

 

† Stanford Clinton Sr. Professor of Law, Director of the Demography of Diversity Project,

Northwestern University School of Law. Chair, AALS Section on Law and Social Science. J.D.,

University of Chicago; B.A., Yale University; currently Ph.D. student, Sociology, University of

Chicago. This Review was funded by the Searle Fund at Northwestern University. I would like to

thank the many unnamed people who contributed to my understanding of Arming America, but

particularly those whose work is discussed here—Randolph Roth on homicides, Robert Churchill

on militia arms and gun censuses, Justin Heather on edge weapons and probate, Eugene Volokh

on legal history, and Clayton Cramer on travel accounts, gun ownership restrictions, and other

matters. I am also indebted to those who located and carefully examined documents in Contra

Costa County—David Golden, Betty Maffei, Dean McLeod, Kathy Beals, and Kathleen Mero—

and to David Golden and the staff of the History Center for their help during my research visit

there. While this Review deals mostly with issues other than probate records, the probate

discussion in particular is based on James Lindgren & Justin L. Heather, Counting Guns in Early

America, 43 WM. & MARY L. REV. (forthcoming 2002), where most of that discussion first

appeared. I benefited from the comments of participants on that paper at faculty workshops at

Yale, Princeton, Columbia, Chicago, William & Mary, Pennsylvania, Berkeley, Indiana, North

Carolina, and Virginia. This Review reflects the state of the dispute in January 2002, although

there are occasional citations to works published later.

 

* Professor of History, Emory University.

 

1. Garry Wills, Editorial, Spiking the Gun Myth, N.Y. TIMES, Sept. 10, 2000, § 7, at 5.

2. James Lindgren, Organizational and Other Constraints on Controlling the Use of Deadly

Force by Police, 455 ANNALS AM. ACAD. POL. & SOC. SCI. 110 (1981); James Lindgren &

Franklin E. Zimring, Regulation of Guns, in 2 ENCYCLOPEDIA OF CRIME AND JUSTICE 836

(Sanford H. Kadish ed., 1983).

3. MICHAEL A. BELLESILES, ARMING AMERICA: THE ORIGINS OF A NATIONAL GUN

CULTURE 110 (2000).

4. See, e.g., id. at 445 tbl.1.

5. See, e.g., id. at 13, 109.

6. Id. at 73.

7. Id. at 106.

8. See, e.g., id. at 390.

9. Id. at 110.

10. Id. at 313 (attributing to a hunter the statement that axes made very good weapons).

11. Id. at 67.

12. Id. at 74-75.

13. Id. at 81, 353.

14. Id. at 305-22.

15. Id. at 389-90.

16. Id. at 267.

17. Id. at 309.

18. Id. at 223.

19. Id. at 87-88, 140-41, 146-53, 178-79, 182-83, 193-98.

20. See id. at 434, 436.

21. Michael A. Bellesiles, The Origins of Gun Culture in the United States, 1760-1865, 83 J.

AM. HIST. 425 (1996).

22. GLORIA L. MAIN, TOBACCO COLONY: LIFE IN EARLY MARYLAND, 1650-1720, at 242

(1982); Anna Hawley, The Meaning of Absence: Household Inventories in Surry County,

Virginia, 1690-1715, in EARLY AMERICAN PROBATE INVENTORIES 23, 27-29 (Peter Benes ed.,

1987); Judith A. McGaw, “So Much Depends upon a Red Wheelbarrow”: Agricultural Tool

Ownership in the Eighteenth-Century Mid-Atlantic, in EARLY AMERICAN TECHNOLOGY:

MAKING AND DOING THINGS FROM THE COLONIAL ERA TO 1850, at 328, 340 (Judith A. McGaw

ed., 1994).

23. ALICE HANSON JONES, AMERICAN COLONIAL WEALTH: DOCUMENTS AND METHODS

(1978).

24. For a full discussion of this point, see James Lindgren & Justin L. Heather, Counting

Guns in Early America, 43 WM. & MARY L. REV. (forthcoming 2002) (manuscript at 53-54, on

file with author). See also infra note 207.

25. For example, if Bellesiles had listed fewer than 200 estates for sixteen Southern counties

for the twenty-six years 1765-1790, it would have been obvious that the count could not be

correct. There would be more than 200 estates in just a few years of one large Southern county. If

Bellesiles had listed a plausible count of, for example, 3000-8000 cases from the South, then the

overall mean of 14.7% would have been obviously impossible, since he reports only 1200 cases

from the frontier, the only region below the mean. See id. (manuscript at 51-54 & nn.105-13).

26. OAH Binkley Stephenson Award Winners, at http://www.oah.org/activities/awards/

binkleystephenson/winners.html (last visited Apr. 17, 2002).

27. The most thorough and persistent critic since the 1996 article was published has been

Clayton Cramer, some of whose criticisms are confirmed in this Review. See Clayton E. Cramer

& Dave Kopel, Disarming Errors, NAT’L REV., Oct. 9, 2000, at 54; Clayton E. Cramer, Firearms

Ownership & Manufacturing in Early America (Apr. 4, 2001), at http://www.claytoncramer.com/

ArmingAmericaLong.pdf [hereinafter Cramer, Firearms Ownership]; Clayton E. Cramer, Gun

Scarcity in the Early Republic? (Nov. 19, 2001), at http://www.claytoncramer.com/

GunScarcity.pdf.

28. Arms and the Man, ECONOMIST, July 3, 1999, at 17.

29. Anthony Ramirez, The Lock and Load Myth; A Disarming Heritage, N.Y. TIMES, Apr.

23, 2000, § 4, at 3.

30. David Bowman, The Reasonable Gun Nut, SALON.COM, Sept. 7, 2000, at

http://www.salon.com/books/feature/2000/09/07/bellesiles (reproducing a transcript of a taped

interview with Michael Bellesiles). In this interview, Bellesiles stated:

 

I wrote him [Heston] an open letter because he wrote an editorial in Guns & Ammo

attacking my research from a very postmodern perspective: Evidence doesn’t matter.

He said I had too much time on my hands. I pointed out that I write history and what

use people make of it is their business, not mine.

 

Id.

31. Id.

32. Differing Views on the Second Amendment (Apr. 3, 2000), at http://www.kentlaw.edu/

news/advisory/adv000403.html.

33. See Bowman, supra note 30. Bellesiles told Salon:

 

I’d like to know what his evidence is. When Professor Heston gets his Ph.D. and does

the research, I might be open to persuasion. This is one area of law that in colonial

America was far stricter and much more rigorously enforced than it is today. Cheating

on probate was a very great crime because resources were thinly stretched. When

someone died, every single item owned—everything, even broken things—was

recorded. Guns had to be listed. So unless Charlton Heston can come up with evidence

that they made an exception for guns, he should keep quiet. The British Common Law

saw guns as belonging to the state. The state had all priority rights over firearms. They

could appropriate them at any time without recompense. There was actually greater

value placed on recording firearms than any other single item.

 

Id.; see also BELLESILES, supra note 3, at 13, 267, 484-85 n.132 (claiming that gifts before death

were recorded); Ramirez, supra note 29 (explaining that Bellesiles questioned the evidence for

Gary Kleck’s argument that guns would have been passed on before death).

34. Bellesiles is virtually alone among historians who work with probate records in thinking

that they are more or less complete. Compare BELLESILES, supra note 3, at 13, 109, 266, with

Lindgren & Heather, supra note 24 (manuscript at 56-59) (explaining the general consensus

among scholars that probate inventories are incomplete). Bellesiles offers no evidence for the idea

that probate records are so detailed that they record both all estate assets and most lifetime gifts.

Nor does he offer any evidence for the idea that firearms were more likely to be listed in probate

inventories than other items. On both issues, the historians he cites directly contradict his claims.

See Lindgren & Heather, supra note 24 (manuscript at 56-59). Clothes and land, for example,

were frequently omitted. Id.

 

In Arming America, Bellesiles raises few hints that probate inventories are not complete.

There is, however, an eloquent general comment about the limitations of using quantitative

records. BELLESILES, supra note 3, at 262.

35. See Wills, supra note 1.

36. Edmund S. Morgan, In Love with Guns, N.Y. REV. BOOKS, Oct. 19, 2000, at 30.

37. Cramer & Kopel, supra note 27.

38. Joyce Lee Malcolm, Concealed Weapons, REASON, Jan. 2001, at 47.

39. Robert H. Churchill, Guns and the Politics of History, 29 REVS. AM. HIST. 329 (2001).

40. Joyce Malcolm, Arming America, 79 TEX. L. REV. 1657 (2001) (book review).

41. Forum, Historians and Guns, 59 WM. & MARY Q. 203 (2002); Ira D. Gruber, Of Arms

and Men: Arming America and Military History, 59 WM. & MARY Q. 217 (2002); Gloria L.

Main, Many Things Forgotten: The Use of Probate Records in Arming America, 59 WM. &

MARY Q. 211 (2002); Randolph Roth, Guns, Gun Culture, and Homicide: The Relationship

Between Firearms, the Uses of Firearms, and Interpersonal Violence, 59 WM. & MARY Q. 223

(2002).

42. Lindgren & Heather, supra note 24.

---

This review first appeared in the Yale Law Journal (June 2002), Vol. 111, pages 2195-2249 and is reprinted with permission.

Comments (6)