Mr. Dan Collins

October 2, 2002

Chairman Leahy, Senator Hatch, and Members of the Committee, I appreciate the opportunity to testify here today on this important subject. The sexual abuse of children is an evil that no decent and civilized society can or should tolerate in any form. The harm inherent in abusive sexual conduct is bad enough; the fact that such abuse may be photographed or videotaped only multiplies the scope of the harm inflicted on the young victims. As the Supreme Court has recognized, because child pornography "permanently record[s] the victim's abuse," the very existence of such materials "causes the child victims continuing harm by haunting the children in years to come." Osborne v. Ohio, 495 U.S. 103, 111 (1990). With the advent of the Internet, this harm has been magnified exponentially: pedophiles can, with the click of a few keys, instantly make such materials available to literally thousands of persons. Moreover, as the Supreme Court has also stated, "evidence suggests that pedophiles use child pornography to seduce other children into sexual activity." Id. Accordingly, the Court has properly held that the First Amendment provides no protection to such materials and that the government has compelling interests that justify "attempting to stamp out this vice at all levels." Id. at 110; see also New York v. Ferber, 458 U.S. 747, 756-58 (1982).
Over the years, the Congress - by large bipartisan majorities - has enacted a number of statutes designed to address the serious problems presented by the manufacture, possession, and trafficking of child pornography. One such law, the Child Pornography Prevention Act of 1996, was favorably reported by this Committee by a vote of 16-2. The Department of Justice, in both the current and prior Administrations, vigorously defended the validity of this important law in the courts. Unfortunately, in April of this year, a divided Supreme Court held that this legislation was, in part, facially unconstitutional. The Department was obviously disappointed by the Court's decision. Nonetheless, we believe that the Court's decision and the Constitution leave the Congress with ample authority to enact a new, more narrowly focused statute that will allow the Government to accomplish its legitimate and compelling objectives without interfering with First Amendment freedoms.
The Department is deeply grateful to the leadership shown by the Congress in moving promptly to work with us to address this important issue. A bipartisan group of Representatives and Senators joined the Attorney General on May 1 to announce a legislative proposal aimed at strengthening the child pornography laws in the wake of the Supreme Court's decision. As that bill, which was introduced as H.R. 4623, moved through the House Judiciary Committee, we were pleased to work with Members on both sides of the aisle in further revising the bill so as to ensure that it would provide maximum protection to our Nation's children while complying with the Supreme Court's decision and the Constitution. Likewise, I have been pleased to meet with members of the staff of this Committee, on both the majority and minority side, in connection with the drafting of S.2520. As I will explain in more detail below, the overall approach of the two bills is, at a conceptual level, very similar. There are points of difference, to be sure, but we start from a fairly large area of common ground.
In order to explain how the two bills address the constitutional deficiencies in the 1996 Act that were identified by the Supreme Court, I would like first to briefly outline the Court's ruling. In Ashcroft v. Free Speech Coalition, 122 S.Ct. 1389 (2002), the Court addressed the constitutionality of two provisions of law. The first was 18 U.S.C. §2256(8)(B), which defines "child pornography" to include virtual child pornography, i.e., visual depictions that "appear[] to be" minors engaging in sexually explicit conduct. The second was 18 U.S.C. §2256(8)(D), which defines "child pornography" also to include materials that are pandered as child pornography - that is, visual depictions that are "advertised, promoted, presented, described, or distributed in such a manner that conveys the impression that the material is or contains a visual depiction of a minor engaging in sexually explicit conduct."
The Supreme Court found these two definitional provisions to be unconstitutionally overbroad. In particular, with respect to "virtual child pornography" covered by §2256(8)(B), the Court concluded that the definition extended far beyond the traditional reach of obscenity as described in Miller v. California, 413 U.S. 15 (1973), and thus could not be justified as a proscription of obscenity, see 122 S.Ct. at 1400-01; that New York v. Ferber, 458 U.S. 747 (1982), could not be extended to support a complete ban on virtual child pornography, see 122 S.Ct. at 1401-02; and that the "reasons the Government offers in support" of the prohibition of virtual child pornography were insufficient under the First Amendment, id. at 1405.
In particular, in defending the 1996 Act, the Government had argued that the existence of virtual child pornography threatened to render the laws against child pornography unenforceable, and that a ban on virtual child pornography, coupled with an affirmative defense allowing some defendants to prove that the material was made using only adults, struck a proper constitutional balance. Without reaching the question whether any sort of "affirmative defense" approach could be constitutional, the Court held that the affirmative defense in the 1996 Act was "incomplete and insufficient." 122 S.Ct. at 1405. In particular, the Court noted that the affirmative defense did not extend to possession offenses and that it only extended to materials produced with youthful-looking adults; materials made by using computer imaging were not eligible for the affirmative defense.
The Government had also argued that child pornography, whether actual or virtual, "whets the appetites" of pedophiles to engage in molestation. In concluding that this could not sustain the 1996 Act's virtual child pornography definition, the Court held that the Government had "shown no more than a remote connection between speech that might encourage thoughts or impulses and any resulting child abuse." 122 S.Ct. at 1403. The Court held that "[w]ithout a significantly stronger, more direct connection, the Government may not prohibit speech on the ground that it may encourage pedophiles to engage in illegal conduct." Id.
With respect to the "pandering" provision in 18 U.S.C. §2256(8)(D), the Court held that the provision was overbroad because it criminalized speech based "on how the speech is presented" rather than "on what is depicted." 122 S.Ct. at 1405.
By invalidating these important features of the 1996 Act, the Court's decision leaves the Government in an unsatisfactory position that the Department believes warrants a prompt legislative response. Already, defendants often contend that there is "reasonable doubt" as to whether a given computer image - and most prosecutions involve materials stored and exchanged on computers - was produced with an actual child or as a result of some other process. There are experts who are willing to testify to the same effect on the defendants' behalf. Moreover, as computer technology continues its rapid evolution, this problem will grow increasingly worse: trials will increasingly devolve into jury-confusing battles of experts arguing over the method of generating an image that, to all appearances, looks like it is the real thing. The end result would be that the Government may be able to prosecute effectively only in very limited cases, such as those in which it happens to be able to match the depictions to pictures in pornographic magazines produced before the development of computer imaging software or in which it can establish the identity of the victim. See, e.g., United States v. Fox, 248 F.3d 394, 403 (5th Cir. 2001) (government's computer expert testified on cross-examination that there was no way to determine whether the individuals depicted even exist), vacated, 122 S.Ct. 1602 (2002).
As Justice Thomas noted in his concurring opinion, "if technological advances thwart prosecution of 'unlawful speech,' the Government may well have a compelling interest in barring or otherwise regulating some narrow category of 'lawful speech' in order to enforce effectively laws against pornography made through the abuse of real children." 122 S.Ct. at 1406-07 (Thomas, J., concurring in the judgment). Similarly, Justice O'Connor noted in her opinion concurring in part and dissenting in part that, "given the rapid pace of advances in computer-graphics technology, the Government's concern is reasonable." Id. at 1409. Moreover, to avert serious harms, Congress may rely on reasonable predictive judgments, even when legislating in an area implicating freedom of speech. See Turner Broad. Sys. Inc. v. FCC, 520 U.S. 180, 210-11 (1997). We believe that Congress has a strong basis for concluding that the very existence of sexually explicit computer images that are virtually indistinguishable from images of real minors engaged in sexually explicit conduct poses a serious danger to future prosecutions involving child pornography. Indeed, we already have some sense of the impact of the Court's decision. The Ninth Circuit had invalidated the same provisions of law in 1999, and all accounts indicate that the number and scope of child pornography prosecutions brought by our prosecutors in the Ninth Circuit has been adversely impacted.
Inaction is, therefore, unacceptable. But let me also emphasize that, while we are disappointed with the Court's decision, we strongly believe that any legislation must respect the Court's decision and endeavor in good faith to resolve the constitutional deficiencies in the prior law that were identified by the Court.
The Supreme Court's decision in Free Speech Coalition leaves open two primary means for addressing the problems presented by virtual child pornography. First, the Court specifically left open the possibility that a more narrowly tailored regulation of virtual child pornography, coupled with a broader affirmative defense, could be constitutional. In addressing the adequacy of the affirmative defense contained in the 1996 Act, the Court noted that the use of an affirmative defense could raise constitutional issues, but the Court explicitly stated that it was not holding that an affirmative-defense approach was unconstitutional:
We need not decide, however, whether the Government could impose this burden on a speaker. Even if an affirmative defense can save a statute from First Amendment challenge, here the defense is incomplete and insufficient, even on its own terms.
122 S.Ct. at 1405. As Justice Thomas correctly noted in his concurring opinion, the majority opinion thus explicitly "leave[s] open the possibility that a more complete affirmative defense could save a statute's constitutionality." Id. at 1407.
Second, the Court's opinion in Free Speech Coalition leaves open the possibility of enacting additional laws designed to more effectively prohibit obscene materials containing depictions of children. The Court concluded that, because of its breadth, the prior law was "much more than a supplement to the existing federal prohibition on obscenity." 122 S.Ct. at 1399. But it did not foreclose the possibility that supplemental legislation aimed specifically at obscene depictions of children could properly be enacted. On the contrary, the Court went out of its way to note that obscenity doctrine may give the Government greater leeway when it comes to graphic depictions of sexual acts involving very young children:
While we have not had occasion to consider the question, we may assume that the apparent age of persons engaged in sexual conduct is relevant to whether a depiction offends community standards. Pictures of young children engaged in certain acts might be obscene where similar depictions of adults, or perhaps even older adolescents, would not.
122 S.Ct. at 1396.
H.R. 4623 and S.2520 properly draw on both of these approaches in crafting a range of complementary provisions that aim to further the Government's compelling interest in protecting children, while avoiding infringement of First Amendment rights.
Let me first address how both bills implement the "affirmative-defense" approach that was explicitly left open by the Supreme Court. Section 3 of H.R. 4623 would significantly revise the existing law's coverage of virtual child pornography by substantially narrowing the scope of the definition of "child pornography" and by simultaneously expanding the affirmative defense. Section 3 of H.R. 4623 eliminates both of the problems identified by the Court in the prior affirmative defense, and more narrowly focuses the statute on the Government's core concern about enforceability. Specifically, section 3 would make at least five significant changes to the prior law:
? The definition of virtual child pornography is explicitly limited to "computer image[s]" or "computer-generated image[s]." As a practical matter, it is the use of computers to traffic images of child pornography that implicates the core of the Government's practical concern about enforceability. The resulting prohibition is one that extends, not to the suppression of any idea, but rather to uses of particular instruments, such as computers, in a way that directly implicates the Government's compelling interest in keeping the child pornography laws enforceable.
? The definition of virtual child pornography is also revised to reach only images that are "indistinguishable" from actual child pornography. Again, the idea is that the Government's core interests are implicated by the sort of materials that, to an ordinary observer, could pass for the real thing. "[D]rawings, cartoons, sculptures, or paintings" - which cannot pass for the real thing - are specifically excluded from the scope of this provision.
? The definition of "sexually explicit conduct" has been narrowed with respect to virtual child pornography. In particular, "simulated" sexual intercourse would be covered only if it the depiction is "lascivious" and involves the exhibition of the "genitals, breast, or pubic area" of any person. Notably, this change alone eliminates most of the overbreadth identified by the Court; it was the breadth of the definition of "sexually explicit conduct" that led to distracting and unhelpful arguments over whether movies such as "Traffic" and "American Beauty" were covered.
? The affirmative defense is explicitly amended to include possession offenses.
? The affirmative defense is also amended so that a defendant could prevail simply by showing that no children were used in the production of the materials. Prior law only granted an affirmative defense for productions involving youthful-looking adults.
With these changes, "'whatever overbreadth may exist should be cured through case-by-case analysis of the fact situations to which its sanctions, assertedly, may not be applied.'" Ferber, 458 U.S. at 773-74 (quoting Broadrick v. Oklahoma, 413 U.S. 601, 615-16 (1973)).
Three provisions of S.2520 would, taken together, also appear aimed at implementing the "affirmative-defense" approach to regulating virtual child pornography. As a general matter, section 2 of the bill would amend 18 U.S.C. §2252A(c) so that the affirmative defense would apply to possession offenses and would also be available to any defendant who could show that no children were used in the production of the materials. As noted above, these were the two deficiencies that led the Supreme Court to hold that the existing affirmative defense was constitutionally inadequate. Section 4 of the bill would create a new subsection "(E)" to the definition of "child pornography" in 18 U.S.C. §2256(8); this new subsection, when read together with a third provision, would appear to extend to certain "virtual" materials. Specifically, proposed section 2256(8)(E) would define as proscribed child pornography any materials whose production "involves the use of an identifiable minor engaging in sexually explicit conduct." Section 4(3) of S.2520 would, in turn, amend the existing definition of "identifiable minor" in section 2256(9) so that it would include certain images depicting persons "who [are] virtually indistinguishable from an actual minor."
For several reasons, we believe that H.R. 4623's implementation of the "affirmative-defense" approach to regulating virtual child pornography is preferable. First, as a purely technical matter, S.2520's use of the definition of "identifiable minor" as the vehicle for covering virtual materials is unduly complicated and, as drafted, may be internally inconsistent. Literally construed, S.2520 would amend section 2256(9) so that it would only cover virtual materials that meet all of the following criteria: (1)the production of the material involves use of "a person," see 18 U.S.C. §2256(8)(E) (as proposed by S.2520); id., §2256(9)(A); (2)the person either (a) was a minor at the time of production or (b) is one whose image as a minor was used in the production, see id., §2256(9)(A)(i); and (3) the person is either (a) recognizable as an actual person or (b) virtually indistinguishable from an actual minor, see id., §2256(9)(A)(ii) (as proposed to be amended by S.2520). From the context, criteria (1) and (2) seem clearly to presume that "person" means an actual person, and that the person, as depicted in the image, is a minor. This would appear directly to conflict with criteria (3)'s effort to cover images that are "virtually indistinguishable" from an actual minor. Indeed, if an image must satisfy §2256(9)(a)(i)'s requirement that it be produced with a minor or with the image of a minor (criteria (2)), then, as a literal matter, the addition of a further requirement that the image be "virtually indistinguishable" from an actual minor is largely surplusage with no practical effect.
Second, although S.2520 and H.R. 4623 both fix the affirmative defense, S.2520 does not contain any of the three additional narrowing elements that H.R. 4623 would enact in order to reduce the substantial overbreadth in the underlying prohibition of virtual materials. As noted above, H.R. 4623 would narrow the definition of "sexually explicit conduct" insofar as it would apply to virtual materials; would specifically and narrowly define "indistinguishable"; and would only reach virtual materials that were computer-generated or that are stored in a computerized format. These additional limitations are meant to help ensure that the Government will best be able to defend the legislation as being sufficiently narrowly tailored. Because S.2520's regulation of virtual materials lacks these features, it is clearly more vulnerable to constitutional challenge than H.R. 4623.
For these reasons, we strongly urge adoption of H.R. 4623's provisions implementing an "affirmative-defense" approach to regulating virtual child pornography.
Let me now address how the two bills would strengthen the ability to go after obscene depictions of children. H.R. 4623 and S.2520 each contain two provisions that seek to implement this approach.
First, both bills would enact a specific prohibition of "obscene" materials that depict minors engaged in sexually explicit conduct. Section 5 of H.R. 4623 would add a new section 1466B to title 18 that would prohibit "obscene" materials "of any kind, including a drawing, cartoon, sculpture, or painting" that "depicts a pre-pubescent child engaging in sexually explicit conduct." No real children need be depicted and no real children need have been used in the production. Notably, proposed §1466B would explicitly incorporate the stricter penalties applicable to child pornography, and section 5(b) of the bill would likewise require application of the guidelines applicable to child pornography, which are substantially more strict than those applicable to obscenity offenses generally. S.2520 would amend existing §2256(8)(B) so that it would define as proscribed child pornography any "obscene" material that "is, or appears to be, of a minor engaging in sexually explicit conduct." Because the amendment is contained in §2256, the harsher penalty structure of the child pornography laws would likewise be carried over by the Senate bill.
These corresponding provisions of the two bills are thus quite close in purpose and effect. Nonetheless, with one important exception, we prefer the House version. By retaining the phrase "is, or appears to be, of a minor," S.2520 may create an argument that its prohibition is limited to obscene materials that are "virtually indistinguishable" from those of actual minors, since that was the meaning that the Justice Department had urged be given to this phrase in the existing statute. Because S.2520 would require proof that the material is "obscene" in the sense described in Miller v. California, 413 U.S. 15 (1973), there is no reason to impose any such additional restrictions (or to use language that could be read to import them). On the other hand, for the same exact reason, there is no reason why proposed §1466B in H.R. 4623 should be limited to depictions of "pre-pubescent" children. In addition, S.2520 does not contain an explicit directive to use the harsher sentencing guidelines associated with child pornography. This is needed to ensure that the courts will not view these as subject to the more lenient guidelines applicable to "obscenity" generally. Moreover, as a technical matter, S.2520's inclusion of the obscenity provision in a definition in the child pornography statute makes that statute somewhat complicated; it necessitates, for example, an explicit exception carving out "obscene" materials from the affirmative defense contained in that statute. It also conceivably could raise a question as to whether S.2520 intends to create two separate offenses that can both be charged with respect to the same set of materials, i.e., can the Government charge two violations of the same exact child pornography statute using two distinct definitions of "child pornography." Enacting an entirely separate obscenity provision, as in H.R. 4623, is cleaner and more straightforward.
Second, both bills contain a second provision that would regulate a narrowly defined class of materials as proscribable obscenity without requiring, in every case, a case-by-case examination of all three of the elements of the Miller test for obscenity.
Section 5 of H.R. 4623 would add a new section 1466A to title 18. Proposed §1466A would create a new obscenity offense that would generally prohibit the production, distribution, or possession of visual depictions of pre-pubescent children engaging in sexually explicit conduct, whether real or virtual. An individualized assessment of the Miller factors would not be required in every case. The penalties imposed on this subset of obscene materials would be the same as those for proposed §1466B, discussed above.
By creating a new provision that more narrowly focuses on pre-pubescent materials, proposed §1466A takes into account the fact that the Free Speech Coalition Court relied entirely on post-pubescent materials in finding that the prior law was substantially overbroad. Moreover, the Court specifically noted in its opinion that the age of the child depicted was an important consideration in determining whether a particular depiction was constitutionally unprotected obscenity: "Pictures of young children engaged in certain acts might be obscene where similar depictions of adults, or perhaps even older adolescents, would not." 122 S.Ct. at 1396.
Congress may reasonably conclude that the very narrow class of materials covered by proposed §1466A are the sort that would invariably satisfy the constitutional standards for obscenity set out in Miller, and that such materials therefore may be fully proscribed because they are constitutionally unprotected obscenity. The narrow class of images reached by proposed §1466A are precisely the sort that appeal to the worst form of prurient interest, that are patently offensive in light of any applicable community standards, and that lack serious literary, artistic, political, or scientific value in virtually any context. Once again, to the extent that there is any residual overbreadth, it is not substantial and may be satisfactorily addressed through case-by-case adjudication.
Section 4 of S.2520 would enact a new §2256(8)(D) that would effectively proscribe any depiction that is, or appears to be, a minor "actually engaging" in certain defined sexual activities and that "lacks serious literary, artistic, political, or scientific value." Thus, whereas the House bill would not require, in all cases, a case-by-case analysis of the Miller factors with respect to a narrowly defined class of "pre-pubescent" sexually explicit materials, the Senate bill would require a case-by-case analysis of only one of the Miller factors with respect to a somewhat more broadly defined class of materials that includes post-pubescent materials.
These approaches are, conceptually, not very far apart from one another. The most conservative approach would be to adopt the House bill's narrower definition of the subset of materials that would be subject to a "per se" obscenity approach and to combine it with the Senate bill's requirement of a case-by-case determination with respect to the third Miller factor, i.e., whether the material lacks serious literary, artistic, political, or scientific value. On the other hand, it may seem unwarranted to require a jury to ask whether, for example, a graphic depiction of the sexual abuse of a five-year-old has "literary, artistic, political, or scientific value." On balance, we prefer the House version as it stands.
Notably, the obscenity provisions of both bills would extend their prohibitions to simple possession of the obscene materials at issue. We strongly endorse this feature of both H.R. 4623 and S.2520. We do not believe that these provisions would be unconstitutional, notwithstanding the Supreme Court's 1969 decision in Stanley v. Georgia, 394 U.S. 557 (1969), which held that a State could not constitutionally criminalize the simple possession of obscenity in the privacy of a person's residence. Several points are worth noting in this regard:
? In Obsorne v. Ohio, 495 U.S. 103 (1990), the Court held that Stanley does not apply to the possession of child pornography involving actual children, and the Court specifically cautioned that "Stanley should not be read too broadly." Id. at108.
? The Court has explicitly rejected the contention "that [because] Stanley has firmly established the right to possess obscene material in the privacy of the home[, . . . ] this creates a correlative right to receive it, transport it, or distribute it." United States v. Orito, 413 U.S. 139, 141 (1973). The lower courts have likewise extended the rationale of Orito to, in effect, cover "home receipt" situations under several federal obscenity and child pornography laws. See, e.g., United States v. Hurt, 795 F.2d 765 (9th Cir. 1986), cert. denied, 484 U.S. 816 (1987); United States v. Kuennen, 901 F.2d 103 (8th Cir.), cert. denied, 498 U.S. 958 (1990); United States v. Hale, 784 F.2d 1465 (9th Cir.), cert. denied, 479 U.S. 829 (1986). Virtually all "possession" cases that would be prosecuted under the House and Senate bills will involve obscene materials that the defendant almost certainly received from someone else, and it makes little sense from a constitutional perspective to require the government to go through the mechanics of proving that the materials possessed by the defendant were unlawfully received.
? The possession prohibitions in the House and Senate bills are not premised "on the desirability of controlling a person's private thoughts." Stanley, 394 U.S. at 566. Instead, they are premised on the Government's substantial and legitimate interest in preventing such obscenity from "entering the stream of commerce" in the first instance, see Orito, 413 U.S. at 143. The vast majority of the materials in question are computer-generated images that are easily susceptible of being transmitted by possessors over interactive computer networks to others seeking the same sorts of images. This fundamentally distinguishes a possession case under the proposed bills from Stanley.
? Recent evidence establishing a significant causal link between possession of child pornography and molestation (or other sex crimes) also provides an additional basis for the prohibition on possession of such obscene materials.
In addition to these various provisions aimed at remedying the Supreme Court's invalidation of the virtual child pornography prohibition in 18 U.S.C. §2256(8)(B), both H.R. 4623 and S.2520 contain new provisions designed to replace the "pandering" provision (18 U.S.C. §2256(8)(D)) that was also struck down by the Court.
Section 3 of H.R. 4623 creates a new "pandering" provision that avoids the problems of the prior law. The Court sharply criticized the fact that prior law criminalized materials based on how they were marketed. Section 3, by contrast, would regulate the marketing itself by enacting a comprehensive prohibition on any offer to sell or buy "real" child pornography, without having to prove that any material was ever produced. This section presents no constitutional difficulty. There is no constitutional limitation on the ability of the legislature to establish inchoate offenses (attempt, conspiracy, solicitation, etc.) respecting conduct that is aimed at unlawful transactions. For example, offering to provide or sell illegal drugs can be criminalized, even where the offeror does not actually have such drugs in hand.
Section 2 of S.2520 would likewise enact a new prohibition on the advertising or promotion of any material "in a manner that conveys the impression that the material is, or contains, an obscene visual depiction of a minor engaging in sexually explicit conduct." This provision is, in our view, too narrow. There is no reason why a prohibition on pandering needs to be limited to "obscene" materials. Just as it is permissible to make it illegal to offer to sell drugs, even though the drugs are not in hand, so too the Congress can prohibit offers to buy or sell real child pornography, even if such materials are not in hand. Moreover, requiring that the Government show that the defendant has "convey[ed] the impression that the material is ... obscene" could be construed to require the Government to prove that the recipient developed a belief with respect to each of the three prongs of the Miller test. There is no reason to impose such a potentially onerous proof requirement. In addition, section 2 of S.2520 would treat as illegal "pandering" the act of "describing" any material as child pornography. The literal breadth of this prohibition undoubtedly exceeds what is intended (it would literally forbid, for example, a verbal description of the materials at issue in a child pornography prosecution). To avoid constitutional difficulties, the word "describes" should be deleted. Moreover, section 2's coverage of would-be receivers of child pornography - as opposed to purveyors - is unclear. Because the "conveys the impression" clause of section 2 seems to presume knowledge of the contents of a particular visual depiction, this phrasing does not fit recipients as comfortably as it does purveyors. For all of this reasons, we favor the provision in section 3 of H.R. 4623.
Both bills also contain provisions aimed at prohibiting the use of sexually explicit materials to facilitate offenses against minors. Section 6 of H.R. 4623 would criminalize the knowing use of, inter alia, obscenity, child pornography, or "indistinguishable" virtual pre-pubescent pornography in connection with the commission of certain specified offenses against minors. Section 6 of H.R. 4623 would also enact a straightfoward prohibition on showing any such materials to children. In turn, section 2 of S.2520 would prohibit the use of child pornography or virtual child pornography "for purposes of inducing or persuading [a] minor to participate in any activity that is illegal." We believe that the more expansive provisions in the House bill are preferable.
The remaining sections of H.R. 4623 and S.2520 would make a number of other important changes to strengthen the law in this vital area. Both bills contain provisions to strengthen penalties for repeat offenders. Both bills provide for strengthening and clarifying the existing reporting requirements applicable to internet service providers. Both bills would strengthen the extraterritorial application of child pornography laws.
S.2520 contains a number of additional provisions not found in the House bill. Among these are a requirement that a defendant provide advance notice of his intention to assert an affirmative defense under the child pornography laws, and civil remedies for victims of child pornography. We support inclusion of these two provisions in any legislation on this subject. We note, however, that section 2 of S.2520, as currently drafted, provides that, if a defendant fails to comply with the requirement to provide advance notice of an affirmative defense, the court should generally prohibit "the defendant from asserting a defense." We assume that the breadth of this phrasing, which would raise serious constitutional concerns, is unintentional and that what is intended is that the defendant would be precluded from asserting "such a defense."
We are sympathetic to the aims of section 5 of S.2520, which would amend the existing statute relating to recordkeeping requirements for the production of visual depictions of actual sexually explicit conduct. Section 5 would amend this provision so that it would extend to "computer generated image[s]". We believe that this provision needs more careful study. It may be more appropriate to craft a more tailored recordkeeping provision that would specifically apply to computer generated materials. Simply amending 18 U.S.C. §2257 so that is applies to computer generated images creates some degree of ambiguity. For example, it is not at all clear what is meant by a "computer generated image" of "actual sexually explicit conduct"; arguably, this is a contradiction in terms. Moreover, section 5 of S.2520 would only insert "computer generated image" in one subsection of the statute without making other conforming changes that would seemingly be required in other subsections. We would we happy to work with the Committee in attempting to craft a more suitable recordkeeping requirement. We do, however, support section 5(1)'s elimination of current law's unjustifiable limitation of the use of §2257 records in criminal prosecutions. We would also suggest that it would be appropriate to make a technical change to section 2257(f)(4), by changing "produce" to "produced".
We do not support section 3 of S.2520, which would make inadmissible "the name, address, or other identifying information, other than the age or approximate age, of any minor who is depicted in any child pornography." As literally drafted, this provision could preclude the admission of evidence necessary to prove that a particular item of child pornography was in fact produced using a real child. We believe that the privacy protection provisions of 18 U.S.C. §3509(d), together with the discretion afforded to district judges under the Federal Rules of Evidence, should be sufficient to address any concerns in this area.
We also do not support sections 12(a) or 12(b) of S.2520. While we are certainly sympathetic to section 12(a)'s requirement that additional attorneys be appointed to prosecute child pornography cases, S.2520 would provide no funding for that purpose. Section 12(b) would impose detailed reporting requirements that would divert resources that could be better spent on prosecuting child pornography cases.
We are not opposed to section 12(c), which would require the Sentencing Commission to promulgate guidelines with respect to certain of the amendments made by S.2520.
I would be pleased to answer any questions the Committee might have on this subject.