Notice of Proposed Rulemaking Proposed Mark for Sexually Oriented Spam February 17, 2004

BEFORE THE FEDERAL TRADE COMMISSION WASHINGTON, D.C.

In the Matter of

Project No. P044405

Notice of Proposed Rulemaking Proposed Mark for Sexually Oriented Spam

Comments of the Center for Democracy and Technology

The Center for Democracy and Technology (CDT) is an independent, non-profit public interest organization advocating democratic values and constitutional liberties in the digital age. CDT has been actively involved in policy and consumer education efforts to reduce the amount of spam. CDT has conducted research on the origins of spam and what kinds of online behavior make computer users more likely to receive spam. We have convened consultations of experts, industry representatives, and consumer advocates to consider proposed technology and policy solutions, their effectiveness, their viability for industry, and their potential impact on speech online. On the basis of the insights gathered through these ongoing efforts, CDT respectfully submits the following comments.

1. The provision in the CAN-SPAM Act requiring the labeling of email that contains sexually explicit content does not pass constitutional muster.

Question 16 in the Commission's Notice of Proposed (NPRM) asks that commentors identify any relevant federal, state or local rules that may . . . conflict with the proposed rule. CDT appreciates that the Commission is mandated to adopt the rule, and we know that the Commission cannot decide that an act of Congress is unconstitutional. However, we must reiterate briefly our position that there is indeed a very important federal rule that conflicts with the proposed labeling requirement, namely the First Amendment to the Constitution. Because of the burdens the proposed label will place on senders of constitutionally protected email, the labeling requirement suffers from the same flaws that have ensnared other government efforts to regulate content on the Internet.

First, it is unclear whether the labeling requirement will do much to further the government's interest in protecting consumers from unwanted sexually oriented email messages, assuming that is a constitutionally legitimate government interest. At best the labeling requirement will address only spam originating in the United States. The requirement will likely not be enforceable against off shore spammers, so that consumers will not receive the warning label on spam from overseas, where a large percentage of spam originates. Further, domestic spammers are notoriously elusive and not distinguished by their compliance with the law. A recent study suggests that a fairly low level of compliance with another disclosure provision of the CAN SPAM Act.[1] In the first two weeks of January, EmailLabs, an email technology provider, conducted an audit of email messages from online retailers, media and publishing companies. The review found that while more than 95% included an unsubscribe mechanism as required by the new law, only 56% of the emails complied with the requirement to include a postal mailing address. And the survey focused on legitimate marketers and publishers. In light of these findings, compliance with the sexually-oriented labeling requirement may be spotty, even among spammers within the United States.

However, to the extent that the labeling requirement is effective, it will be effective in an unconstitutional way, for the label, as used by service providers, will prevent senders of lawful material from reaching willing recipients. Question 10 of the NPRM highlights this constitutional vulnerability when it asks whether the proposed rule would unduly burden either entities selling sexually oriented material through email messages or consumers who were interested in purchasing sexually oriented material offered to them through email messages. Precisely because the label is intended to interface with ISP filters, the rule would burden senders of lawful, sexually oriented material.

A fundamental flaw in the labeling approach taken by the statute is that it is designed to promote filtering by the ISPs and takes control away from the end user. Ideally, mechanisms to reduce spam should be modeled on a user empowerment approach, wherein the user, taking advantage of filtering software on her computer, is the ultimate arbiter of what content she receives. The Commission should consider whether the labeling provisions of the CAN SPAM Act could be implemented in such a way as to provide the user with information about the nature of the content of the email without allowing ISPs to make that choice.

Not only does the CAN SPAM Act[2] cover clearly legal material, its definitions also lack clarity. While some legal content falls squarely within the terms of the rule, other senders of lawful content that discusses, for example, issues related to sexuality or sexual health, will be placed in the difficult position of guessing whether their content must be labeled. Recipients of that legitimate material will be burdened because the filtering technologies that the labeling provision is specifically intended to facilitate will prevent it from reaching their email boxes.

2. The language of the labeling provision is internally conflicting. The Commission is not required to perpetuate this irrationality in regulations, as it proposes to do in Section 316.1.

The labeling provisions of the CAN-SPAM Act are susceptible to a damned if you do, damned if you don't reading that is facially unconstitutional. Section 5(d)(1) of the CAN-SPAM Act says that it is unlawful to send a message that includes sexually oriented material and (A) fail to label it or (B) fail to provide that the message when initially opened does not include sexually oriented material but only instructions on how to access sexually oriented material. This means that if one sends a message with sexually oriented content and labels it, one still violates 5(d) because one has failed to send the message in a way that includes only instructions on how to access the sexually oriented material. In essence, the placement of the fail to language makes it illegal to fail to do (A) and illegal to fail to do (B). And (B) is a prohibition against including sexually oriented material in the initial message. The effect of 5(d)(1)(B) is to prohibit sending commercial email that includes anything more than instructions on how to access sexually oriented material, even if the material is lawful. Thus, 5(d)(1)(B) becomes not a labeling requirement, but a prohibition against including lawful sexually oriented material directly in a commercial email. This is clearly unconstitutional.

To compound the confusion, if one complies with 5(d)(1)(B), and doesn't actually include sexually oriented material but only instructions on how to access sexually oriented material, then the whole section doesn't apply. But (B) also says that even if one doesn't include sexually oriented material, (B)(iii), one must still label one's message as including sexually oriented material, (B)(i). This section, in its circularity, fails the constitutional requirement of clarity in legislation regulating speech.

Again, we appreciate that the Commission will not in this context opine on the constitutionality of an enacted law. However, the Commission does not have to repeat the internally inconsistent language of 5(d)(1) in its own rule. Section 5(d)(3) of the Act requires the Commission only to prescribe clearly identifiable marks or notices to be included in or associated with commercial electronic mail that contains sexually oriented material . . .[3] Thus, we urge the Commission to limit its actions with respect to the labeling requirement to the single function mandated by 5(d)(3) of the statute and to refrain from perpetuating in its regulations the provisions of 5(d)(1) that make no sense.

3. We support Commission efforts to monitor the effectiveness of this measure at reducing consumer exposure to unwanted sexually oriented email messages.

Question 11 of the NPRM asks how the Commission can measure the effectiveness of the proposed rule in protecting consumers from unwanted sexually oriented email messages[.] We support the Commission's apparent plans to monitor the effectiveness of the labeling requirements and believe that it is well positioned to do so. We believe that it will be important to wait until the law and the labeling requirement has been in effect for at least 6-9 months before conducting this kind of assessment. At that time it may be useful to request public comment on the effectiveness of the labeling provision. It may also be useful to monitor consumer complaint levels at the FTC. We caution against involving ISPs in efforts to monitor effectiveness, as doing so could have the effect of encouraging more aggressive filtering to demonstrate that the law is working. Such a response would have the negative effective of strengthening government influence on ISPs' decisions about filtering, thus exacerbating the unconstitutional aspects of the statute.

CDT appreciates the opportunity to submit these comments.

Respectfully submitted,

Paula J. Bruening
James X. Dempsey
Center for Democracy and Technology

February 17, 2004

Notes

[1] Opt-in Email Marketers Already Comply With CAN Spam Ð But Not All of It, Internet Retailer, January 27, 2004

[2] The CAN-SPAM Act defines sexually oriented material by reference to 18 USC 2256 which defines sexually explicit conduct as actual or simulated -- (A) sexual intercourse, including genital-genital, oral-genital, anal-genital, or oral-anal, whether between persons of the same or opposite sex; (B) bestiality; (C) masturbation; (D) sadistic or masochistic abuse; or (E) lascivious exhibition of the genitals or pubic area of any person.

[3] Section 5(d)(3) provides that [n]ot later than 120 days after the date of the enactment of this Act, the Commission in consultation with the Attorney General shall prescribe clearly identifiable marks or notices to be included in or associated with commercial electronic mail that contains sexually oriented material, in order to inform the recipient of that fact and to facilitate filtering of such electronic mail.

Free Speech | Data Privacy | Government Surveillance | Cryptography | Domain Names | International | Bandwidth | Security | Internet Standards, Technology and Policy Project | Terrorism | Authentication | Right to Know | Spam

Our Mission / Get Involved / Staff / Publications / Links / Search CDT / Jobs / Action! Previous Headlines | Legislative Tracking | CDT's Privacy Policy