Mandatory Gun Ownership, the Militia Census of 1806, and Background Assumptions concerning the Early American Right to Arms: A Cautious Response to Robert Churchill

In “Gun Ownership in Early America,” published in the William and Mary Quarterly in 2003,[1] Robert Churchill drew on probate inventories and militia records from the  to make the case that arms ownership was pervasive in late colonial, revolutionary, and early national America. Churchill concluded with the observation that “[i]t is time to ponder what these guns meant to their owners and how that meaning changed over time.”[2] In his substantial contribution to this volume of Law and History Review,[3] Churchill takes up that challenge himself and advances the claim that widespread arms ownership engendered a sense of possessory entitlement, and that this notion of right informed constitutional sensibilities respecting guns and the Second Amendment. He acknowledges that a civic republican understanding focused on the militia was central to the framers’ conception of the right to arms, but urges that another stream of discourse—individualistic, personal, and divorced from militia linked obligations—was present from the beginning. By the early nineteenth century, Churchill argues, this purely private view of the right to arms had become ascendant.

Churchill’s most intriguing claim is that arms possession (in large measure because of its alleged ubiquity) acquired an aura of immunity against at least some assertions of government power during the period in which the Constitution and Bill of Rights were drafted and ratified. In colonial times, says Churchill, provincial and imperial military authority extended to seizure of guns for purposes of arming militia and ensuring security, but by the 1780s, these statist claims against privately held weapons were abating, never to return. The police power still allowed civil authorities to regulate firearms usage to preserve safety in towns and on public roads, promote public decorum, and protect the population of game, but assertions of governmental power to seize (rather than merely regulate) guns rapidly petered out in the new nation. Perhaps like other royal prerogatives that died a death of desuetude in the Whigish narrative of English history, the abeyance of governmental authority to confiscate arms begat a negative liberty against gun seizure, and this liberty soon took up a prideful place in the orthodox (or at least popular) understanding of constitutional rights. Churchill’s argument is interesting, in several respects novel, and in many ways enlightening. But it is by no means clear that he accurately captures all the evidence on which he relies, or that his thesis can fully account for some important evidence that he glosses over or ignores. In fact, vital material Churchill misreads or omits points squarely back to the civic-republican focused reading of the constitutional right to arms he aims to play down or read away.

Consider, for instance, Congressional inquiry into the arming of the militia. On April 2, 1806, Joseph Varnum, then a six-term Republican Congressman from Massachusetts and major general in the Commonwealth’s militia, presented a report from the “committee instructed to inquire what measures are necessary to be adopted to complete the arming of the militia of the United States” to the House of Representatives.[4] The report, partly reprinted below, is difficult to reconcile with Churchill’s findings concerning gun ownership among militia members in the revolutionary and early national periods, on which his argument that the Second Amendment protected a private as well as a militia-focused right to arms largely depends. Congressman Varnum drew on data familiar to the committee and reprinted in the Militia Census of 1806, a document officially communicated to Congress by the president nine days later on April 11, 1806.[5] The results of that census, and the question of how they square with Churchill’s account, are taken up next. In the report itself, Varnum’s committee informed Congress
[t]hat, by the laws of the United States, each citizen enrolled in the militia is put under obligations to provide himself with a good musket or rifle, and all the other military equipments prescribed by law. From the best estimate which the committee have been able to form, there is upwards of 250,000 fire arms and rifles in the hands of the militia, which have, a few instances excepted, been provided by, and are the property of, the individuals who hold them. It is highly probable, that many more of the militia would have provided themselves with fire arms in the same way, if they had been for sale in those parts of the United States where the deficiencies have happened; but the wars in Europe have had a tendency to prevent the importation of fire arms from thence into the United States, which, together with the limited establishments for the manufacture of that implement in the United States, has rendered it impossible for individuals to procure them.[6]

The committee went on to say that the number of stands of public arms in the arsenals of the various states had not been ascertained, that there were about 120,000 fire arms fit for use and 12,000 in need of repair in the magazines of the United States, and that the committee was of the opinion that further public monies ($62,100 to be exact) should be set aside for manufacture of fire arms in the armories of the United States “to provide for the exigency of war.”[7]

The Militia Census listed the total numbers of men enrolled for each state rank by rank and the total numbers of rifles and muskets each state reported. Assuming that privates and noncommissioned officers but not officers were expected to carry long guns as required by the Militia Act of 1792, the percentages of militia members each state reported as armed with rifles or muskets are as follows:[8]
New Hampshire Infantry: 19,100 privates and 1,108 sergeants, 12,500 muskets; or 61.9 percent armed.
Massachusetts Infantry: 53,316 privates and 1,108 sergeants, 46,218 muskets and 397 rifles; or 85.7 percent armed.
Rhode Island Infantry: 4,414 privates and 302 sergeants, 3,052 muskets; or 64.7 percent armed.
Connecticut Infantry: 13,952 privates, 1,144 corporals, 1,293 sergeants, 15,085 muskets; or 92.0 percent armed.
Vermont Infantry: 13,708 privates, 1,011 sergeants, 8,824 muskets; or 59.9 percent armed.
New York Infantry: 63,744 privates, 3,885 sergeants, 39,919 muskets and 1,928 rifles; or 61.9 percent armed.
New Jersey Infantry: 21,742 privates, 1,142 sergeants, 12,423 muskets and 86 rifles; or 54.7 percent armed.
Pennsylvania Infantry and Riflemen: 80,061 privates, 2,881 sergeants, 3,352 riflemen, 20,000 muskets, 3,352 rifles; or 27.1 percent armed.
Delaware, not reporting.
Maryland, not reporting.
Virginia Infantry: 61,962 privates, 3,388 sergeants, 10,490 muskets, 2,734 rifles; or 21.3 percent armed.
North Carolina Infantry: 37,871 privates, 1,774 sergeants, 16,571 muskets, 2,343 rifles; or 47.7 percent armed.
South Carolina Infantry, Riflemen, and Light Infantry: 29,082 privates and rank and file, 245 pioneers, 165 corporals, 1,245 sergeants, 5,916 muskets, 5,731 rifles; or 37.9 percent armed.
Georgia Infantry and Riflemen: 16,650 infantry and rank and file, 835 sergeants, 1,782 muskets, 1,955 rifles; or 21.4 percent armed.
Kentucky Infantry: 29,386 privates, 1,679 sergeants, 3,966 muskets, 15,567 rifles; or 62.9 percent armed.
Tennessee Infantry: 14,285 privates, 308 corporals, 308 sergeants, 4,647 muskets; or 31.2 percent armed.
Ohio Infantry: 8,031 privates, 456 sergeants, 277 muskets, 3,238 rifles; or 41.4 percent armed.

Varnum’s report points to the committee’s concerns over a national militia less than fully armed and then proposes to rectify this problem by Congressional spending on arms production in federal arsenals for distribution (via the market? state purchase and resale? loan? outright grant?) to unarmed militia members. As the numbers above make clear, the census he and the committee consulted in reaching this decision indicated that the New England militia was substantially armed and that the middle state militia (except in Pennsylvania, home to large populations of Quakers and other conscientious objectors) was above half armed. But serious problems arose in the South and West (excepting Kentucky), and these problems were nowhere as acute as in Virginia. Churchill, however, argues that that disarmament there was more apparent than real. The census, he claims, counted only state owned arms (and not privately owned arms) in Virginia and several other states. Yet the committee reported that it did not know how many arms were held in the arsenals of the states, and this is very hard to incorporate into Churchill’s interpretation of the census, unless his point is that Virginians who had been issued state owned arms kept them at home and brought them to muster where these guns (unlike the guns still in the arsenals) were counted.

Churchill’s main thrust on this issue is that most Virginia militia members actually owned their own guns (why would they have so many fewer than their northern compatriots?), but refused to bring them to muster in large measure because of a state history of confiscation. Here again Churchill’s thesis stumbles over its own inconsistencies. The claim that Virginians were still influenced by expectations of confiscation in 1806 is not wholly in harmony with Churchill’s larger argument that the power to seize atrophied in the 1780s even as a sense of immunity against confiscation took hold in the popular mind.

Varnum’s report and the census finding of low armament in Virginia is troubling for the Churchill thesis in at least one other sense as well. If Churchill is right that Virginians had guns but did not bring them to muster, it becomes necessary to explain why a Jeffersonian controlled Congress closely tied to the Virginian president was unaware of this issue. This holds particularly for Major General Varnum himself, given his life-long service with Massachusetts citizen soldiery, his national responsibilities for militia oversight, and his personal relations with the president—he became Jefferson’s candidate for Speaker of the House in the next Congress and won appointment when former Speaker John Randolph’s faction broke with the administration. If the cause of the Virginia militia’s seeming unreadiness was as Churchill supposes, it stands to reason that Jefferson’s Virginia connections, including the state’s three most recent governors—James Monroe (1799–1802), John Page (1802–1805), and William Cabell (1805–1808), all Jefferson loyalists—would have informed the president, and that Jefferson would have passed to word to Varnum, one of his leading New England lieutenants in the House and chair of the committee responsible for supervising arming of the militia. Churchill’s assumption is equally hard to square with then Governor Monroe’s behavior six years earlier in 1800, when he was called on to consider the Virginia militia’s potential effectiveness as a potential counterweight to a Hamiltonian army unwilling to yield the presidency in the event of a Republican victory in the national elections. Monroe made it a point to order arms from overseas, not to order Virginians to bring their arms out of hiding.[9]

In a cordial email to this reviewer, Churchill has stressed that the assumption of widespread arms ownership that underlies his thesis is the product of his detailed research into a variety of sources, including probate records, and the local militia rolls, which he found formed the basis of state figures included in federal militia censuses. In truth, my disagreements with Churchill have less to do with the prevalence of guns in early national culture (my sense is that the Census of 1806 is about right, his studied retort is that it substantially undercounts) than with the purpose and meaning that Americans attached to their ownership of guns, and the question of how that fed into their thinking (such as there was on this point) about the Second Amendment. And in this respect, Churchill’s argument appears based on an oddly ambiguous set of assumptions about statutory compliance. His reasoning relies on two premises. First, Americans complied willingly and broadly with colonial and state level militia-linked requirements to acquire guns. Second, they later followed the federal Militia Act’s command that white men of arms-bearing years obtain a musket or a rifle. But he builds on these assumptions to argue that once Americans came into compliance, and became accustomed to a culture of arms bearing, the statutory purposes behind their acquisition of guns were subordinated. Ownership of guns took on an individualistic valence says Churchill, with hostility to gun confiscation reflecting less and less solicitude for the communitarian militia, and more and more a property-focused sense of private immunity. This understanding in turn became imbued with qualities perhaps more readily associated with modern Takings Clause jurisprudence (and its late eighteenth-century precursors), and the sort of “Lockean” rhetoric Locke may not have recognized, than with the anti-army trappings of old Commonwealth Whiggery.

I believe that Churchill reads too much libertarianism and too little republicanism into the problem, and that along the way he smoothes over some important ambiguities that his evidence, fairly read, will not resolve. Once more, the Varnum report is instructive. Varnum suggests that most militia eligible Americans wanted to comply with the Militia Act’s requirement of arming themselves, but that many were unable to do so because guns were scarce. Whether Varnum was too charitable respecting the causes of wide-spread non-compliance (lack of guns as opposed to lack of will), the fact remains that, unless Churchill’s largely conclusory surmise that many southerners and westerners were hiding their guns is true, nearly half the militia eligible population was non-compliant. Non-compliance was not an uncommon theme in recent American history. Churchill himself claims that non-compliance with the Act of 1792 (failure to appear armed on muster day) actually explains the alleged undercounting of guns in Virginia. Far more famously, the Sugar Act, Stamp Act, Townsend Duties, and Tea Act come to mind as late colonial statutes generating less than optimal compliance, as do the Whiskey Tax and Window Tax from the Federalist period. To be sure, these were imperial or at least national as opposed to provincial or state laws such as those Churchill cites to support his claim for a wide distribution of arms. But other provincial or state laws, including prohibitions against unlicensed preaching and absenting oneself from the established church in a manner not contemplated in the Toleration Act, were notoriously under-enforced or unenforceable as well.[10]

If Americans were as widely out of compliance with late colonial militia-linked mandatory arming laws as their successors were with the U.S. Act of 1792, then serious problems arise at the beginning of Churchill’s chronological chain linking the presumption of wide spread gun possession (required by statute) to familiarity to possessory impulses to claims of right to assertions of immunity to constitutionalization. And it is in the earlier period, where Churchill insists the discourse that ripened into rights talk began, that he relies most strongly on unadorned assumptions of statutory compliance, for the evidence from probate inventories and censuses becomes thicker only as the colonial period ends. If late colonial Americans were as non-compliant in regard to gun ownership as they were respecting tax payment and religious establishment, perhaps they were less obsessed with clinging to guns they did not have for individualistic property-focused purposes than they were animated with pro-militia and anti-army rhetoric for civic and republican ends.

Churchill’s essay is problematic not simply for evidentiary reasons. His argument builds principally on the theoretical distinction between military authority to seize and police power to regulate guns, but this theoretical distinction may require substantial rethinking. It is premised at least in part on the assumption that measures relying on military authority were extraordinary and rare, while exercises of the police power were quotidian and norm defining. Even if this were true, however, it would in no way undermine the theory that fears of the standing armies and executive usurpation were central to the Second Amendment, for it was in extraordinary times of crises real or imagined or pre-textual that efforts to disarm the militia and set up a corrupt regime buttressed by the army were most to be expected. Churchill’s underlying assumption, however, is in fact not true for the generation that experienced the Revolution and the constitutional crisis. As Alan Taylor among others reminds us, imperial wars between Britain and France were more common than not in the late colonial period, and those wars increasingly focused on the North American theater and increasingly mobilized the North American population. A native born American aged fifty when the new national government convened in 1789 had known more years of war than peace.[11] The great imperial and national political debates of that person’s lifetime had focused on war, taxes to fund war, and the dangers of a government capable of enacting and enforcing the tax regime required to finance war or hold together a country sufficiently powerful to avoid war.

If war, or fear of war, or the need to pay for or avoid war was the norm for the founding generation, perhaps this does not so much undermine Churchill’s principal claim as suggest that national attitudes were bound to change. Fears of undue assertions of military authority subsided in the decades that followed the revolutionary period, with a clearly civilian-controlled Jeffersonian system of governance firmly in place in the substantially demilitarized nation that became the ante-bellum republic. But this does not get Churchill wholly off the hook. His premises remain problematic for the earlier period in which he roots his analysis, and for colonial times, his terminological distinction between military powers and police powers in some respects is itself anachronistic.

The sharp distinction between military and police powers makes much more sense under the system of federalism and separation of powers adopted in the U.S. Constitution of 1788 than it does for the colonial system of governance. The national Constitution conveyed certain specified powers to the United States Congress (such as the power to raise and support armies), rested the Commander-in-Chief power in the federal presidency, and reserved the bulk of unspecified powers to the states, the latter including the general authority inherent in their quasi sovereign status and partly confirmed in the Tenth Amendment to make general policy respecting health, safety, and morals. Whether the distinction between military and police powers will bear as much weight as Churchill would load on it in colonial times is a more doubtful proposition. The constitutional settlement reached during the Glorious Revolution in England left intact the royal prerogatives that Prince William insisted on keeping to make the Crown worthwhile, including those related to war and peace and command of the military. In the eighteenth century, however, these devolved in practice to the cabinet and the prime minister. From 1689, military funding was required to flow from Parliament, and the Bill of Rights spoke of allowing protestant subjects such arms as were allowed by law.[12]

On an ad hoc and imperfect basis, each colony’s government of provincial assembly, governor, and council mirrored the British system, and during war time (which was, as mentioned above, as normal as not) military relations between colonial and royal government frequently became complicated and confused by the presence of regulars responsible directly to the Crown under whose commanders colonial militia sometimes served. How meaningful it is in this context to attempt to label weapons seizures for purposes of militia arming (often carried out under imperial pressure to get more local troops in the field) exercises of either police or military powers is difficult to say.[13] The authority behind confiscation appears to have been sometimes imperial, sometimes local, sometimes prerogative, sometimes statutory, and sometimes a matter of ad hoc necessity. It is likewise open to question whether the coming of independence marked any conceptual sharpening of the distinctions between police and military powers respecting the issue of guns, so much as it did a general heightening of the popular preference for militia over regulars, and realization by the revolutionary leadership that regulars were as necessary during war as they were dangerous to peace and to republican principles.

Churchill’s essay is engaging and thought provoking throughout. His central insight that power to regulate and power to confiscate are not one and the same is of crucial importance, both to understanding the meaning of the American right to arms at its origins, and to understanding the fevered politics that envelop that right in our own times. I have doubts, however, that Churchill’s distinction between authority under police and military powers offers a complete and accurate account of changing attitudes towards the militia and the right to arms in the founding and early national periods. In his contribution to this volume, Saul Cornell has pointed to grave problems concerning Churchill’s use of evidence from the constitutional period and from the nineteenth century.[14] My own concerns focus on Churchill’s extrapolations from assumptions perhaps too hastily drawn about the meaning of arms ownership to Americans in the late eighteenth and early nineteenth centuries. While Churchill has raised interesting questions and offered intriguing insights, he has also made generalizations that his evidence is not strong enough to support. In the end, I remain convinced that David Konig, Saul Cornell, and my late friend and mentor Richard Uviller and I were correct to stress the civic, militia-focused meaning of the right to arms that dominated discussion at the time of the Second Amendment’s framing, and (as Cornell ably shows in this forum) continued to predominate long into the nineteenth century. There were, to be sure, countervailing voices. To ignore them would be false to the historical record and would wrongly deprive many enthusiastic supporters of a broad right to own guns of a sense of provenance to which they attach much meaning. Churchill is right to take those voices seriously. But in the article discussed here, Churchill exaggerates their importance and forges for them a kinship with the mainstream that a careful reading of the record cannot always confirm.

William G. Merkel is a professor at Washburn University School of Law.


1. Robert H. Churchill, “Gun Ownership in Early America: A Survey of Manuscript Militia Returns,” The William and Mary Quarterly, 3d ser., 60 (2003): 615.

2. Ibid., 642.

3. Robert H. Churchill, “Gun Regulation, the Police Power, and the Right to Keep Arms in Early America: The Legal Context of the Second Amendment,” Law and History Review 25 (2007): 139–75.

4. American State Papers, 5, Military Affairs, 1:198–99. For information on Varnum, see the entry in American National Biography 22:278–79 by Edward W. Hanson of the Massachusetts Historical Society.

5. American State Papers, 5, Military Affairs, 1:199.

6. Ibid., 199.

7. Ibid., 198–99.

8. These are my rough calculations; I have left out sergeant majors and quartermaster sergeants because of their insignificant numbers and my uncertainty as to whether they were expected to muster with long guns. I have also omitted the reports for the District of Columbia, Mississippi Territory, and Indiana Territory, whose militia were few in number. The Orleans and Louisiana territories did not report. For the raw numbers see American State Papers, 5, Military Affairs, 1:202–3. For Churchill’s important reservations about reading too much into these numbers see Churchill, “Gun Ownership in Early America.” Churchill makes the case that the census should not be taken at face value; instead, he maintains, it is important to look at the documents generated at the brigade level on which the census was based. By failing to do so, Churchill cautions that historians will be mislead because the census undercounts by measuring guns brought to muster, not guns held at home. But if this were a serious problem, one would expect Major General Varnum, with thirty years militia experience, to have been aware of it. He reported “upwards of 250,000 firearms and rifles in the hands of the militia”; the census lists by my count some 204,200 muskets and 52,900 rifles, suggesting strongly that it provided the basis of Varnum’s figures. If Varnum knew of systemic undercounting, he failed to tell Congress, and mislead his colleagues in the process. See American State Papers, 5, Military Affairs, 1:199.

9. See William G. Merkel, “To See Oneself as a Target of a Justified Revolution: Thomas Jefferson and Gabriel’s Rebellion,” American Nineteenth Century History 4.1 (2003): 1–31.

10. See Rhys Isaac, The Transformation of Virginia: Community, Religion, and Authority, 1740–1790 (Chapel Hill: University of North Carolina Press, 1982).

11. Alan Taylor, American Colonies (New York: Viking, 2001), 420–43.

12. See detailed discussion in Lois G. Schwoerer, “To Hold and Bear Arms: The English Perspective,” Chicago-Kent Law Review 76 (2000): 27, and in H. Richard Uviller and William G. Merkel, The Militia and the Right to Arms, Or How the Second Amendment Fell Silent (Durham: North Carolina Press, 2002), 47–56.

13. See, e.g., Michael A. McDonnell, “The Politics of Mobilization in Revolutionary Virginia: Military Culture and Political and Social Relations, 1774–1783” (Oxford University D.Phil. thesis, 1995).

14. Saul Cornell, “Early American Gun Regulation and the Second Amendment: A Closer Look at the Evidence,” Law and History Review 25 (2007): 197–204.