delivered the opinion of the Court.

At issue in this case is the constitutionality of a New York criminal statute which prohibits persons from knowingly promoting sexual performances by children under the age of 16 by distributing material which depicts such performances.

I

In recent years, the exploitive use of children in the production of pornography has become a serious national problem.1 The Federal Government and 47 States have sought to combat the problem with statutes specifically directed at the production of child pornography. At least half of such statutes do not require that the materials produced be legally obscene. Thirty-five States and the United States Congress have also passed legislation prohibiting the distribution of such materials; 20 States prohibit the distribution of material depicting children engaged in sexual conduct without requiring that the material be legally obscene.2

*750New York is one of the 20. In 1977, the New York Legislature enacted Article 263 of its Penal Law. N. Y. Penal Law, Art. 263 (McKinney 1980). Section 263.05 criminalizes as a class C felony the use of a child in a sexual performance:

“A person is guilty of the use of a child in a sexual performance if knowing the character and content thereof he employs, authorizes or induces a child less than sixteen years of age to engage in a sexual performance or being a parent, legal guardian or custodian of such child, *751he consents to the participation by such child in a sexual performance.”

A “[s]exual performance” is defined as “any performance or part thereof which includes sexual conduct by a child less than sixteen years of age.” § 263.00(1). “Sexual conduct” is in turn defined in §263.00(3):

“ ‘Sexual conduct’ means actual or simulated sexual intercourse, deviate sexual intercourse, sexual bestiality, masturbation, sado-masochistic abuse, or lewd exhibition of the genitals.”

A performance is defined as “any play, motion picture, photograph or dance” or “any other visual representation exhibited before an audience.” §263.00(4).

At issue in this case is §263.15, defining a class D felony:3

“A person is guilty of promoting a sexual performance by a child when, knowing the character and content thereof, he produces, directs or promotes any performance which includes sexual conduct by a child less than sixteen years of age.”

To “promote” is also defined:

“‘Promote’ means to procure, manufacture, issue, sell, give, provide, lend, mail, deliver, transfer, transmute, publish, distribute, circulate, disseminate, present, exhibit or advertise, or to offer or agree to do the same.” §263.00(5).

A companion provision bans only the knowing dissemination of obscene material. §263.10.

This case arose when Paul Ferber, the proprietor of a Man*752hattan bookstore specializing in sexually oriented products, sold two films to an undercover police officer. The films are devoted almost exclusively to depicting young boys masturbating. Ferber was indicted on two counts of violating §263.10 and two counts of violating §263.15, the two New York laws controlling dissemination of child pornography.4 After a jury trial, Ferber was acquitted of the two counts of promoting an obscene sexual performance, but found guilty of the two counts under §263.15, which did not require proof that the films were obscene. Ferber’s convictions were affirmed without opinion by the Appellate Division of the New York State Supreme Court. 74 App. Div. 2d 558, 424 N. Y. S. 2d 967 (1980).

The New York Court of Appeals reversed, holding that §263.15 violated the First Amendment. 52 N. Y. 2d 674, 422 N. E. 2d 523 (1981). The court began by noting that in light of § 263.10’s explicit inclusion of an obscenity standard, §263.15 could not be construed to include such a standard. Therefore, “the statute would . . . prohibit the promotion of materials which are traditionally entitled to constitutional protection from government interference under the First Amendment.” 52 N. Y. 2d, at 678, 422 N. E. 2d, at 525. Although the court recognized the State’s “legitimate interest in protecting the welfare of minors” and noted that this “interest may transcend First Amendment concerns,” id., at 679, 422 N. E. 2d, at 525-526, it nevertheless found two fatal defects in the New York statute. Section 263.15 was under-inclusive because it discriminated against visual portrayals of children engaged in sexual activity by not also prohibiting the distribution of films of other dangerous activity. It was also overbroad because it prohibited the distribution of materials produced outside the State, as well as materials, such as medical books and educational sources, which *753“deal with adolescent sex in a realistic but nonobscene manner.” 52 N. Y. 2d, at 681, 422 N. E. 2d, at 526. Two judges dissented. We granted the State’s petition for certiorari, 454 U. S. 1052 (1981), presenting the single question:

“To prevent the abuse of children who are made to engage in sexual conduct for commercial purposes, could the New York State Legislature, consistent with the First Amendment, prohibit the dissemination of material which shows children engaged in sexual conduct, regardless of whether such material is obscene?”

II

The Court of Appeals proceeded on the assumption that the standard of obscenity incorporated in §263.10, which follows the guidelines enunciated in Miller v. California, 413 U. S. 15 (1973),5 constitutes the appropriate line dividing protected from unprotected expression by which to measure a regulation directed at child pornography. It was on the premise that “nonobscene adolescent sex” could not be singled out for special treatment that the court found §263.15 “strikingly underinclusive.” Moreover, the assumption that the constitutionally permissible regulation of pornography could not be more extensive with respect to the distribution of material depicting children may also have led the court to conclude that a narrowing construction of §263.15 was unavailable.

The Court of Appeals’ assumption was not unreasonable in light of our decisions. This case, however, constitutes our first examination of a statute directed at and limited to depictions of sexual activity involving children. We believe our inquiry should begin with the question of whether a State has somewhat more freedom in proscribing works which portray sexual acts or lewd exhibitions of genitalia by children.

*754A

In Chaplinsky v. New Hampshire, 315 U. S. 568 (1942), the Court laid the foundation for the excision of obscenity from the realm of constitutionally protected expression:

“There are certain well-defined and narrowly limited classes of speech, the prevention and punishment of which have never been thought to raise any Constitutional problem. These include the lewd and obscene .... It has been well observed that such utterances are no essential part of any exposition of ideas, and are of such slight social value as a step to truth that any benefit that may be derived from them is clearly outweighed by the social interest in order and morality.” Id., at 571-572 (footnotes omitted).

Embracing this judgment, the Court squarely held in Roth v. United States, 354 U. S. 476 (1957), that “obscenity is not within the area of constitutionally protected speech or press.” Id., at 485. The Court recognized that “rejection of obscenity as utterly without redeeming social importance” was implicit in the history of the First Amendment: The original States provided for the prosecution of libel, blasphemy, and profanity, and the “universal judgment that obscenity should be restrained [is] reflected in the international agreement of over 50 nations, in the obscenity laws of all of the 48 states, and in the 20 obscenity laws enacted by Congress from 1842 to 1956.” Id., at 484-485 (footnotes omitted).

Roth was followed by 15 years during which this Court struggled with “the intractable obscenity problem.” Interstate Circuit, Inc. v. Dallas, 390 U. S. 676, 704 (1968) (opinion of Harlan, J.). See, e. g., Redrup v. New York, 386 U. S. 767 (1967). Despite considerable vacillation over the proper definition of obscenity, a majority of the Members of the Court remained firm in the position that “the States have a legitimate interest in prohibiting dissemination or exhibition of obscene material when the mode of dissemination carries with it a significant danger of offending the sensibil*755ities of unwilling recipients or of exposure to juveniles.” Miller v. California, supra, at 18-19 (footnote omitted); Stanley v. Georgia, 394 U. S. 557, 567 (1969); Ginsberg v. New York, 390 U. S. 629, 637-643 (1968); Interstate Circuit, Inc. v. Dallas, supra, at 690; Redrup v. New York, supra, at 769; Jacobellis v. Ohio, 378 U. S. 184, 195 (1964).

Throughout this period, we recognized “the inherent dangers of undertaking to regulate any form of expression.” Miller v. California, supra, at 23. Consequently, our difficulty was not only to assure that statutes designed to regulate obscene materials sufficiently defined what was prohibited, but also to devise substantive limits on what fell within the permissible scope of regulation. In Miller v. California, supra, a majority of the Court agreed that a “state offense must also be limited to works which, taken as a whole, appeal to the prurient interest in sex, which portray sexual conduct in a patently offensive way, and which, taken as a whole, do not have serious literary, artistic, political, or scientific value.” Id., at 24. Over the past decade, we have adhered to the guidelines expressed in Miller,6 which subsequently has been followed in the regulatory schemes of most States.7

*756B

The Miller standard, like its predecessors, was an accommodation between the State’s interests in protecting the “sensibilities of unwilling recipients” from exposure to pornographic material and the dangers of censorship inherent in unabashedly content-based laws. Like obscenity statutes, laws directed at the dissemination of child pornography run the risk of suppressing protected expression by allowing the hand of the censor to become unduly heavy. For the following reasons, however, we are persuaded that the States are entitled to greater leeway in the regulation of pornographic depictions of children.

First. It is evident beyond the need for elaboration that a State’s interest in “safeguarding the physical and psychologi*757cal well-being of a minor” is “compelling.” Globe Newspaper Co. v. Superior Court, 457 U. S. 596, 607 (1982). “A democratic society rests, for its continuance, upon the healthy, well-rounded growth of young people into full maturity as citizens.” Prince v. Massachusetts, 321 U. S. 158, 168 (1944). Accordingly, we have sustained legislation aimed at protecting the physical and emotional well-being of youth even when the laws have operated in the sensitive area of constitutionally protected rights. In Prince v. Massachusetts, supra, the Court held that a statute prohibiting use of a child to distribute literature on the street was valid notwithstanding the statute’s effect on a First Amendment activity. In Ginsberg v. New York, supra, we sustained a New York law protecting children from exposure to nonobscene literature. Most recently, we held that the Government’s interest in the “well-being of its youth” justified special treatment of indecent broadcasting received by adults as well as children. FCC v. Pacifica Foundation, 438 U. S. 726 (1978).

The prevention of sexual exploitation and abuse of children constitutes a government objective of surpassing importance. The legislative findings accompanying passage of the New York laws reflect this concern:

“[T]here has been a proliferation of exploitation of children as subjects in sexual performances. The care of children is a sacred trust and should not be abused by those who seek to profit through a commercial network based upon the exploitation of children. The public policy of the state demands the protection of children from exploitation through sexual performances.” 1977 N. Y. Laws, ch. 910, § l.8

*758We shall not second-guess this legislative judgment. Respondent has not intimated that we do so. Suffice it to say that virtually all of the States and the United States have passed legislation proscribing the production of or otherwise combating “child pornography.” The legislative judgment, as well as the judgment found in the relevant literature, is that the use of children as subjects of pornographic materials is harmful to the physiological, emotional, and mental health of the child.9 That judgment, we think, easily passes muster under the First Amendment.

*759Second. The distribution of photographs and films depicting sexual activity by juveniles is intrinsically related to the sexual abuse of children in at least two ways. First, the materials produced are a permanent record of the children’s participation and the harm to the child is exacerbated by their circulation.10 Second, the distribution network for child pornography must be closed if the production of material which requires the sexual exploitation of children is to be effectively controlled. Indeed, there is no serious contention that the legislature was unjustified in believing that it is difficult, if *760not impossible, to halt the exploitation of children by pursuing only those who produce the photographs and movies. While the production of pornographic materials is a low-profile, clandestine industry, the need to market the resulting products requires a visible apparatus of distribution. The most expeditious if not the only practical method of law enforcement may be to dry up the market for this material by imposing severe criminal penalties on persons selling, advertising, or otherwise promoting the product. Thirty-five States and Congress have concluded that restraints on the distribution of pornographic materials are required in order to effectively combat the problem, and there is a body of literature and testimony to support these legislative conclusions.11 Cf. United States v. Darby, 312 U. S. 100 (1941) (upholding federal restrictions on sale of goods manufactured in violation of Fair Labor Standards Act).

Respondent does not contend that the State is unjustified in pursuing those who distribute child pornography. Rather, he argues that it is enough for the State to prohibit the distribution of materials that are legally obscene under the Miller test. While some States may find that this approach properly accommodates its interests, it does not fol*761low that the First Amendment prohibits a State from going further. The Miller standard, like all general definitions of what may be banned as obscene, does not reflect the State’s particular and more compelling interest in prosecuting those who promote the sexual exploitation of children. Thus, the question under the Miller test of whether a work, taken as a whole, appeals to the prurient interest of the average person bears no connection to the issue of whether a child has been physically or psychologically harmed in the production of the work. Similarly, a sexually explicit depiction need not be “patently offensive” in order to have required the sexual exploitation of a child for its production. In addition, a work which, taken on the whole, contains serious literary, artistic, political, or scientific value may nevertheless embody the hardest core of child pornography. “It is irrelevant to the child [who has been abused] whether or not the material. . . has a literary, artistic, political or social value.” Memorandum of Assemblyman Lasher in Support of §263.15. We therefore cannot conclude that the Miller standard is a satisfactory solution to the child pornography problem.12

Third. The advertising and selling of child pornography provide an economic motive for and are thus an integral part of the production of such materials, an activity illegal throughout the Nation.13 “It rarely has been suggested that *762the constitutional freedom for speech and press extends its immunity to speech or writing used as an integral part of conduct in violation of a valid criminal statute.” Giboney v. Empire Storage & Ice Co., 336 U. S. 490, 498 (1949).14 We note that were the statutes outlawing the employment of children in these films and photographs fully effective, and the constitutionality of these laws has not been questioned, the First Amendment implications would be no greater than that presented by laws against distribution: enforceable production laws would leave no child pornography to be marketed.15

Fourth. The value of permitting live performances and photographic reproductions of children engaged in lewd sexual conduct is exceedingly modest, if not de minimis. We consider it unlikely that visual depictions of children performing sexual acts or lewdly exhibiting their genitals would often constitute an important and necessary part of a literary per*763formance or scientific or educational work. As a state judge in this case observed, if it were necessary for literary or artistic value, a person over the statutory age who perhaps looked younger could be utilized.16 Simulation outside of the prohibition of the statute could provide another alternative. Nor is there any question here of censoring a particular literary theme or portrayal of sexual activity. The First Amendment interest is limited to that of rendering the portrayal somewhat more “realistic” by utilizing or photographing children.

Fifth. Recognizing and classifying child pornography as a category of material outside the protection of the First Amendment is not incompatible with our earlier decisions. “The question whether speech is, or is not, protected by the First Amendment often depends on the content of the speech.” Young v. American Mini Theatres, Inc., 427 U. S. 50, 66 (1976) (opinion of Stevens, J., joined by Burger, C. J., and White and Rehnquist, JJ.). See also FCC v. Pacifica Foundation, 438 U. S. 726, 742-748 (1978) (opinion of Stevens, J., joined by Burger, C. J., and Rehnquist, J.). “[I]t is the content of [an] utterance that determines whether it is a protected epithet or an unprotected ‘fighting comment.’” Young v. American Mini Theatres, Inc., supra, at 66. See Chaplinsky v. New Hampshire, 315 U. S. 568 (1942). Leaving aside the special considerations when public officials are the target, New York Times Co. v. Sullivan, 376 U. S. 254 (1964), a libelous publication is not protected by the Constitution. Beauhamais v. Illinois, 343 U. S. 250 (1952). Thus, it is not rare that a content-based classification of speech has been accepted because it may be appropriately generalized that within the confines of the given classification, the evil to be restricted so overwhelmingly outweighs *764the expressive interests, if any, at stake, that no process of case-by-case adjudication is required. When a definable class of material, such as that covered by § 26B. 15, bears so heavily and pervasively on the welfare of children engaged in its production, we think the balance of competing interests is clearly struck and that it is permissible to consider these materials as without the protection of the First Amendment.

C

There are, of course, limits on the category of child pornography which, like obscenity, is unprotected by the First Amendment. As with all legislation in this sensitive area, the conduct to be prohibited must be adequately defined by the applicable state law, as written or authoritatively construed. Here the nature of the harm to be combated requires that the state offense be limited to works that visually depict sexual conduct by children below a specified age.17 The category of “sexual conduct” proscribed must also be suitably limited and described.

The test for child pornography is separate from the obscenity standard enunciated in Miller, but may be compared to it for the purpose of clarity. The Miller formulation is adjusted in the following respects: A trier of fact need not find that the material appeals to the prurient interest of the average person; it is not required that sexual conduct portrayed be done so in a patently offensive manner; and the material at issue need not be considered as a whole. We note that the distri*765bution of descriptions or other depictions of sexual conduct, not otherwise obscene, which do not involve live performance or photographic or other visual reproduction of live performances, retains First Amendment protection. As with obscenity laws, criminal responsibility may not be imposed without some element of scienter on the part of the defendant. Smith v. California, 361 U. S. 147 (1959); Hamling v. United States, 418 U. S. 87 (1974).

D

Section 263.15’s prohibition incorporates a definition of sexual conduct that comports with the above-stated principles. The forbidden acts to be depicted are listed with sufficient precision and represent the kind of conduct that, if it were the theme of a work, could render it legally obscene: “actual or simulated sexual intercourse, deviate sexual intercourse, sexual bestiality, masturbation, sado-masochistic abuse, or lewd exhibition of the genitals.” § 263.00(3). The term “lewd exhibition of the genitals” is not unknown in this area and, indeed, was given in Miller as an example of a permissible regulation. 413 U. S., at 25. A performance is defined only to include live or visual depictions: “any play, motion picture, photograph or dance . . . [or] other visual representation exhibited before an audience.” §263.00(4). Section 263.15 expressly includes a scienter requirement.

We hold that §263.15 sufficiently describes a category of material the production and distribution of which is not entitled to First Amendment protection. It is therefore clear that there is nothing unconstitutionally “underinclusive” about a statute that singles out this category of material for proscription.18 It also follows that the State is not barred by *766the First Amendment from prohibiting the distribution of unprotected materials produced outside the State.19

Ill

It remains to address the claim that the New York statute is unconstitutionally overbroad because it would forbid the distribution of material with serious literary, scientific, or educational value or material which does not threaten the harms sought to be combated by the State. Respondent prevailed on that ground below, and it is to that issue that we now turn.

The New York Court of Appeals recognized that over-breadth scrutiny has been limited with respect to conduct-related regulation, Broadrick v. Oklahoma, 413 U. S. 601 (1973), but it did not apply the test enunciated in Broadrick because the challenged statute, in its view, was directed at “pure speech.” The court went on to find that §263.15 was fatally overbroad: “[T]he statute would prohibit the showing of any play or movie in which a child portrays a defined sexual act, real or simulated, in a nonobscene manner. It would also prohibit the sale, showing, or distributing of medical or educational materials containing photographs of such acts. *767Indeed, by its terms, the statute would prohibit those who oppose such portrayals from providing illustrations of what they oppose.” 52 N. Y. 2d, at 678, 422 N. E. 2d, at 525.

While the construction that a state court gives a state statute is not a matter subject to our review, Wainwright v. Stone, 414 U. S. 21, 22-28 (1973); Gooding v. Wilson, 405 U. S. 518, 520 (1972), this Court is the final arbiter of whether the Federal Constitution necessitated-the invalidation of a state law. It is only through this process of review that we may correct erroneous applications of the Constitution that err on the side of an overly broad reading of our doctrines and precedents, as well as state-court decisions giving the Constitution too little shrift. A state court is not free to avoid a proper facial attack on federal constitutional grounds. Bigelow v. Virginia, 421 U. S. 809, 817 (1975). By the same token, it should not be compelled to entertain an overbreadth attack when not required to do so by the Constitution.

A

The traditional rule is that a person to whom a statute may constitutionally be applied may not challenge that statute on the ground that it may conceivably be applied unconstitutionally to others in situations not before the Court. Broadrick v. Oklahoma, supra, at 610; United States v. Raines, 362 U. S. 17, 21 (1960); Carmichael v. Southern Coal & Coke Co., 301 U. S. 495, 513 (1937); Yazoo & M. V. R. Co. v. Jackson Vinegar Co., 226 U. S. 217, 219-220 (1912). In Broadrick, we recognized that this rule reflects two cardinal principles of our constitutional order: the personal nature of constitutional rights, McGowan v. Maryland, 366 U. S. 420, 429 (1961), and prudential limitations on constitutional adjudication.20 In United States v. Raines, supra, at 21, we *768noted the “incontrovertible proposition” that it “‘would indeed be undesirable for this Court to consider every conceivable situation which might possibly arise in the application of complex and comprehensive legislation/” (quoting Barrows v. Jackson, 346 U. S. 249, 256 (1953)). By focusing on the factual situation before us, and similar cases necessary for development of a constitutional rule,21 we face “flesh-and-blood”22 legal problems with data “relevant and adequate to an informed judgment.”23 This practice also fulfills a valuable institutional purpose: it allows state courts the opportunity to construe a law to avoid constitutional infirmities.

What has come to be known as the First Amendment overbreadth doctrine is one of the few exceptions to this principle and must be justified by “weighty countervailing policies.” United States v. Raines, supra, at 22-23. The doctrine is predicated on the sensitive nature of protected expression: “persons whose expression is constitutionally protected may well refrain from exercising their rights for fear of criminal sanctions by a statute susceptible of application to protected expression.” Village of Schaumburg v. *769Citizens for a Better Environment, 444 U. S. 620, 634 (1980); Gooding v. Wilson, supra, at 521. It is for this reason that we have allowed persons to attack overly broad statutes even though the conduct of the person making the attack is clearly unprotected and could be proscribed by a law drawn with the requisite specificity. Dombrowski v. Pfister, 380 U. S. 479, 486 (1965); Thornhill v. Alabama, 310 U. S. 88, 97-98 (1940); United States v. Raines, supra, at 21-22; Gooding v. Wilson, supra, at 521.

The scope of the First Amendment overbreadth doctrine, like most exceptions to established principles, must be carefully tied to the circumstances in which facial invalidation of a statute is truly warranted. Because of the wide-reaching effects of striking down a statute on its face at the request of one whose own conduct may be punished despite the First Amendment, we have recognized that the overbreadth doctrine is “strong medicine” and have employed it with hesitation, and then “only as a last resort.” Broadrick, 413 U. S., at 613. We have, in consequence, insisted that the over-breadth involved be “substantial” before the statute involved will be invalidated on its face.24

*770In Broadrick, we explained the basis for this requirement:

“[T]he plain import of our cases is, at the very least, that facial overbreadth adjudication is an exception to our traditional rules of practice and that its function, a limited one at the outset, attenuates as the otherwise unprotected behavior that it forbids the State to sanction moves from ‘pure speech’ toward conduct and that conduct — even if expressive — falls within the scope of otherwise valid criminal laws that reflect legitimate state interests in maintaining comprehensive controls over harmful, constitutionally unprotected conduct. Although such laws, if too broadly worded, may deter protected speech to some unknown extent, there comes a point where that effect — at best a prediction — cannot, with confidence, justify invalidating a statute on its face and so prohibiting a State from enforcing the statute against conduct that is admittedly within its power to proscribe. Cf. Alderman v. United States, 394 U. S. 165, 174-175 (1969).” Id., at 615.

We accordingly held that “particularly where conduct and not merely speech is involved, we believe that the overbreadth of a statute must not only be real, but substantial as well, judged in relation to the statute’s plainly legitimate sweep.” Ibid.25

*771Broadrick examined a regulation involving restrictions on political campaign activity, an area not considered “pure speech,” and thus it was unnecessary to consider the proper overbreadth test when a law arguably reaches traditional forms of expression such as books and films. As we intimated in Broadrick, the requirement of substantial over-breadth extended “at the very least” to cases involving conduct plus speech. This case, which poses the question squarely, convinces us that the rationale of Broadrick is sound and should be applied in the present context involving the harmful employment of children to make sexually explicit materials for distribution.

The premise that a law should not be invalidated for overbreadth unless it reaches a substantial number of impermissible applications is hardly novel. On most occasions involving facial invalidation, the Court has stressed the embracing sweep of the statute over protected expression.26 *772Indeed, Justice Brennan observed in his dissenting opinion in Broadrick:

“We have never held that a statute should be held invalid on its face merely because it is possible to conceive of a single impermissible application, and in that sense a requirement of substantial overbreadth is already implicit in the doctrine.” Id., at 630.

The requirement of substantial overbreadth is directly derived from the purpose and nature of the doctrine. While a sweeping statute, or one incapable of limitation, has the potential to repeatedly chill the' exercise of expressive activity by many individuals, the extent of deterrence of protected speech can be expected to decrease with the declining reach of the regulation.27 This observation appears equally applicable to the publication of books and films as it is to activities, such as picketing or participation in election campaigns, which have previously been categorized as involving conduct plus speech. We see no appreciable difference between the position of a publisher or bookseller in doubt as to the reach of New York’s child pornography law and the situation faced by the Oklahoma state employees with respect to that State’s restriction on partisan political activity. Indeed, it could reasonably be argued that the bookseller, with an economic incentive to sell materials that may fall within the statute’s scope, may be less likely to be deterred than the employee who wishes to engage in political campaign activity. Cf. Bates v. State Bar of Arizona, 433 U. S. 350, 380-381 (1977) (overbreadth analysis inapplicable to commercial speech).

This requirement of substantial overbreadth may justifiably be applied to statutory challenges which arise in defense *773of a criminal prosecution as well as civil enforcement or actions seeking a declaratory judgment. Cf. Parker v. Levy, 417 U. S. 733, 760 (1974). Indeed, the Court’s practice when confronted with ordinary criminal laws that are sought to be applied against protected conduct is not to invalidate the law in toto, but rather to reverse the particular conviction. Cantwell v. Connecticut, 310 U. S. 296 (1940); Edwards v. South Carolina, 372 U. S. 229 (1973). We recognize, however, that the penalty to be imposed is relevant in determining whether demonstrable overbreadth is substantial. We simply hold that the fact that a criminal prohibition is involved does not obviate the need for the inquiry or a priori warrant a finding of substantial overbreadth.

B

Applying these principles, we hold that § 263.15 is not substantially overbroad. We consider this the paradigmatic case of a state statute whose legitimate reach dwarfs its arguably impermissible applications. New York, as we have held, may constitutionally prohibit dissemination of material specified in §263.15. While the reach of the statute is directed at the hard core of child pornography, the Court of Appeals was understandably concerned that some protected expression, ranging from medical textbooks to pictorials in the National Geographic would fall prey to the statute. How often, if ever, it may be necessary to employ children to engage in conduct clearly within the reach of §263.15 in order to produce educational, medical, or artistic works cannot be known with certainty. Yet we seriously doubt, and it has not been suggested, that these arguably impermissible applications of the statute amount to more than a tiny fraction of the materials within the statute’s reach. Nor will we assume that the New York courts will widen the possibly invalid reach of the statute by giving an expansive construction to the proscription on “lewd exhibition[s] of the genitals.” Under these circumstances, §263.15 is “not substantially overbroad and . . . whatever overbreadth may exist *774should be cured through case-by-case analysis of the fact situations to which its sanctions, assertedly, may not be applied.” Broadrick v. Oklahoma, 413 U. S., at 615-616.

IV

Because § 263.15 is not substantially overbroad, it is unnecessary to consider its application to material that does not depict sexual conduct of a type that New York may restrict consistent with the First Amendment. As applied to Paul Ferber and to others who distribute similar material, the statute does not violate the First Amendment as applied to the States through the Fourteenth.28 The judgment of the New York Court of Appeals is reversed, and the case is remanded to that court for further proceedings not inconsistent with this opinion.

So ordered.

concurring.

Although I join the Court’s opinion, I write separately to stress that the Court does not hold that New York must except “material with serious literary, scientific, or educational value,” ante, at 766, from its statute. The Court merely holds that, even if the First Amendment shelters such material, New York’s current statute is not sufficiently overbroad to support respondent’s facial attack. The compelling interests identified in today’s opinion, see ante, at 756-764, suggest that the Constitution might in fact permit New York to ban knowing distribution of works depicting minors engaged in explicit sexual conduct, regardless of the social value of the depictions. For example, a 12-year-old child photographed while *775masturbating surely suffers the same psychological harm whether the community labels the photograph “edifying” or “tasteless.” The audience’s appreciation of the depiction is simply irrelevant to New York’s asserted interest in protecting children from psychological, emotional, and mental harm.

An exception for depictions of serious social value, moreover, would actually increase opportunities for the content-based censorship disfavored by the First Amendment. As drafted, New York’s statute does not attempt to suppress the communication of particular ideas. The statute permits discussion of child sexuality, forbidding only attempts to render the “portrayal[s] somewhat more ‘realistic’ by utilizing or photographing children.” Ante, at 763. Thus, the statute attempts to protect minors from abuse without attempting to restrict the expression of ideas by those who might use children as live models.

On the other hand, it is quite possible that New York’s statute is overbroad because it bans depictions that do not actually threaten the harms identified by the Court. For example, clinical pictures of adolescent sexuality, such as those that might appear in medical textbooks, might not involve the type of sexual exploitation and abuse targeted by New York’s statute. Nor might such depictions feed the poisonous “kiddie porn” market that New York and other States have attempted to regulate. Similarly, pictures of children engaged in rites widely approved by their cultures, such as those that might appear in issues of the National Geographic, might not trigger the compelling interests identified by the Court. It is not necessary to address these possibilities further today, however, because this potential overbreadth is not sufficiently substantial to warrant facial invalidation of New York’s statute.

concurring in the judgment.

I agree with much of what is said in the Court’s opinion. As I made clear in the opinion I delivered for the Court in *776Ginsburg v. New York, 390 U. S. 629 (1968), the State has a special interest in protecting the well-being of its youth. Id., at 638-641. See also Globe Newspaper Co. v. Superior Court, 457 U. S. 596, 607 (1982). This special and compelling interest, and the particular vulnerability of children, afford the State the leeway to regulate pornographic material, the promotion of which is harmful to children, even though the State does not have such leeway when it seeks only to protect consenting adults from exposure to such material. Ginsburg v. New York, supra, at 637, 638, n. 6, 642-643, n. 10. See also Jacobellis v. Ohio, 378 U. S. 184, 195 (1964) (opinion of Brennan, J.). I also agree with the Court that the “tiny fraction,” ante, at 773, of material of serious artistic, scientific, or educational value that could conceivably fall within the reach of the statute is insufficient to justify striking the statute on the grounds of overbreadth. See Broadrick v. Oklahoma, 413 U. S. 601, 630 (1973) (Brennan, J., dissenting).

But in my view application of §263.15 or any similar statute to depictions of children that in themselves do have serious literary, artistic, scientific, or medical value, would violate the First Amendment. As the Court recognizes, the limited classes of speech, the suppression of which does not raise serious First Amendment concerns, have two attributes. They are of exceedingly “slight social value,” and the State has a compelling interest in their regulation. See Chaplinsky v. New Hampshire, 315 U. S. 568, 571-572 (1942). The First Amendment value of depictions of children that are in themselves serious contributions to art, literature, or science, is, by definition, simply not “de minimis.” See ante, at 761. At the same time, the State’s interest in suppression of such materials is likely to be far less compelling. For the Court’s assumption of harm to the child resulting from the “permanent record” and “circulation” of the child’s “participation,” ante, at 759, lacks much of its force where the depiction is a serious contribution to art or science. The production of materials of serious value is not the “low-*777profile, clandestine industry” that according to the Court produces purely pornographic materials. See ante, at 760. In short, it is inconceivable how a depiction of a child that is itself a serious contribution to the world of art or literature or science can be deemed “material outside the protection of the First Amendment.” See ante, at 763.

I, of course, adhere to my view that, in the absence of exposure, or particular harm, to juveniles or unconsenting adults, the State lacks power to suppress sexually oriented materials. See, e. g., Paris Adult Theatre I v. Slaton, 413 U. S. 49, 73 (1973) (Brennan, J., dissenting). With this understanding, I concur in the Court’s judgment in this case.

concurring in the judgment.

Two propositions seem perfectly clear to me. First, the specific conduct that gave rise to this criminal prosecution is not protected by the Federal Constitution; second, the state statute that respondent violated prohibits some conduct that is protected by the First Amendment. The critical question, then, is whether this respondent, to whom the statute may be applied without violating the Constitution, may challenge the statute on the ground that it conceivably may be applied unconstitutionally to others in situations not before the Court. I agree with the Court’s answer to this question but not with its method of analyzing the issue.

Before addressing that issue, I shall explain why respondent’s conviction does not violate the Constitution. The two films that respondent sold contained nothing more than lewd exhibition; there is no claim that the films included any material that had literary, artistic, scientific, or educational value.1 Respondent was a willing participant in a commercial market that the State of New York has a legitimate interest in suppressing. The character of the State’s interest in protecting children from sexual abuse justifies the imposition *778of criminal sanctions against those who profit, directly or indirectly, from the promotion of such films. In this respect my evaluation of this case is different from the opinion I have expressed concerning the imposition of criminal sanctions for the promotion of obscenity in other contexts.2

A holding that respondent may be punished for selling these two films does not require us to conclude that other users of these very films, or that other motion pictures containing similar scenes, are beyond the pale of constitutional protection. Thus, the exhibition of these films before a legislative committee studying a proposed amendment to a state law, or before a group of research scientists studying human behavior, could not, in my opinion, be made a crime. Moreover, it is at least conceivable that a serious work of art, a documentary on behavioral problems, or a medical or psychiatric teaching device, might include a scene from one of these films and, when viewed as a whole in a proper setting, be entitled to constitutional protection. The question whether a specific act of communication is protected by the First Amendment always requires some consideration of both its content and its context.

The Court’s holding that this respondent may not challenge New York’s statute as overbroad follows its discussion of the contours of the category of nonobscene child pornography that New York may legitimately prohibit. Having defined that category in an abstract setting,3 the Court makes the *779empirical judgment that the arguably impermissible application of the New York statute amounts to only a “tiny fraction of the materials within the statute’s reach.” Ante, at 773. Even assuming that the Court’s empirical analysis is sound,4 I believe a more conservative approach to the issue would adequately vindicate the State’s interest in protecting its children and cause less harm to the federal interest in free expression.

A hypothetical example will illustrate my concern. Assume that the operator of a New York motion picture theater specializing in the exhibition of foreign feature films is offered a full-length movie containing one scene that is plainly lewd if viewed in isolation but that nevertheless is part of a serious work of art. If the child actor resided abroad, New York’s interest in protecting its young from sexual exploitation would be far less compelling than in the case before us. The federal interest in free expression would, however, be just as strong as if an adult actor had been used. There are at least three different ways to deal with the statute’s potential application to that sort of case.

First, at one extreme and as the Court appears to hold, the First Amendment inquiry might be limited to determining *780whether the offensive scene, viewed in isolation, is lewd. When the constitutional protection is narrowed in this drastic fashion, the Court is probably safe in concluding that only a tiny fraction of the materials covered by the New York statute is protected. And with respect to my hypothetical exhibitor of foreign films, he need have no uncertainty about the permissible application of the statute; for the one lewd scene would deprive the entire film of any constitutional protection.

Second, at the other extreme and as the New York Court of Appeals correctly perceived, the application of this Court’s cases requiring that an obscenity determination be based on the artistic value of a production taken as a whole would afford the exhibitor constitutional protection and result in a holding that the statute is invalid because of its overbreadth. Under that approach, the rationale for invalidating the entire statute is premised on the concern that the exhibitor’s understanding about its potential reach could cause him to engage in self-censorship. This Court’s approach today substitutes broad, unambiguous, state-imposed censorship for the self-censorship that an overbroad statute might produce.

Third, as an intermediate position, I would refuse to apply overbreadth analysis for reasons unrelated to any prediction concerning the relative number of protected communications that the statute may prohibit. Specifically, I would postpone decision of my hypothetical case until it actually arises. Advocates of a liberal use of overbreadth analysis could object to such postponement on the ground that it creates the risk that the exhibitor’s uncertainty may produce self-censorship. But that risk obviously interferes less with the interest in free expression than does an abstract, advance ruling that the film is simply unprotected whenever it contains a lewd scene, no matter how brief.

My reasons for avoiding overbreadth analysis in this case are more qualitative than quantitative. When we follow our *781traditional practice of adjudicating difficult and novel constitutional questions only in concrete factual situations, the adjudications tend to be crafted with greater wisdom. Hypothetical rulings are inherently treacherous and prone to lead us into unforeseen errors; they are qualitatively less reliable than the products of case-by-case adjudication.

Moreover, it is probably safe to assume that the category of speech that is covered by the New York statute generally is of a lower quality than most other types of communication. On a number of occasions, I have expressed the view that the First Amendment affords some forms of speech more protection from governmental regulation than other forms of speech.5 Today the Court accepts this view, putting the category of speech described in the New York statute in its rightful place near the bottom of this hierarchy. Ante, at 761-763. Although I disagree with the Court’s position that such speech is totally without First Amendment protection, I agree that generally marginal speech does not warrant the extraordinary protection afforded by the overbreadth doctrine.6

Because I have no difficulty with the statute’s application in this case, I concur in the Court’s judgment.