478 F. 3d 370, affirmed.

HTML version
PDF version
HTML version
PDF version
HTML version
PDF version
HTML version
PDF version

Breyer, J., dissenting



on writ of certiorari to the united states court ofappeals for the district of columbia circuit

[June 26, 2008]

Justice Breyer, with whom Justice Stevens, Justice Souter, and Justice Ginsburg join, dissenting.

We must decide whether a District of Columbia law that prohibits the possession of handguns in the home violates the Second Amendment . The majority, relying upon its view that the Second Amendment seeks to protect a right of personal self-defense, holds that this law violates that Amendment. In my view, it does not.


The majority’s conclusion is wrong for two independent reasons. The first reason is that set forth by Justice Stevens—namely, that the Second Amendment protects militia-related, not self-defense-related, interests. These two interests are sometimes intertwined. To assure 18th-century citizens that they could keep arms for militia purposes would necessarily have allowed them to keep arms that they could have used for self-defense as well. But self-defense alone, detached from any militia-related objective, is not the Amendment’s concern.

The second independent reason is that the protection the Amendment provides is not absolute. The Amendment permits government to regulate the interests that it serves. Thus, irrespective of what those interests are—whether they do or do not include an independent interest in self-defense—the majority’s view cannot be correct unless it can show that the District’s regulation is unreasonable or inappropriate in Second Amendment terms. This the majority cannot do.

In respect to the first independent reason, I agree with Justice Stevens, and I join his opinion. In this opinion I shall focus upon the second reason. I shall show that the District’s law is consistent with the Second Amendment even if that Amendment is interpreted as protecting a wholly separate interest in individual self-defense. That is so because the District’s regulation, which focuses upon the presence of handguns in high-crime urban areas, represents a permissible legislative response to a serious, indeed life-threatening, problem.

Thus I here assume that one objective (but, as the majority concedes, ante, at 26, not the primary objective) of those who wrote the Second Amendment was to help assure citizens that they would have arms available for purposes of self-defense. Even so, a legislature could reasonably conclude that the law will advance goals of great public importance, namely, saving lives, preventing injury, and reducing crime. The law is tailored to the urban crime problem in that it is local in scope and thus affects only a geographic area both limited in size and entirely urban; the law concerns handguns, which are specially linked to urban gun deaths and injuries, and which are the overwhelmingly favorite weapon of armed criminals; and at the same time, the law imposes a burden upon gun owners that seems proportionately no greater than restrictions in existence at the time the Second Amendment was adopted. In these circumstances, the District’s law falls within the zone that the Second Amendment leaves open to regulation by legislatures.


The Second Amendment says that: “A well regulated Militia, being necessary to the security of a free State, the right of the people to keep and bear Arms, shall not be infringed.” In interpreting and applying this Amendment, I take as a starting point the following four propositions, based on our precedent and today’s opinions, to which I believe the entire Court subscribes:

(1) The Amendment protects an “individual” right—i.e., one that is separately possessed, and may be separately enforced, by each person on whom it is conferred. See, e.g., ante, at 22 (opinion of the Court); ante, at 1 (Stevens, J., dissenting).

(2) As evidenced by its preamble, the Amendment was adopted “[w]ith obvious purpose to assure the continuation and render possible the effectiveness of [militia] forces.” United States v. Miller, 307 U. S. 174, 178 (1939) ; see ante, at 26 (opinion of the Court); ante, at 1 (Stevens, J., dissenting).

(3) The Amendment “must be interpreted and applied with that end in view.” Miller, supra, at 178.

(4) The right protected by the Second Amendment is not absolute, but instead is subject to government regulation. See Robertson v. Baldwin, 165 U. S. 275, 281–282 (1897) ; ante, at 22, 54 (opinion of the Court).

My approach to this case, while involving the first three points, primarily concerns the fourth. I shall, as I said, assume with the majority that the Amendment, in addition to furthering a militia-related purpose, also furthers an interest in possessing guns for purposes of self-defense, at least to some degree. And I shall then ask whether the Amendment nevertheless permits the District handgun restriction at issue here.

Although I adopt for present purposes the majority’s position that the Second Amendment embodies a general concern about self-defense, I shall not assume that the Amendment contains a specific untouchable right to keep guns in the house to shoot burglars. The majority, which presents evidence in favor of the former proposition, does not, because it cannot, convincingly show that the Second Amendment seeks to maintain the latter in pristine, unregulated form.

To the contrary, colonial history itself offers important examples of the kinds of gun regulation that citizens would then have thought compatible with the “right to keep and bear arms,” whether embodied in Federal or State Constitutions, or the background common law. And those examples include substantial regulation of firearms in urban areas, including regulations that imposed obstacles to the use of firearms for the protection of the home.

Boston, Philadelphia, and New York City, the three largest cities in America during that period, all restricted the firing of guns within city limits to at least some degree. See Churchill, Gun Regulation, the Police Power, and the Right to Keep Arms in Early America, 25Law & Hist. Rev. 139, 162 (2007); Dept. of Commerce, Bureau of Census, C. Gibson, Population of the 100 Largest Cities and Other Urban Places in the United States: 1790 to 1990 (1998) (Table 2), online at http://www.census.gov/population/documentation/twps0027/tab02.txt (all Internet materials as visited June 19, 2008, and available in Clerk of Court’s case file). Boston in 1746 had a law prohibiting the “discharge” of “any Gun or Pistol charged with Shot or Ball in the Town” on penalty of 40 shillings, a law that was later revived in 1778. See Act of May 28, 1746, ch. 10; An Act for Reviving and Continuing Sundry Laws that are Expired, and Near Expiring, 1778 Massachusetts Session Laws, ch. 5, pp. 193, 194. Philadelphia prohibited, on penalty of 5 shillings (or two days in jail if the fine were not paid), firing a gun or setting off fireworks in Philadelphia without a “governor’s special license.” See Act of Aug. 26, 1721, §4, in 3 Mitchell, Statutes at Large of Pennsylvania 253–254. And New York City banned, on penalty of a 20-shilling fine, the firing of guns (even in houses) for the three days surrounding New Year’s Day. 5 Colonial Laws of New York, ch. 1501, pp. 244–246 (1894); see also An Act to Suppress the Disorderly Practice of Firing Guns, & c., on the Times Therein Mentioned, 8 Statutes at Large of Pennsylvania 1770–1776, pp. 410–412 (1902) (similar law for all “inhabited parts” of Pennsylvania). See also An Act for preventing Mischief being done in the Town of Newport, or in any other Town in this Government, 1731, Rhode Island Session Laws (prohibiting, on penalty of 5 shillings for a first offense and more for subsequent offenses, the firing of “any Gun or Pistol … in the Streets of any of the Towns of this Government, or in any Tavern of the same, after dark, on any Night whatsoever”).

Furthermore, several towns and cities (including Philadelphia, New York, and Boston) regulated, for fire-safety reasons, the storage of gunpowder, a necessary component of an operational firearm. See Cornell & DeDino, A Well Regulated Right, 73 Fordham L. Rev. 487, 510–512 (2004). Boston’s law in particular impacted the use of firearms in the home very much as the District’s law does today. Boston’s gunpowder law imposed a 10 fine upon “any Person” who “shall take into any Dwelling-House, Stable, Barn, Out-house, Ware-house, Store, Shop, or other Building, within the Town of Boston, any … Fire-Arm, loaded with, or having Gun-Powder.” An Act in Addition to the several Acts already made for the prudent Storage of Gun-Powder within the Town of Boston, ch. XIII, 1783 Mass. Acts 218–219; see also 1 S. Johnson, A Dictionary of the English Language 751 (4th ed. 1773) (defining “firearms” as “[a]rms which owe their efficacy to fire; guns”). Even assuming, as the majority does, see ante, at 59–60, that this law included an implicit self-defense exception, it would nevertheless have prevented a homeowner from keeping in his home a gun that he could immediately pick up and use against an intruder. Rather, the homeowner would have had to get the gunpowder and load it into the gun, an operation that would have taken a fair amount of time to perform. See Hicks, United States Military Shoulder Arms, 1795–1935, 1 Am. Military Hist. Foundation 23, 30 (1937) (experienced soldier could, with specially prepared cartridges as opposed to plain gunpowder and ball, load and fire musket 3-to-4 times per minute); id., at 26–30 (describing the loading process); see also Grancsay, The Craft of the Early American Gunsmith, 6 Metropolitan Museum of Art Bulletin 54, 60 (1947) (noting that rifles were slower to load and fire than muskets).

Moreover, the law would, as a practical matter, have prohibited the carrying of loaded firearms anywhere in the city, unless the carrier had no plans to enter any building or was willing to unload or discard his weapons before going inside. And Massachusetts residents must have believed this kind of law compatible with the provision in the Massachusetts Constitution that granted “the people … a right to keep and to bear arms for the common defence”—a provision that the majority says was interpreted as “secur[ing] an individual right to bear arms for defensive purposes.” Art. XVII (1780), in 3 The Federal and State Constitutions, Colonial Charters, and Other Organic Laws 1888, 1892 (F. Thorpe ed. 1909) (hereinafter Thorpe); ante, at 28–29 (opinion of the Court).

The New York City law, which required that gunpowder in the home be stored in certain sorts of containers, and laws in certain Pennsylvania towns, which required that gunpowder be stored on the highest story of the home, could well have presented similar obstacles to in-home use of firearms. See Act of April 13, 1784, ch. 28, 1784 N. Y. Laws p. 627; An Act for Erecting the Town of Carlisle, in the County of Cumberland, into a Borough, ch. XIV, §XLII, 1782 Pa. Laws p. 49; An Act for Erecting the Town of Reading, in the County of Berks, into a Borough, ch. LXXVI, §XLII, 1783 Pa. Laws p. 211. Although it is unclear whether these laws, like the Boston law, would have prohibited the storage of gunpowder inside a firearm, they would at the very least have made it difficult to reload the gun to fire a second shot unless the homeowner happened to be in the portion of the house where the extra gunpowder was required to be kept. See 7 United States Encyclopedia of History 1297 (P. Oehser ed. 1967) (“Until 1835 all small arms [were] single-shot weapons, requiring reloading by hand after every shot”). And Pennsylvania, like Massachusetts, had at the time one of the self-defense-guaranteeing state constitutional provisions on which the majority relies. See ante, at 28 (citing Pa. Declaration of Rights, Art. XIII (1776), in 5 Thorpe 3083).

The majority criticizes my citation of these colonial laws. See ante, at 59–62. But, as much as it tries, it cannot ignore their existence. I suppose it is possible that, as the majority suggests, see ante, at 59–61, they all in practice contained self-defense exceptions. But none of them expressly provided one, and the majority’s assumption that such exceptions existed relies largely on the preambles to these acts—an interpretive methodology that it elsewhere roundly derides. Compare ibid. (interpreting 18th-century statutes in light of their preambles), with ante, at 4–5, and n. 3 (contending that the operative language of an 18th-century enactment may extend beyond its preamble). And in any event, as I have shown, the gunpowder-storage laws would have burdened armed self-defense, even if they did not completely prohibit it.

This historical evidence demonstrates that a self-defense assumption is the beginning, rather than the end, of any constitutional inquiry. That the District law impacts self-defense merely raises questions about the law’s constitutionality. But to answer the questions that are raised (that is, to see whether the statute is unconstitutional) requires us to focus on practicalities, the statute’s rationale, the problems that called it into being, its relation to those objectives—in a word, the details. There are no purely logical or conceptual answers to such questions. All of which to say that to raise a self-defense question is not to answer it.


I therefore begin by asking a process-based question: How is a court to determine whether a particular firearm regulation (here, the District’s restriction on handguns) is consistent with the Second Amendment ? What kind of constitutional standard should the court use? How high a protective hurdle does the Amendment erect?

The question matters. The majority is wrong when it says that the District’s law is unconstitutional “[u]nder any of the standards of scrutiny that we have applied to enumerated constitutional rights.” Ante, at 56. How could that be? It certainly would not be unconstitutional under, for example, a “rational basis” standard, which requires a court to uphold regulation so long as it bears a “rational relationship” to a “legitimate governmental purpose.” Heller v. Doe, 509 U. S. 312, 320 (1993) . The law at issue here, which in part seeks to prevent gun-related accidents, at least bears a “rational relationship” to that “legitimate” life-saving objective. And nothing in the three 19th-century state cases to which the majority turns for support mandates the conclusion that the present District law must fall. See Andrews v. State, 50 Tenn. 165, 177, 186–187, 192 (1871) (striking down, as violating a state constitutionalprovision adopted in 1870, a statewide ban on acarrying a broad class of weapons, insofar as it applied to revolvers); Nunn v. State, 1 Ga. 243, 246, 250–251 (1846) (striking down similarly broad ban on openly carrying weapons, based on erroneous view that the Federal Second Amendment applied to the States); State v. Reid, 1 Ala. 612, 614–615, 622 (1840) (upholding a concealed-weapon ban against a state constitutional challenge). These cases were decided well (80, 55, and 49 years, respectively) after the framing; they neither claim nor provide any special insight into the intent of the Framers; they involve laws much less narrowly tailored that the one before us; and state cases in any event are not determinative of federal constitutional questions, see, e.g., Garcia v. San Antonio Metropolitan Transit Authority, 469 U. S. 528, 549 (1985) (citing Martin v. Hunter’s Lessee, 1 Wheat. 304 (1816)).

Respondent proposes that the Court adopt a “strict scrutiny” test, which would require reviewing with care each gun law to determine whether it is “narrowly tailored to achieve a compelling governmental interest.” Abrams v. Johnson, 521 U. S. 74, 82 (1997) ; see Brief for Respondent 54–62. But the majority implicitly, and appropriately, rejects that suggestion by broadly approving a set of laws—prohibitions on concealed weapons, forfeiture by criminals of the Second Amendment right, prohibitions on firearms in certain locales, and governmental regulation of commercial firearm sales—whose constitutionality under a strict scrutiny standard would be far from clear. See ante, at 54.

Indeed, adoption of a true strict-scrutiny standard for evaluating gun regulations would be impossible. That is because almost every gun-control regulation will seek to advance (as the one here does) a “primary concern of every government—a concern for the safety and indeed the lives of its citizens.” United States v. Salerno, 481 U. S. 739, 755 (1987) . The Court has deemed that interest, as well as “the Government’s general interest in preventing crime,” to be “compelling,” see id., at 750, 754, and the Court has in a wide variety of constitutional contexts found such public-safety concerns sufficiently forceful to justify restrictions on individual liberties, see e.g., Brandenburg v. Ohio, 395 U. S. 444, 447 (1969) (per curiam) ( First Amendment free speech rights); Sherbert v. Verner, 374 U. S. 398, 403 (1963) ( First Amendment religious rights); Brigham City v. Stuart, 547 U. S. 398, 403–404 (2006) ( Fourth Amendment protection of the home); New York v. Quarles, 467 U. S. 649, 655 (1984) ( Fifth Amendment rights under Miranda v. Arizona, 384 U. S. 436 (1966) ); Salerno, supra, at755 ( Eighth Amendment bail rights). Thus, any attempt in theory to apply strict scrutiny to gun regulations will in practice turn into an interest-balancing inquiry, with the interests protected by the Second Amendment on one side and the governmental public-safety concerns on the other, the only question being whether the regulation at issue impermissibly burdens the former in the course of advancing the latter.

I would simply adopt such an interest-balancing inquiry explicitly. The fact that important interests lie on both sides of the constitutional equation suggests that review of gun-control regulation is not a context in which a court should effectively presume either constitutionality (as in rational-basis review) or unconstitutionality (as in strict scrutiny). Rather, “where a law significantly implicates competing constitutionally protected interests in complex ways,” the Court generally asks whether the statute burdens a protected interest in a way or to an extent that is out of proportion to the statute’s salutary effects upon other important governmental interests. See Nixon v. Shrink Missouri Government PAC, 528 U. S. 377, 402 (2000) (Breyer, J., concurring). Any answer would take account both of the statute’s effects upon the competing interests and the existence of any clearly superior less restrictive alternative. See ibid. Contrary to the majority’s unsupported suggestion that this sort of “proportionality” approach is unprecedented, see ante, at 62, the Court has applied it in various constitutional contexts, including election-law cases, speech cases, and due process cases. See 528 U. S., at 403 (citing examples where the Court has taken such an approach); see also, e.g., Thompson v. Western States Medical Center, 535 U. S. 357, 388 (2002) (Breyer, J., dissenting) (commercial speech); Burdick v. Takushi, 504 U. S. 428, 433 (1992) (election regulation); Mathews v. Eldridge, 424 U. S. 319, 339–349 (1976) (procedural due process); Pickering v. Board of Ed. of Township High School Dist. 205, Will Cty., 391 U. S. 563, 568 (1968) (government employee speech).

In applying this kind of standard the Court normally defers to a legislature’s empirical judgment in matters where a legislature is likely to have greater expertise and greater institutional factfinding capacity. See Turner Broadcasting System, Inc. v. FCC, 520 U. S. 180, 195–196 (1997) ; see also Nixon, supra, at 403(Breyer, J., concurring). Nonetheless, a court, not a legislature, must make the ultimate constitutional conclusion, exercising its “independent judicial judgment” in light of the whole record to determine whether a law exceeds constitutional boundaries. Randall v. Sorrell, 548 U. S. 230, 249 (2006) (opinion of Breyer, J.) (citing Bose Corp. v. Consumers Union of United States, Inc., 466 U. S. 485, 499 (1984) ).

The above-described approach seems preferable to a more rigid approach here for a further reason. Experience as much as logic has led the Court to decide that in one area of constitutional law or another the interests are likely to prove stronger on one side of a typical constitutional case than on the other. See, e.g., United States v. Virginia, 518 U. S. 515, 531–534 (1996) (applying heightened scrutiny to gender-based classifications, based upon experience with prior cases); Williamson v. Lee Optical of Okla., Inc., 348 U. S. 483, 488 (1955) (applying rational-basis scrutiny to economic legislation, based upon experience with prior cases). Here, we have little prior experience. Courts that do have experience in these matters have uniformly taken an approach that treats empirically-based legislative judgment with a degree of deference. See Winkler, Scrutinizing the Second Amendment , 105 Mich. L. Rev. 683, 687, 716–718 (2007) (describing hundreds of gun-law decisions issued in the last half-century by Supreme Courts in 42 States, which courts with “surprisingly little variation,” have adopted a standard more deferential than strict scrutiny). While these state cases obviously are not controlling, they are instructive. Cf., e.g., Bartkus v. Illinois, 359 U. S. 121, 134 (1959) (looking to the “experience of state courts” as informative of a constitutional question). And they thus provide some comfort regarding the practical wisdom of following the approach that I believe our constitutional precedent would in any event suggest.


The present suit involves challenges to three separate District firearm restrictions. The first requires a license from the District’s Chief of Police in order to carry a “pistol,” i.e., a handgun, anywhere in the District. See D. C. Code §22–4504(a) (2001); see also §§22–4501(a), 22–4506. Because the District assures us that respondent could obtain such a license so long as he meets the statutory eligibility criteria, and because respondent concedes that those criteria are facially constitutional, I, like the majority, see no need to address the constitutionality of the licensing requirement. See ante, at 58–59.

The second District restriction requires that the lawful owner of a firearm keep his weapon “unloaded and disassembled or bound by a trigger lock or similar device” unless it is kept at his place of business or being used for lawful recreational purposes. See §7–2507.02. The only dispute regarding this provision appears to be whether the Constitution requires an exception that would allow someone to render a firearm operational when necessary for self-defense (i.e., that the firearm may be operated under circumstances where the common law would normally permit a self-defense justification in defense against a criminal charge). See Parker v. District of Columbia, 478 F. 3d 370, 401 (2007) (case below); ante, at 57–58 (opinion of the Court); Brief for Respondent 52–54. The District concedes that such an exception exists. See Brief for Petitioners 56–57. This Court has final authority (albeit not often used) to definitively interpret District law, which is, after all, simply a species of federal law. See, e.g., Whalen v. United States, 445 U. S. 684, 687–688 (1980) ; see also Griffin v. United States, 336 U. S. 704, 716–718 (1949) . And because I see nothing in the District law that would preclude the existence of a background common-law self-defense exception, I would avoid the constitutional question by interpreting the statute to include it. See Ashwander v. TVA, 297 U. S. 288, 348 (1936) (Brandeis, J., concurring).

I am puzzled by the majority’s unwillingness to adopt a similar approach. It readily reads unspoken self-defense exceptions into every colonial law, but it refuses to accept the District’s concession that this law has one. Compare ante, at 59–61, with ante, at 57–58. The one District case it cites to support that refusal, McIntosh v. Washington, 395 A. 2d 744, 755–756 (1978), merely concludes that the District Legislature had a rational basis for applying the trigger-lock law in homes but not in places of business. Nowhere does that case say that the statute precludes a self-defense exception of the sort that I have just described. And even if it did, we are not bound by a lower court’s interpretation of federal law.

The third District restriction prohibits (in most cases) the registration of a handgun within the District. See §7–2502.02(a)(4). Because registration is a prerequisite to firearm possession, see §7–2502.01(a), the effect of this provision is generally to prevent people in the District from possessing handguns. In determining whether this regulation violates the Second Amendment , I shall ask how the statute seeks to further the governmental interests that it serves, how the statute burdens the interests that the Second Amendment seeks to protect, and whether there are practical less burdensome ways of furthering those interests. The ultimate question is whether the statute imposes burdens that, when viewed in light of the statute’s legitimate objectives, are disproportionate. See Nixon, 528 U. S., at 402 (Breyer, J., concurring).


No one doubts the constitutional importance of the statute’s basic objective, saving lives. See, e.g., Salerno, 481 U. S., at 755. But there is considerable debate about whether the District’s statute helps to achieve that objective. I begin by reviewing the statute’s tendency to secure that objective from the perspective of (1) the legislature (namely, the Council of the District of Columbia) that enacted the statute in 1976, and (2) a court that seeks to evaluate the Council’s decision today.


First, consider the facts as the legislature saw them when it adopted the District statute. As stated by the local council committee that recommended its adoption, the major substantive goal of the District’s handgun restriction is “to reduce the potentiality for gun-related crimes and gun-related deaths from occurring within the District of Columbia.” Hearing and Disposition before the House Committee on the District of Columbia, 94th Cong., 2d Sess., on H. Con. Res. 694, Ser. No. 94–24, p. 25 (1976) (herinafter DC Rep.) (reproducing, inter alia, the Council committee report). The committee concluded, on the basis of “extensive public hearings” and “lengthy research,” that “[t]he easy availability of firearms in the United States has been a major factor contributing to the drastic increase in gun-related violence and crime over the past 40 years.” Id., at 24, 25. It reported to the Council “startling statistics,” id., at 26, regarding gun-related crime, accidents, and deaths, focusing particularly on the relation between handguns and crime and the proliferation of handguns within the District. See id., at 25–26.

The committee informed the Council that guns were “responsible for 69 deaths in this country each day,” for a total of “[a]pproximately 25,000 gun-deaths … each year,” along with an additional 200,000 gun-related injuries. Id., at 25. Three thousand of these deaths, the report stated, were accidental. Ibid. A quarter of the victims in those accidental deaths were children under the age of 14. Ibid. And according to the committee,“[f]or every intruder stopped by a homeowner with a firearm, there are 4 gun-related accidents within the home.” Ibid.

In respect to local crime, the committee observed that there were 285 murders in the District during 1974—a record number. Id., at 26. The committee also stated that, “[c]ontrary to popular opinion on the subject, firearms are more frequently involved in deaths and violence among relatives and friends than in premeditated criminal activities.” Ibid. Citing an article from the American Journal of Psychiatry, the committee reported that “[m]ost murders are committed by previously law-abiding citizens, in situations where spontaneous violence is generated by anger, passion or intoxication, and where the killer and victim are acquainted.” Ibid. “Twenty-five percent of these murders,” the committee informed the Council, “occur within families.” Ibid.

The committee report furthermore presented statistics strongly correlating handguns with crime. Of the 285 murders in the District in 1974, 155 were committed with handguns. Ibid. This did not appear to be an aberration, as the report revealed that “handguns [had been] used in roughly 54% of all murders” (and 87% of murders of law enforcement officers) nationwide over the preceding several years. Ibid. Nor were handguns only linked to murders, as statistics showed that they were used in roughly 60% of robberies and 26% of assaults. Ibid. “A crime committed with a pistol,” the committee reported, “is 7 times more likely to be lethal than a crime committed with any other weapon.” Id., at 25. The committee furthermore presented statistics regarding the availability of handguns in the United States, ibid., and noted that they had “become easy for juveniles to obtain,” even despite then-current District laws prohibiting juveniles from possessing them, id., at 26.

In the committee’s view, the current District firearms laws were unable “to reduce the potentiality for gun-related violence,” or to “cope with the problems of gun control in the District” more generally. Ibid. In the absence of adequate federal gun legislation, the committee concluded, it “becomes necessary for local governments to act to protect their citizens, and certainly the District of Columbia as the only totally urban statelike jurisdiction should be strong in its approach.” Id., at 27. It recommended that the Council adopt a restriction on handgun registration to reflect “a legislative decision that, at this point in time and due to the gun-control tragedies and horrors enumerated previously” in the committee report, “pistols … are no longer justified in this jurisdiction.” Id., at 31; see also ibid. (handgun restriction “denotes a policy decision that handguns … have no legitimate use in the purely urban environment of the District”).

The District’s special focus on handguns thus reflects the fact that the committee report found them to have a particularly strong link to undesirable activities in the District’s exclusively urban environment. See id., at 25–26. The District did not seek to prohibit possession of other sorts of weapons deemed more suitable for an “urban area.” See id., at 25. Indeed, an original draft of the bill, and the original committee recommendations, had sought to prohibit registration of shotguns as well as handguns, but the Council as a whole decided to narrow the prohibition. Compare id., at 30 (describing early version of the bill), with D. C. Code §7–2502.02).


Next, consider the facts as a court must consider them looking at the matter as of today. See, e.g., Turner, 520 U. S., at 195 (discussing role of court as factfinder in a constitutional case). Petitioners, and their amici, have presented us with more recent statistics that tell much the same story that the committee report told 30 years ago. At the least, they present nothing that would permit us to second-guess the Council in respect to the numbers of gun crimes, injuries, and deaths, or the role of handguns.

From 1993 to 1997, there were 180,533 firearm-related deaths in the United States, an average of over 36,000 per year. Dept. of Justice, Bureau of Justice Statistics, M. Zawitz & K. Strom, Firearm Injury and Death from Crime, 1993–97, p. 2 (Oct. 2000), online at http://www.ojp.usdoj.gov/bjs/pub/pdf/fidc9397.pdf (hereinafter Firearm Injury and Death from Crime). Fifty-one percent were suicides, 44% were homicides, 1% were legal interventions, 3% were unintentional accidents, and 1% were of undetermined causes. See ibid. Over that same period there were an additional 411,800 nonfatal firearm-related injuries treated in U. S. hospitals, an average of over 82,000 per year. Ibid. Of these, 62% resulted from assaults, 17% were unintentional, 6% were suicide attempts, 1% were legal interventions, and 13% were of unknown causes. Ibid.

The statistics are particularly striking in respect to children and adolescents. In over one in every eight firearm-related deaths in 1997, the victim was someone under the age of 20. American Academy of Pediatrics, Firearm-Related Injuries Affecting the Pediatric Population, 105 Pediatrics 888 (2000) (hereinafter Firearm-Related Injuries). Firearm-related deaths account for 22.5% of all injury deaths between the ages of 1 and 19. Ibid. More male teenagers die from firearms than from all natural causes combined. Dresang, Gun Deaths in Rural and Urban Settings, 14 J. Am. Bd. Family Practice 107 (2001). Persons under 25 accounted for 47% of hospital-treated firearm injuries between June 1, 1992 and May 31, 1993. Firearm-Related Injuries891.

Handguns are involved in a majority of firearm deaths and injuries in the United States. Id., at 888. From 1993 to 1997, 81% of firearm-homicide victims were killed by handgun. Firearm Injury and Death from Crime 4; see also Dept. of Justice, Bureau of Justice Statistics, C. Perkins, Weapon Use and Violent Crime, p. 8 (Sept. 2003), (Table 10), http://www.ojp.usdoj.gov/bjs/pub/pdf/wuvc01.pdf (hereinafter Weapon Use and Violent Crime) (statistics indicating roughly the same rate for 1993–2001). In the same period, for the 41% of firearm injuries for which the weapon type is known, 82% of them were from handguns. Firearm Injury and Death From Crime 4. And among children under the age of 20, handguns account for approximately 70% of all unintentional firearm-related injuries and deaths. Firearm-Related Injuries 890. In particular, 70% of all firearm-related teenage suicides in 1996 involved a handgun. Id., at 889; see also Zwerling, Lynch, Burmeister, & Goertz, The Choice of Weapons in Firearm Suicides in Iowa, 83 Am. J. Public Health 1630, 1631 (1993) (Table 1) (handguns used in 36.6% of all firearm suicides in Iowa from 1980–1984 and 43.8% from 1990–1991).

Handguns also appear to be a very popular weapon among criminals. In a 1997 survey of inmates who were armed during the crime for which they were incarcerated, 83.2% of state inmates and 86.7% of federal inmatessaid that they were armed with a handgun. See Dept. ofJustice, Bureau of Justice Statistics, C. Harlow, Firearm Use by Offenders, p. 3 (Nov. 2001), online at http://www.ojp.usdoj.gov/bjs/pub/pdf/fuo.pdf; see also Weapon Use and Violent Crime 2 (Table 2) (statistics indicating that handguns were used in over 84% of nonlethal violent crimes involving firearms from 1993 to 2001). And handguns are not only popular tools for crime, but popular objects of it as well: the FBI received on average over 274,000 reports of stolen guns for each year between 1985 and 1994, and almost 60% of stolen guns are handguns. Dept. of Justice, Bureau of Justice Statistics, M. Zawitz, Guns Used in Crime, p. 3 (July 1995), online at http://www.ojp.usdoj.gov/bjs/pub/pdf/guic.pdf. Department of Justice studies have concluded that stolen handguns in particular are an important source of weapons for both adult and juvenile offenders. Ibid.

Statistics further suggest that urban areas, such as the District, have different experiences with gun-related death, injury, and crime, than do less densely populated rural areas. A disproportionate amount of violent and property crimes occur in urban areas, and urban criminals are more likely than other offenders to use a firearm during the commission of a violent crime. See Dept. of Justice, Bureau of Justice Statistics, D. Duhart, Urban, Suburban, and Rural Victimization, 1993–98, pp. 1, 9 (Oct. 2000), online at http://www.ojp.usdoj.gov/bjs/pub/pdf/usrv98.pdf. Homicide appears to be a much greater issue in urban areas; from 1985 to 1993, for example, “half of all homicides occurred in 63 cities with 16% of the nation’s population.” Wintemute, The Future of Firearm Violence Prevention, 282JAMA 475 (1999). One study concluded that although the overall rate of gun death between 1989 and 1999 was roughly the same in urban than rural areas, the urban homicide rate was three times as high; even after adjusting for other variables, it was still twice as high. Branas, Nance, Elliott, Richmond, & Schwab, Urban-Rural Shifts in Intentional Firearm Death, 94 Am. J. Public Health 1750, 1752 (2004); see also ibid. (noting that rural areas appear to have a higher rate of firearm suicide). And a study of firearm injuries to children and adolescents in Pennsylvania between 1987 and 2000 showed an injury rate in urban counties 10 times higher than in nonurban counties. Nance & Branas, The Rural-Urban Continuum, 156 Archives of Pediatrics & Adolescent Medicine 781, 782 (2002).

Finally, the linkage of handguns to firearms deaths and injuries appears to be much stronger in urban than in rural areas. “[S]tudies to date generally support the hypothesis that the greater number of rural gun deaths are from rifles or shotguns, whereas the greater number of urban gun deaths are from handguns.” Dresang, supra, at 108. And the Pennsylvania study reached a similar conclusion with respect to firearm injuries—they are much more likely to be caused by handguns in urban areas than in rural areas. See Nance & Branas, supra, at 784.


Respondent and his many amici for the most part do not disagree about the figures set forth in the preceding subsection, but they do disagree strongly with the District’s predictive judgment that a ban on handguns will help solve the crime and accident problems that those figures disclose. In particular, they disagree with the District Council’s assessment that “freezing the pistol … population within the District,” DC Rep., at 26, will reduce crime, accidents, and deaths related to guns. And they provide facts and figures designed to show that it has not done so in the past, and hence will not do so in the future.

First, they point out that, since the ban took effect, violent crime in the District has increased, not decreased. See Brief for Criminologists et al. as Amici Curiae 4–8, 3a (hereinafter Criminologists’ Brief); Brief for Congress of Racial Equality as Amicus Curiae 35–36; Brief for National Rifle Assn. et al. as Amici Curiae 28–30 (hereinafter NRA Brief). Indeed, a comparison with 49 other major cities reveals that the District’s homicide rate is actually substantially higher relative to these other cities than it was before the handgun restriction went into effect. See Brief for Academics as Amici Curiae 7–10 (hereinafter Academics’ Brief); see also Criminologists’ Brief 6–9, 3a–4a, 7a. Respondent’s amici report similar results in comparing the District’s homicide rates during that period to that of the neighboring States of Maryland and Virginia (neither of which restricts handguns to the same degree), and to the homicide rate of the Nation as a whole. SeeAcademics’ Brief 11–17; Criminologists’ Brief 6a, 8a.

Second, respondent’s amici point to a statistical analysis that regresses murder rates against the presence or absence of strict gun laws in 20 European nations. See Criminologists’ Brief 23 (citing Kates & Mauser, Would Banning Firearms Reduce Murder and Suicide? 30 Harv. J. L. & Pub. Pol’y 649, 651–694 (2007)). That analysis concludes that strict gun laws are correlated with more murders, not fewer. See Criminologists’ Brief 23;see also id., at 25–28.They also cite domestic studies, based on data from various cities, States, and the Nation as a whole, suggesting that a reduction in the number of guns does not lead to a reduction in the amount of violent crime. See id., at 17–20. They further argue that handgun bans do not reduce suicide rates, see id., at 28–31, 9a, or rates of accidents, even those involving children, see Brief for International Law Enforcement Educators and Trainers Assn. et al. as Amici Curiae App. 7–15 (hereinafter ILEETA Brief).

Third, they point to evidence indicating that firearm ownership does have a beneficial self-defense effect. Based on a 1993 survey, the authors of one study estimated that there were 2.2-to-2.5 million defensive uses of guns (mostly brandishing, about a quarter involving the actual firing of a gun) annually. See Kleck & Gertz, Armed Resistance to Crime, 86J. Crim. L. & C. 150, 164 (1995); see also ILEETA Brief App. 1–6 (summarizing studies regarding defensive uses of guns). Another study estimated that for a period of 12 months ending in 1994, there were 503,481 incidents in which a burglar found himself confronted by an armed homeowner, and that in 497,646 (98.8%) of them, the intruder was successfully scared away. See Ikida, Dahlberg, Sacks, Mercy, & Powell, Estimating Intruder-Related Firearms Retrievals in U. S. Households, 12 Violence & Victims 363 (1997). A third study suggests that gun-armed victims are substantially less likely than non-gun-armed victims to be injured in resisting robbery or assault. Barnett & Kates, Under Fire, 45 Emory L. J. 1139, 1243–1244, n. 478 (1996). And additional evidence suggests that criminals are likely to be deterred from burglary and other crimes if they know the victim is likely to have a gun. See Kleck, Crime Control Through the Private Use of Armed Force, 35 Social Problems 1, 15 (1988) (reporting a substantial drop in the burglary rate in an Atlanta suburb that required heads of households to own guns); see also ILEETA Brief 17–18 (describing decrease in sexual assaults in Orlando when women were trained in the use of guns).

Fourth, respondent’s amici argue that laws criminalizing gun possession are self-defeating, as evidence suggests that they will have the effect only of restricting law-abiding citizens, but not criminals, from acquiring guns. See, e.g., Brief for President Pro Tempore of Senate of Pennsylvania as Amicus Curiae 35, 36, and n. 15. That effect, they argue, will be especially pronounced in the District, whose proximity to Virginia and Maryland will provide criminals with a steady supply of guns. See Brief for Heartland Institute as Amicus Curiae 20.

In the view of respondent’s amici, this evidence shows that other remedies—such as less restriction on gun ownership, or liberal authorization of law-abiding citizens to carry concealed weapons—better fit the problem. See, e.g., Criminologists’ Brief 35–37 (advocating easily obtainable gun licenses); Brief for Southeastern Legal Foundation, Inc. et al. as Amici Curiae 15 (hereinafter SLF Brief) (advocating “widespread gun ownership” as a deterrent to crime); see also J. Lott, More Guns, Less Crime (2d ed. 2000). They further suggest that at a minimum the District fails to show that its remedy, the gun ban, bears a reasonable relation to the crime and accident problems that the District seeks to solve. See, e.g., Brief for Respondent 59–61.

These empirically based arguments may have proved strong enough to convince many legislatures, as a matter of legislative policy, not to adopt total handgun bans. But the question here is whether they are strong enough to destroy judicial confidence in the reasonableness of a legislature that rejects them. And that they are not. For one thing, they can lead us more deeply into the uncertainties that surround any effort to reduce crime, but they cannot prove either that handgun possession diminishes crime or that handgun bans are ineffective. The statistics do show a soaring District crime rate. And the District’s crime rate went up after the District adopted its handgun ban. But, as students of elementary logic know, after it does not mean because of it. What would the District’s crime rate have looked like without the ban? Higher? Lower? The same? Experts differ; and we, as judges, cannot say.

What about the fact that foreign nations with strict gun laws have higher crime rates? Which is the cause and which the effect? The proposition that strict gun laws cause crime is harder to accept than the proposition that strict gun laws in part grow out of the fact that a nation already has a higher crime rate. And we are then left with the same question as before: What would have happened to crime without the gun laws—a question that respondent and his amici do not convincingly answer.

Further, suppose that respondent’s amici are right when they say that householders’ possession of loaded handguns help to frighten away intruders. On that assumption, one must still ask whether that benefit is worth the potential death-related cost. And that is a question without a directly provable answer.

Finally, consider the claim of respondent’s amici that handgun bans cannot work; there are simply too many illegal guns already in existence for a ban on legal guns to make a difference. In a word, they claim that, given the urban sea of pre-existing legal guns, criminals can readily find arms regardless. Nonetheless, a legislature might respond, we want to make an effort to try to dry up that urban sea, drop by drop. And none of the studies can show that effort is not worthwhile.

In a word, the studies to which respondent’s amici point raise policy-related questions. They succeed in proving that the District’s predictive judgments are controversial. But they do not by themselves show that those judgments are incorrect; nor do they demonstrate a consensus, academic or otherwise, supporting that conclusion.

Thus, it is not surprising that the District and its amici support the District’s handgun restriction with studies of their own. One in particular suggests that, statistically speaking, the District’s law has indeed had positive life-saving effects. See Loftin, McDowall, Weirsema, & Cottey, Effects of Restrictive Licensing of Handguns on Homicide and Suicide in the District of Columbia, 325 New England J. Med. 1615 (1991) (hereinafter Loftin study). Others suggest that firearm restrictions as a general matter reduce homicides, suicides, and accidents in the home. See, e.g., Duggan, More Guns, More Crime, 109J. Pol. Econ. 1086 (2001); Kellerman, Somes, Rivara, Lee, & Banton, Injuries and Deaths Due to Firearms in the Home, 45 J. Trauma, Infection & Critical Care 263 (1998); Miller, Azrael, & Hemenway, Household Firearm Ownership and Suicide Rates in the United States, 13 Epidemiology 517 (2002). Still others suggest that the defensive uses of handguns are not as great in number as respondent’s amici claim. See, e.g., Brief for American Public Health Assn. et al. as Amici Curiae 17–19 (hereinafter APHA Brief) (citing studies).

Respondent and his amici reply to these responses; and in doing so, they seek to discredit as methodologically flawed the studies and evidence relied upon by the District. See, e.g., Criminologists’ Brief 9–17, 20–24; Brief for Assn. Am. Physicians and Surgeons, Inc. as Amicus Curiae 12–18; SLF Brief17–22; Britt, Kleck, & Bordua, A Reassessment of the D.C. Gun Law, 30Law & Soc. Rev. 361 (1996) (criticizing the Loftin study). And, of course, the District’s amici produce counter-rejoinders, referring to articles that defend their studies. See, e.g., APHA Brief 23, n. 5 (citing McDowall, Loftin, & Wiersema et al., Using Quasi-Experiments to Evaluate Firearm Laws, 30Law & Soc. Rev. 381 (1996)).

The upshot is a set of studies and counterstudies that, at most, could leave a judge uncertain about the proper policy conclusion. But from respondent’s perspective any such uncertainty is not good enough. That is because legislators, not judges, have primary responsibility for drawing policy conclusions from empirical fact. And, given that constitutional allocation of decisionmaking responsibility, the empirical evidence presented here is sufficient to allow a judge to reach a firm legal conclusion.

In particular this Court, in First Amendment cases applying intermediate scrutiny, has said that our “sole obligation” in reviewing a legislature’s “predictive judgments” is “to assure that, in formulating its judgments,” the legislature “has drawn reasonable inferences based on substantial evidence.” Turner, 520 U. S., at 195 (internal quotation marks omitted). And judges, looking at the evidence before us, should agree that the District legislature’s predictive judgments satisfy that legal standard. That is to say, the District’s judgment, while open to question, is nevertheless supported by “substantial evidence.”

There is no cause here to depart from the standard set forth in Turner, for the District’s decision represents the kind of empirically based judgment that legislatures, not courts, are best suited to make. See Nixon, 528 U. S., at 402 (Breyer, J., concurring). In fact, deference to legislative judgment seems particularly appropriate here, where the judgment has been made by a local legislature, with particular knowledge of local problems and insight into appropriate local solutions. See Los Angeles v. Alameda Books, Inc., 535 U. S. 425, 440 (2002) (plurality opinion) (“[W]e must acknowledge that the Los Angeles City Council is in a better position than the Judiciary to gather an evaluate data on local problems”); cf. DC Rep., at 67 (statement of Rep. Gude) (describing District’s law as “a decision made on the local level after extensive debate and deliberations”).Different localities may seek to solve similar problems in different ways, and a “city must be allowed a reasonable opportunity to experiment with solutions to admittedly serious problems.” Renton v. Playtime Theatres, Inc., 475 U. S. 41, 52 (1986) (internal quotation marks omitted). “The Framers recognized that the most effective democracy occurs at local levels of government, where people with firsthand knowledge of local problems have more ready access to public officials responsible for dealing with them.” Garcia v. San Antonio Metropolitan Transit Authority, 469 U. S. 528, 575, n. 18 (1985) (Powell, J., dissenting) (citing The Federalist No. 17, p. 107 (J. Cooke ed. 1961) (A. Hamilton)). We owe that democratic process some substantial weight in the constitutional calculus.

For these reasons, I conclude that the District’s statute properly seeks to further the sort of life-preserving and public-safety interests that the Court has called “compelling.” Salerno, 481 U. S., at 750, 754.


I next assess the extent to which the District’s law burdens the interests that the Second Amendment seeks to protect. Respondent and his amici, as well as the majority, suggest that those interests include: (1) the preservation of a “well regulated Militia”; (2) safeguarding the use of firearms for sporting purposes, e.g., hunting and marksmanship; and (3) assuring the use of firearms for self-defense. For argument’s sake, I shall consider all three of those interests here.


The District’s statute burdens the Amendment’s first and primary objective hardly at all. As previously noted, there is general agreement among the Members of the Court that the principal (if not the only) purpose of the Second Amendment is found in the Amendment’s text: the preservation of a “well regulated Militia.” See supra, at 3. What scant Court precedent there is on the Second Amendment teaches that the Amendment was adopted “[w]ith obvious purpose to assure the continuation and render possible the effectiveness of [militia] forces” and “must be interpreted and applied with that end in view.” Miller, 307 U. S., at 178. Where that end is implicated only minimally (or not at all), there is substantially less reason for constitutional concern. Compare ibid. (“In the absence of any evidence tending to show that possession or use of a ‘shotgun having a barrel of less than eighteen inches in length’ at this time has some reasonable relationship to the preservation or efficiency of a wellregulated militia, we cannot say that the Second Amendment guarantees the right to keep and bear such aninstrument”).

To begin with, the present case has nothing to do with actual military service. The question presented presumes that respondent is “not affiliated with any state-regulated militia.” 552 U. S. __ (2007) (emphasis added). I am aware of no indication that the District either now or in the recent past has called up its citizenry to serve in a militia, that it has any inkling of doing so anytime in the foreseeable future, or that this law must be construed to prevent the use of handguns during legitimate militia activities. Moreover, even if the District were to call up its militia, respondent would not be among the citizens whose service would be requested. The District does not consider him, at 66 years of age, to be a member of its militia. See D. C. Code §49–401 (2001) (militia includes only male residents ages 18 to 45); App. to Pet. for Cert. 120a (indicating respondent’s date of birth).

Nonetheless, as some amici claim, the statute might interfere with training in the use of weapons, training useful for military purposes. The 19th-century constitutional scholar, Thomas Cooley, wrote that the Second Amendment protects “learning to handle and use [arms] in a way that makes those who keep them ready for their efficient use” during militia service. General Principles of Constitutional Law 271 (1880); ante, at 45 (opinion of the Court); see also ante, at 45–46 (citing other scholars agreeing with Cooley on that point). And former military officers tell us that “private ownership of firearms makes for a more effective fighting force” because “[m]ilitary recruits with previous firearms experience and training are generally better marksmen, and accordingly, better soldiers.” Brief for Retired Military Officers as Amici Curiae 1–2 (hereinafter Military Officers’ Brief). An amicus brief filed by retired Army generals adds that a “well-regulated militia—whether ad hoc or as part of our organized military—depends on recruits who have familiarity and training with firearms—rifles, pistols, and shotguns.” Brief for Major General John D. Altenburg, Jr., et al. as Amici Curiae 4 (hereinafter Generals’ Brief). Both briefs point out the importance of handgun training. Military Officers’ Brief 26–28; Generals’ Brief 4. Handguns are used in military service, see id., at 26, and “civilians who are familiar with handgun marksmanship and safety are much more likely to be able to safely and accurately fire a rifle or other firearm with minimal training upon entering military service,” id., at 28.

Regardless, to consider the military-training objective a modern counterpart to a similar militia-related colonial objective and to treat that objective as falling within the Amendment’s primary purposes makes no difference here. That is because the District’s law does not seriously affect military training interests. The law permits residents to engage in activities that will increase their familiarity with firearms. They may register (and thus possess in their homes) weapons other than handguns, such as rifles and shotguns. See D. C. Code §§7–2502.01, 7–2502.02(a) (only weapons that cannot be registered are sawed-off shotguns, machine guns, short-barreled rifles, and pistols not registered before 1976); compare Generals’ Brief 4 (listing “rifles, pistols, and shotguns” as useful military weapons; emphasis added). And they may operate those weapons within the District “for lawful recreational purposes.” §7–2507.02; see also §7–2502.01(b)(3) (nonresidents “participating in any lawful recreational firearm-related activity in the District, or on his way to or from such activity in another jurisdiction” may carry even weapons not registered in the District). These permissible recreations plainly include actually using and firing the weapons, as evidenced by a specific D. C. Code provision contemplating the existence of local firing ranges. See§7–2507.03.

And while the District law prevents citizens from training with handguns within the District, the District consists of only 61.4 square miles of urban area. See Dept. of Commerce, Bureau of Census, United States: 2000 (pt. 1), p. 11 (2002) (Table 8). The adjacent States do permit the use of handguns for target practice, and those States are only a brief subway ride away. See Md. Crim. Law Code Ann. §4–203(b)(4) (Lexis Supp. 2007) (general handgun restriction does not apply to “the wearing, carrying, or transporting by a person of a handgun used in connection with,” inter alia, “a target shoot, formal or informal target practice, sport shooting event, hunting, [or] a Department of Natural Resources-sponsored firearms and hunter safety class”); Va. Code Ann. §18.2–287.4 (Lexis Supp. 2007) (general restriction on carrying certain loaded pistols in certain public areas does not apply “to any person actually engaged in lawful hunting or lawful recreational shooting activities at an established shooting range or shooting contest”); Washington Metropolitan Area Transit Authority, Metrorail System Map, http://www.wmata.com/metrorail/systemmmap.cfm.

Of course, a subway rider must buy a ticket, and the ride takes time. It also costs money to store a pistol, say, at a target range, outside the District. But given the costs already associated with gun ownership and firearms training, I cannot say that a subway ticket and a short subway ride (and storage costs) create more than a minimal burden. Compare Crawford v. Marion County Election Bd., 553 U. S. ___, ___ (2008) (slip op., at 3) (Breyer, J., dissenting) (acknowledging travel burdens on indigent persons in the context of voting where public transportation options were limited). Indeed, respondent and two of his coplaintiffs below may well use handguns outside the District on a regular basis, as their declarations indicate that they keep such weapons stored there. See App. to Pet. for Cert. 77a (respondent); see also id., at 78a, 84a (coplaintiffs). I conclude that the District’s law burdens the Second Amendment ’s primary objective little, or not at all.


The majority briefly suggests that the “right to keep and bear Arms” might encompass an interest in hunting. See, e.g., ante, at 26. But in enacting the present provisions, the District sought “to take nothing away from sportsmen.” DC Rep., at 33. And any inability of District residents to hunt near where they live has much to do with the jurisdiction’s exclusively urban character and little to do with the District’s firearm laws. For reasons similar to those I discussed in the preceding subsection—that the District’s law does not prohibit possession of rifles or shotguns, and the presence of opportunities for sporting activities in nearby States—I reach a similar conclusion, namely, that the District’s law burdens any sports-related or hunting-related objectives that the Amendment may protect little, or not at all.


The District’s law does prevent a resident from keeping a loaded handgun in his home. And it consequently makes it more difficult for the householder to use the handgun for self-defense in the home against intruders, such as burglars. As the Court of Appeals noted, statistics suggest that handguns are the most popular weapon for self defense. See 478 F. 3d, at 400 (citing Kleck & Gertz, 86 J. Crim. L. & C., at 182–183). And there are some legitimate reasons why that would be the case: Amici suggest (with some empirical support) that handguns are easier to hold and control (particularly for persons with physical infirmities), easier to carry, easier to maneuver in enclosed spaces, and that a person using one will still have a hand free to dial 911. See ILEETA Brief 37–39; NRA Brief 32–33; see also ante, at 57. But see Brief for Petitioners 54–55 (citing sources preferring shotguns and rifles to handguns for purposes of self-defense). To that extent the law burdens to some degree an interest in self-defense that for present purposes I have assumed the Amendment seeks to further.


In weighing needs and burdens, we must take account of the possibility that there are reasonable, but less restrictive alternatives. Are there other potential measures that might similarly promote the same goals while imposing lesser restrictions? See Nixon, 528 U. S., at 402 (Breyer, J., concurring) (“existence of a clearly superior, less restrictive alternative” can be a factor in determining whether a law is constitutionally proportionate). Here I see none.

The reason there is no clearly superior, less restrictive alternative to the District’s handgun ban is that the ban’s very objective is to reduce significantly the number of handguns in the District, say, for example, by allowing a law enforcement officer immediately to assume that any handgun he sees is an illegal handgun. And there is no plausible way to achieve that objective other than to ban the guns.

It does not help respondent’s case to describe the District’s objective more generally as an “effort to diminish the dangers associated with guns.” That is because the very attributes that make handguns particularly useful for self-defense are also what make them particularly dangerous. That they are easy to hold and control means that they are easier for children to use. See Brief for American Academy of Pediatrics et al. as Amici Curiae 19 (“[C]hildren as young as three are able to pull the trigger of most handguns”). That they are maneuverable and permit a free hand likely contributes to the fact that they are by far the firearm of choice for crimes such as rape and robbery. See Weapon Use and Violent Crime 2 (Table 2). That they are small and light makes them easy to steal, see supra, at 19, and concealable, cf. ante, at 54 (opinion of the Court) (suggesting that concealed-weapon bans are constitutional).

This symmetry suggests that any measure less restrictive in respect to the use of handguns for self-defense will, to that same extent, prove less effective in preventing the use of handguns for illicit purposes. If a resident has a handgun in the home that he can use for self-defense, then he has a handgun in the home that he can use to commit suicide or engage in acts of domestic violence. See supra, at 18 (handguns prevalent in suicides); Brief for National Network to End Domestic Violence et al. as Amici Curiae 27 (handguns prevalent in domestic violence). If it is indeed the case, as the District believes, that the number of guns contributes to the number of gun-related crimes, accidents, and deaths, then, although there may be less restrictive, less effective substitutes for an outright ban, there is no less restrictive equivalent of an outright ban.

Licensing restrictions would not similarly reduce the handgun population, and the District may reasonably fear that even if guns are initially restricted to law-abiding citizens, they might be stolen and thereby placed in the hands of criminals. See supra, at 19. Permitting certain types of handguns, but not others, would affect the commercial market for handguns, but not their availability. And requiring safety devices such as trigger locks, or imposing safe-storage requirements would interfere with any self-defense interest while simultaneously leaving operable weapons in the hands of owners (or others capable of acquiring the weapon and disabling the safety device) who might use them for domestic violence or other crimes.

The absence of equally effective alternatives to a complete prohibition finds support in the empirical fact that other States and urban centers prohibit particular types of weapons. Chicago has a law very similar to the District’s, and many of its suburbs also ban handgun possession under most circumstances. See Chicago, Ill., Municipal Code §§8–20–030(k), 8–20–40, 8–20–50(c) (2008); Evanston, Ill., City Code §9–8–2 (2007); Morton Grove, Ill., Village Code §6–2–3(C) (2008); Oak Park, Ill., Village Code §27–2–1 (2007); Winnetka, Ill., Village Ordinance §9.12.020(B) (2008); Wilmette, Ill., Ordinance §12–24(b) (2008). Toledo bans certain types of handguns. Toledo, Ohio, Municipal Code, ch. 549.25 (2007). And San Francisco in 2005 enacted by popular referendum a ban on most handgun possession by city residents; it has been precluded from enforcing that prohibition, however, by state-court decisions deeming it pre-empted by state law. See Fiscal v. City and County of San Francisco, 158 Cal. App. 4th 895, 900–901, 70 Cal. Rptr. 3d 324, 326–328 (2008). (Indeed, the fact that as many as 41 States may pre-empt local gun regulation suggests that the absence of more regulation like the District’s may perhaps have more to do with state law than with a lack of locally perceived need for them. See Legal Community Against Violence, Regulating Guns in America 14 (2006), http://www.lcav.org/Library/reports_analyses/National_Audit_Total_8.16.06.pdf.

In addition, at least six States and Puerto Rico impose general bans on certain types of weapons, in particular assault weapons or semiautomatic weapons. See Cal. Penal Code §12280(b) (West Supp. 2008); Conn. Gen. Stat. §§53–202c (2007); Haw. Rev. Stat. §134–8 (1993); Md. Crim. Law Code Ann. §4–303(a) (Lexis 2002); Mass. Gen. Laws, ch. 140, §131M (West 2006); N. Y. Penal Law Ann. §265.02(7) (West Supp. 2008); 25 Laws P. R. Ann. §456m (Supp. 2006); see also 18 U. S. C. §922(o) (federal machinegun ban). And at least 14 municipalities do the same. See Albany, N. Y., Municipal Code §193–16(A) (2005); Aurora, Ill., Ordinance §29–49(a) (2007); Buffalo, N. Y., City Code §180–1(F) (2000); Chicago, Ill., Municipal Code §8–24–025(a), 8–20–030(h); Cincinnati, Ohio, Admin. Code §708–37(a) (Supp. 2008); Cleveland, Ohio, Ordinance §628.03(a) (2008); Columbus, Ohio, City Code §2323.31 (2007); Denver, Colo., Municipal Code §38–130(e) (2008); Morton Grove, Ill., Village Code §6–2–3(B); N. Y. C. Admin. Code §10–303.1 (2007); Oak Park, Ill., Village Code §27–2-1; Rochester, N. Y., Code §47–5(f) (2008); South Bend, Ind., Ordinance §§13–97(b), 13–98 (2008); Toledo, Ohio, Municipal Code §549.23(a). These bans, too, suggest that there may be no substitute to an outright prohibition in cases where a governmental body has deemed a particular type of weapon especially dangerous.


The upshot is that the District’s objectives are compelling; its predictive judgments as to its law’s tendency to achieve those objectives are adequately supported; the law does impose a burden upon any self-defense interest that the Amendment seeks to secure; and there is no clear less restrictive alternative. I turn now to the final portion of the “permissible regulation” question: Does the District’s law disproportionately burden Amendment-protected interests? Several considerations, taken together, convince me that it does not.

First, the District law is tailored to the life-threatening problems it attempts to address. The law concerns one class of weapons, handguns, leaving residents free to possess shotguns and rifles, along with ammunition. The area that falls within its scope is totally urban. Cf. Lorillard Tobacco Co. v. Reilly, 533 U. S. 525, 563 (2001) (varied effect of statewide speech restriction in “rural, urban, or suburban” locales “demonstrates a lack of narrow tailoring”). That urban area suffers from a serious handgun-fatality problem. The District’s law directly aims at that compelling problem. And there is no less restrictive way to achieve the problem-related benefits that it seeks.

Second, the self-defense interest in maintaining loaded handguns in the home to shoot intruders is not the primary interest, but at most a subsidiary interest, that the Second Amendment seeks to serve. The Second Amendment ’s language, while speaking of a “Militia,” says nothing of “self-defense.” As Justice Stevens points out, the Second Amendment ’s drafting history shows that the language reflects the Framers’ primary, if not exclusive, objective. See ante, at 17–28 (dissenting opinion). And the majority itself says that “the threat that the new Federal Government would destroy the citizens’ militia by taking away their arms was the reason that right . . . was codified in a written Constitution.” Ante, at 26 (emphasis added). The way in which the Amendment’s operative clause seeks to promote that interest—by protecting a right “to keep and bear Arms”—may in fact help further an interest in self-defense. But a factual connection falls far short of a primary objective. The Amendment itself tells us that militia preservation was first and foremost in the Framers’ minds. See Miller, 307 U. S., at 178 (“With obvious purpose to assure the continuation and render possible the effectiveness of [militia] forces the declaration and guarantee of the Second Amendment were made,” and the amendment “must be interpreted and applied with that end in view”).

Further, any self-defense interest at the time of the Framing could not have focused exclusively upon urban-crime related dangers. Two hundred years ago, most Americans, many living on the frontier, would likely have thought of self-defense primarily in terms of outbreaks of fighting with Indian tribes, rebellions such as Shays’ Rebellion, marauders, and crime-related dangers to travelers on the roads, on footpaths, or along waterways. See Dept. of Commerce, Bureau of Census, Population: 1790 to 1990 (1998) (Table 4), online at http://www.census.gov/population/censusdata/table-4.pdf (of the 3,929,214 Americans in 1790, only 201,655—about 5%—lived in urban areas). Insofar as the Framers focused at all on the tiny fraction of the population living in large cities, they would have been aware that these city dwellers were subject to firearm restrictions that their rural counterparts were not. See supra, at 4–7. They are unlikely then to have thought of a right to keep loaded handguns in homes to confront intruders in urban settings as central. And the subsequent development of modern urban police departments, by diminishing the need to keep loaded guns nearby in case of intruders, would have moved any such right even further away from the heart of the amendment’s more basic protective ends. See, e.g., Sklansky, The Private Police, 46 UCLA L. Rev. 1165, 1206–1207 (1999) (professional urban police departments did not develop until roughly the mid-19th century).

Nor, for that matter, am I aware of any evidence that handguns in particular were central to the Framers’ conception of the Second Amendment . The lists of militia-related weapons in the late 18th-century state statutes appear primarily to refer to other sorts of weapons, muskets in particular. See Miller, 307 U. S., at 180–182 (reproducing colonial militia laws). Respondent points out in his brief that the Federal Government and two States at the time of the founding had enacted statutes that listed handguns as “acceptable” militia weapons. Brief for Respondent 47. But these statutes apparently found them “acceptable” only for certain special militiamen (generally, certain soldiers on horseback), while requiring muskets or rifles for the general infantry. See Act of May 8, 1792, ch. XXXIII, 1 Stat. 271; Laws of the State of North Carolina 592 (1791); First Laws of the State of Connecticut 150 (1784); see also 25 Journals of the Continental Congress, pp. 1774–1789 741–742 (1922).

Third,irrespective of what the Framers could have thought, we know what they did think. Samuel Adams, who lived in Boston, advocated a constitutional amendment that would have precluded the Constitution from ever being “construed” to “prevent the people of the United States, who are peaceable citizens, from keeping their own arms.” 6 Documentary History of the Ratification of the Constitution 1453 (J. Kaminski & G. Saladino eds. 2000). Samuel Adams doubtless knew that the Massachusetts Constitution contained somewhat similar protection. And he doubtless knew that Massachusetts law prohibited Bostonians from keeping loaded guns in the house. So how could Samuel Adams have advocated such protection unless he thought that the protection was consistent with local regulation that seriously impeded urban residents from using their arms against intruders? It seems unlikely that he meant to deprive the Federal Government of power (to enact Boston-type weapons regulation) that he know Boston had and (as far as we know) he would have thought constitutional under the Massachusetts Constitution. Indeed, since the District of Columbia (the subject of the Seat of Government Clause, U. S. Const., Art. I, §8, cl. 17) was the only urban area under direct federal control, it seems unlikely that the Framers thought about urban gun control at all. Cf. Palmore v. United States, 411 U. S. 389, 397–398 (1973) (Congress can “legislate for the District in a manner with respect to subjects that would exceed its powers, or at least would be very unusual, in the context of national legislation enacted under other powers delegated to it”).

Of course the District’s law and the colonial Boston law are not identical. But the Boston law disabled an even wider class of weapons (indeed, all firearms). And its existence shows at the least that local legislatures could impose (as here) serious restrictions on the right to use firearms. Moreover, as I have said, Boston’s law, though highly analogous to the District’s, was not the only colonial law that could have impeded a homeowner’s ability to shoot a burglar. Pennsylvania’s and New York’s laws could well have had a similar effect. See supra, at 6–7. And the Massachusetts and Pennsylvania laws were not only thought consistent with an unwritten common-law gun-possession right, but also consistent with written state constitutional provisions providing protections similar to those provided by the Federal Second Amendment . See supra, at 6–7. I cannot agree with the majority that these laws are largely uninformative because the penalty for violating them was civil, rather than criminal. Ante, at 61–62. The Court has long recognized that the exercise of a constitutional right can be burdened by penalties far short of jail time. See, e.g., Murdock v. Pennsylvania, 319 U. S. 105 (1943) (invalidating $7 per week solicitation fee as applied to religious group); see also Forsyth County v. Nationalist Movement, 505 U. S. 123, 136 (1992) (“A tax based on the content of speech does not become more constitutional because it is a small tax”).

Regardless, why would the majority require a precise colonial regulatory analogue in order to save a modern gun regulation from constitutional challenge? After all, insofar as we look to history to discover how we can constitutionally regulate a right to self-defense, we must look, not to what 18th-century legislatures actually did enact, but to what they would have thought they could enact. There are innumerable policy-related reasons why a legislature might not act on a particular matter, despite having the power to do so. This Court has “frequently cautioned that it is at best treacherous to find in congressional silence alone the adoption of a controlling rule of law.” United States v. Wells, 519 U. S. 482, 496 (1997) . It is similarly “treacherous” to reason from the fact that colonial legislatures did not enact certain kinds of legislation an unalterable constitutional limitation on the power of a modern legislature cannot do so. The question should not be whether a modern restriction on a right to self-defense duplicates a past one, but whether that restriction, when compared with restrictions originally thought possible, enjoys a similarly strong justification. At a minimum that similarly strong justification is what the District’s modern law, compared with Boston’s colonial law, reveals.

Fourth, a contrary view, as embodied in today’s decision, will have unfortunate consequences. The decision will encourage legal challenges to gun regulation throughout the Nation. Because it says little about the standards used to evaluate regulatory decisions, it will leave the Nation without clear standards for resolving those challenges. See ante, at 54, and n. 26. And litigation over the course of many years, or the mere specter of such litigation, threatens to leave cities without effective protection against gun violence and accidents during that time.

As important, the majority’s decision threatens severely to limit the ability of more knowledgeable, democratically elected officials to deal with gun-related problems. The majority says that it leaves the District “a variety of tools for combating” such problems. Ante, at 64. It fails to list even one seemingly adequate replacement for the law it strikes down. I can understand how reasonable individuals can disagree about the merits of strict gun control as a crime-control measure, even in a totally urbanized area. But I cannot understand how one can take from the elected branches of government the right to decide whether to insist upon a handgun-free urban populace in a city now facing a serious crime problem and which, in the future, could well face environmental or other emergencies that threaten the breakdown of law and order.


The majority derides my approach as “judge-empowering.” Ante, at 62. I take this criticism seriously, but I do not think it accurate. As I have previously explained, this is an approach that the Court has taken in other areas of constitutional law. See supra, at 10–11. Application of such an approach, of course, requires judgment, but the very nature of the approach—requiring careful identification of the relevant interests and evaluating the law’s effect upon them—limits the judge’s choices; and the method’s necessary transparency lays bare the judge’s reasoning for all to see and to criticize.

The majority’s methodology is, in my view, substantially less transparent than mine. At a minimum, I find it difficult to understand the reasoning that seems to underlie certain conclusions that it reaches.

The majority spends the first 54 pages of its opinion attempting to rebut Justice Stevens’ evidence that the Amendment was enacted with a purely militia-related purpose. In the majority’s view, the Amendment also protects an interest in armed personal self-defense, at least to some degree. But the majority does not tell us precisely what that interest is. “Putting all of [the Second Amendment ’s] textual elements together,” the majority says, “we find that they guarantee the individual right to possess and carry weapons in case of confrontation.” Ante, at 19. Then, three pages later, it says that “we do not read the Second Amendment to permit citizens to carry arms for any sort of confrontation.” Ante, at 22. Yet, with one critical exception, it does not explain which confrontations count. It simply leaves that question unanswered.

The majority does, however, point to one type of confrontation that counts, for it describes the Amendment as “elevat[ing] above all other interests the right of law-abiding, responsible citizens to use arms in defense of hearth and home.” Ante, at 63. What is its basis for finding that to be the core of the Second Amendment right? The only historical sources identified by the majority that even appear to touch upon that specific matter consist of an 1866 newspaper editorial discussing the Freedmen’s Bureau Act, see ante, at 43, two quotations from that 1866 Act’s legislative history, see ante, at 43–44, and a 1980 state court opinion saying that in colonial times the same were used to defend the home as to maintain the militia, see ante, at 52. How can citations such as these support the far-reaching proposition that the Second Amendment ’s primary concern is not its stated concern about the militia, but rather a right to keep loaded weapons at one’s bedside to shoot intruders?

Nor is it at all clear to me how the majority decides which loaded“arms” a homeowner may keep. The majority says that that Amendment protects those weapons “typically possessed by law-abiding citizens for lawful purposes.” Ante, at 53. This definition conveniently excludes machineguns, but permits handguns, which the majority describes as “the most popular weapon chosen by Americans for self-defense in the home.” Ante, at 57; see also ante, at 54–55. But what sense does this approach make? According to the majority’s reasoning, if Congress and the States lift restrictions on the possession and use of machineguns, and people buy machineguns to protect their homes, the Court will have to reverse course and find that the Second Amendment does, in fact, protect the individual self-defense-related right to possess a machinegun. On the majority’s reasoning, if tomorrow someone invents a particularly useful, highly dangerous self-defense weapon, Congress and the States had better ban it immediately, for once it becomes popular Congress will no longer possess the constitutional authority to do so. In essence, the majority determines what regulations are permissible by looking to see what existing regulations permit. There is no basis for believing that the Framers intended such circular reasoning.

I am similarly puzzled by the majority’s list, in Part III of its opinion, of provisions that in its view would survive Second Amendment scrutiny. These consist of (1) “prohibitions on carrying concealed weapons”; (2) “prohibitions on the possession of firearms by felons”; (3) “prohibitions on the possession of firearms by … the mentally ill”; (4) “laws forbidding the carrying of firearms in sensitive places such as schools and government buildings”; and (5) government “conditions and qualifications” attached “to the commercial sale of arms.” Ante, at 54. Why these? Is it that similar restrictions existed in the late 18th century? The majority fails to cite any colonial analogues. And even were it possible to find analogous colonial laws in respect to all these restrictions, why should these colonial laws count, while the Boston loaded-gun restriction (along with the other laws I have identified) apparently does not count? See supra, at 5–6, 38–39.

At the same time the majority ignores a more important question: Given the purposes for which the Framers enacted the Second Amendment , how should it be applied to modern-day circumstances that they could not have anticipated? Assume, for argument’s sake, that the Framers did intend the Amendment to offer a degree of self-defense protection. Does that mean that the Framers also intended to guarantee a right to possess a loaded gun near swimming pools, parks, and playgrounds? That they would not have cared about the children who might pick up a loaded gun on their parents’ bedside table? That they (who certainly showed concern for the risk of fire, see supra, at 5–7) would have lacked concern for the risk of accidental deaths or suicides that readily accessible loaded handguns in urban areas might bring? Unless we believe that they intended future generations to ignore such matters, answering questions such as the questions in this case requires judgment—judicial judgment exercised within a framework for constitutional analysis that guides that judgment and which makes its exercise transparent. One cannot answer those questions by combining inconclusive historical research with judicial ipse dixit.

The argument about method, however, is by far the less important argument surrounding today’s decision. Far more important are the unfortunate consequences that today’s decision is likely to spawn. Not least of these, as I have said, is the fact that the decision threatens to throw into doubt the constitutionality of gun laws throughout the United States. I can find no sound legal basis for launching the courts on so formidable and potentially dangerous a mission. In my view, there simply is no untouchable constitutional right guaranteed by the Second Amendment to keep loaded handguns in the house in crime-ridden urban areas.


For these reasons, I conclude that the District’s measure is a proportionate, not a disproportionate, response to the compelling concerns that led the District to adopt it. And, for these reasons as well as the independently sufficient reasons set forth by Justice Stevens, I would find the District’s measure consistent with the Second Amendment ’s demands.

With respect, I dissent.