Urofsky v. Allen, 995 F. Supp. 634 (E.D. Va. 1998).
On July 1, 1996, Virginia enacted a statute that restricted state employees' access to constitutionally protected speech.(1) Under the new law, Virginia employees could not use state computer equipment to access sexually explicit material without express authorization from the head of a state agency.(2) In Urofsky v. Allen,(3) the United States District Court for the Eastern District of Virginia struck down the statute as an unconstitutional abridgment of First Amendment rights.
On its face, Urofsky appears to be an important victory for those who advocate liberal constructions of the freedom of speech. After all, the court refused to be swayed by Virginia's argument that "state employee computer use is not protected speech under the First Amendment because the employees are acting in their capacities as government employees, not public citizens."(4) It also rejected the argument that expressions of sexuality, while protected under the First Amendment, "`are of such slight social value ... that any benefit that may be derived from them is clearly outweighed by the social interest in order and morality.'"(5) Instead, the court accorded full constitutional protection to sexually explicit materials and condemned the statute as an impermissible restriction on the ability of public employees to "speak on matters of public concern."(6)
But the statute did not restrict this ability, at least not directly. What it did restrict directly was the employees' ability to hear others speak on matters of public concern. Unfortunately, the court failed to distinguish between the First Amendment interest in producing speech and the First Amendment interest in receiving speech, and as a result, its reasoning is strained and ultimately unpersuasive. Furthermore, by failing to identify properly the interest most directly at stake in the case before it, the court lost the opportunity to protect this interest against future incursions that are sure to come.
The facts presented in Urofsky were in little dispute. The Virginia statute, entitled "Restrictions on State Employee Access to Information Infrastructure," made it a criminal offense for any state employee to use state-owned or state-leased computer equipment "to access, download, print or store any information infrastructure files or services having sexually explicit content" without first obtaining written approval.(7) Under the statute, such approval could be granted only by the heads of state agencies and only "to the extent required in conjunction with a bona fide, agency-approved research project or other agency-approved undertaking."(8) Furthermore, the statute defined "sexually explicit content" to include any description or visual representation of "a lewd exhibition of nudity, ... sexual excitement, [or] sexual conduct," and it defined "information infrastructure" so broadly as to encompass all "telecommunications, cable, and computer networks," including "the Internet, the World Wide Web, Usenet, bulletin board systems, on-line systems, and telephone networks."(9) So if Virginia employees wished to view nudity on the Internet, receive sexual messages over e-mail, or even discuss sexuality on office telephones,(10) they had to justify their actions in terms of official state business and obtain the permission of their employer.
The Urofsky plaintiffs, a group of professors at Virginia state colleges and universities, challenged the constitutionality of these restrictions. They claimed that because the statute restricted their access to sexual materials, it interfered with their research and teaching and thereby violated their First Amendment rights.(11) In response, the state argued that the statute's restrictions were necessary to maintain efficiency in the public workplace and to prevent the creation of sexually hostile work environments.(12)
To resolve this controversy, the district court analyzed the Virginia statute as a direct restriction on state employees' freedom of speech. Using a test that the Supreme Court established in Pickering v. Board of Education,(13) the court balanced the "`interests of the [employee] ... in commenting upon matters of public concern'" against "`the interest of the State, as employer, in promoting the efficiency of the public services it performs.'"(14) The court also considered "the public's interest in receiving the speech of government employees."(15) While the court recognized the validity of the government's interests in maintaining workplace efficiency and preventing sexually hostile work environments, it found the statute "both fatally overinclusive and underinclusive" as a method of furthering those interests.(16) Moreover, the court noted that Virginia employees were already subject to content-neutral policies and statutes--that is, those that did not single out expressive material on the basis of its content--that addressed the state's concerns.(17) In the court's view, the "obvious lack of fit between the government's purported interest[s] and the sweep of its restrictions,"(18) combined with the existence of these content-neutral alternatives, cast "serious doubt" on the asserted need for the statute,(19) The court therefore determined that the asserted governmental interests did not outweigh the employees' interest in speaking and the public's interest in hearing their speech. Accordingly, it granted summary judgment to the plaintiffs and invalidated the statute in its entirety.
Urofsky builds upon an extensive First Amendment doctrine relating to speech in the government employment context. In a series of cases, the Supreme Court has explored the extent to which the government may restrict its employees' ability to engage in political activity,(20) criticize government policy,(21) practice political patronage,(22) and affiliate with "subversive" political organizations.(23) It is from these cases that the Urofsky court derived its standard for reviewing the Virginia statute. The court relied heavily on Pickering, in which the Supreme Court granted relief to a teacher who had been fired for publicly criticizing the Board of Education that employed him.(24) The Pickering Court balanced the teacher's First Amendment interest in commenting upon matters of public concern against the state's interest in furthering the efficiency of its services through employee management.(25) Subsequently, in United States v. National Treasury Employees Union(26) (NTEU), which partially invalidated a statute that prohibited government employees from accepting compensation for speeches or articles, the Court broadened the Pickering test to account for the fact that restrictions on government employee expression impinge upon the public's interest in hearing this expression.(27) Drawing upon these cases, the Urofsky court implemented what it referred to as "the Pickering/NTEU test": "[W]hen the government broadly restricts public employee speech, it has the burden of establishing that `the interests of both potential audiences and a vast group of present and future employees in a broad range of present and future expression are outweighed by that expression's necessary impact on the actual operation of the Government.'"(28)
But Urofsky differs from Pickering, NTEU, and other previous government employment cases in a fundamental respect. Each of the precedents involved a direct attempt by the government to prevent its employees from engaging in a certain form of speech; in all of them, the government penalized employees for speech that they themselves had produced. In such situations, application of the Pickering/NTEU standard makes sense. However, Urofsky did not involve a direct attempt by the government to suppress its employees' speech. Rather, the censorship at issue in Urofsky produced this suppressive effect only indirectly (i.e., by denying employees access to materials about which they might speak).(29)
Rather than miscasting the Virginia statute solely as a threat to government employees' right to produce speech, the Urofsky court should have examined it primarily as a threat to their right to receive speech. No court has ever explicitly addressed this latter interest in a government-employment free speech case, a fact that may partially explain the Urofsky court's oversight. To some extent, this has resulted from the nature of the government restrictions that have heretofore arisen in the employment context. However, there have been some cases that have clearly implicated public employees' interests as listeners, most notably those involving restrictions on the provision of information to military personnel.(30) Interestingly, courts have chosen to analyze such restrictions under the public forum doctrine, and the only First Amendment interests discussed in these cases are those of the parties seeking to provide the information. Like Urofsky, these cases have focused solely on government employees' right to produce speech and ignored their right to receive it.
The right to receive speech, while constitutionally derivative of the right to produce it,(31) is distinct and possesses independent legal force. As the Supreme Court has stated, "Freedom of speech presupposes a willing speaker. But where a speaker exists, ... the protection afforded is to the communication, to its source and to its recipients both."(32) Although it has long been established that the fight to receive speech possesses independent constitutional status,(33) this fight has only recently begun to assume a central role in the Court's First Amendment jurisprudence. The emergence--or more accurately, the reemergence--of this fight has been particularly striking in the television regulation context, where the Court has repeatedly emphasized the importance of providing "the widest possible dissemination of information from diverse and antagonistic sources."(34) While the Justices have yet to discuss this right in the government employment context, it would be inconsistent for it not to exist there as well.(35)
Had the Urofsky court followed the Supreme Court's lead and framed the Virginia statute as a restriction on public employees' fight to receive speech, its analysis would have taken on a fundamentally different character. To be sure, the court still could have employed a balancing approach,(36) but the nature of the First Amendment interests considered, and the methodology for evaluating burdens on those interests, would have been substantially different. To evaluate the statute's First Amendment implications from this perspective, a court would examine the extent to which the statute burdened the state employees' right to access the speech of others. In making this examination, a court would need to inquire into such factors as the degree to which state employees use the Internet for informational purposes and the availability of alternative information sources. Only through such inquiries could a court determine the burden the statute placed on employees' First Amendment access rights. It could then go on to explore the extent to which this burden in turn restricted other constitutional rights, including those recognized in Urofsky. For the point is not that the statute's impact on employee speech interests is irrelevant, but that it is secondary. Just as the public's interest in heating speech is only implicated to the extent that the statute restricts the employees' ability to speak, the employees' ability to speak is only implicated to the extent that the statute restricts their access to information and ideas.
Framing the Virginia statute properly--as a restriction on the right to receive speech--does not alter Urofsky's bottom-line result. The Internet, as the court noted, is "arguably the most powerful tool for sharing information ever developed,"(37) and the record in Urofsky made clear that state employees regularly access the Internet's "sexually explicit" content over state computer systems for a variety of professional, personal, and pedagogical reasons.(38) In addition, the statute prohibited employees not only from accessing sexual material on the Internet, but also from receiving it through e-mail and local database networks.(39) Undoubtedly, many state employees depend upon their employment for access to all of these information technologies. The Virginia statute therefore substantially impaired their ability to access vast quantities of information that they could not easily obtain from alternative sources. Neither of the governmental interests advanced in support of the statute can justify such a severe contravention of the employees' First Amendment rights. While furthering operational efficiency and preventing the development of hostile workplaces are valid goals, the statute is, as the court correctly observed, "both fatally overinclusive and underinclusive" as a means of attaining them.(40) Thus, the court was right to strike the statute down as a violation of the First Amendment; it merely erred in its methodology.
The short-term cost of this error appears minimal. Although the court's analysis was disjointed--an uneasy amalgam of precedent, reasoning, and facts--its ultimate decision to invalidate the statute appears correct. But the long-term cost could well be much greater. In disregarding the primary constitutional right at stake, the court failed to produce a principled basis for defending that right against the Virginia statute's threat and against the future threats that almost certainly will arise.
The right to receive speech is likely to be increasingly threatened as a result of several emerging sociopolitical trends. First, information technologies are becoming the primary vehicles of public discourse, and the citizen's ability to access them is therefore assuming a greater constitutional dimension.(41) Furthermore, as these technologies have replaced more individualized methods of communication, the information sources upon which citizens rely have become increasingly concentrated. This, in turn, has increased the likelihood that government regulations will impair public access to information; the more concentrated the sources of information become, the more likely it is that the regulation of an individual source will significantly impair the citizen's ability to receive speech and ideas. At the same time that information technologies are assuming increasing constitutional significance and consolidating the flow of information to the public, both state and federal governments are becoming increasingly active in their efforts to regulate these technologies.(42) Thus, it seems inevitable that future governmental actions will intrude upon the right to receive speech. By failing to recognize the importance of this right, the Urofsky court left it vulnerable to such future intrusions.
Because the court failed to appreciate fully the novelty of the issue before it, what could have been a watershed case may well end up as a minor footnote in the history of First Amendment jurisprudence. The Urofsky court could have been the first to state explicitly that the fight to receive speech has reemerged as a central component of constitutional law. It could have been the first to examine this right as it applies to public employees. And it could have been the first to establish the importance of this right in the Internet context. But it was not. Instead, the court's decision, while correct in the narrow sense, failed to address the central issue that underlay the case.
(1.) See VA. CODE ANN. [subsections] 2.1-804 to -806 (Michie Supp. 1998).
(2.) See id. [sections] 2.1-805.
(3.) 995 F. Supp. 634 (E.D. Va. 1998).
(4.) Id. at 636 (quoting the defendant's trial memorandum). This argument was not without precedential support. As Justice Holmes once stated, "There are few employments for hire in which the servant does not agree to suspend his constitutional fights of free speech ... by the implied terms of his contract," McAuliffe v. Mayor of New Bedford, 29 N.E. 517, 517-18 (Mass. 1892), and recent Fourth Circuit decisions had adhered to this principle in allowing restrictions on state employees' speech, see Boring v. Buncombe County Bd. of Educ., 136 F.3d 364 (4th Cir. 1998); DiMeglio v. Haines, 45 F.3d 790 (4th Cir. 1995). But see, e.g., Perry v. Sindermann, 408 U.S. 593, 597 (1972) (arguing that the government may not deny a person benefits "on a basis that infringes his constitutionally protected interests--especially, his interest in freedom of speech").
(5.) FCC v. Pacifica Found., 438 U.S. 726, 746 (1978) (quoting Chaplinsky v. New Hampshire, 315 U.S. 568, 572 (1942)).
(6.) Urofsky, 995 F. Supp. at 636.
(7.) VA. CODE ANN. [sections] 2.1-805 (Michie Supp. 1998). Only employees of the Department of State Police were exempted from this prohibition. See id. [sections] 2.1-804. Thus, state agencies prohibited from using or maintaining databases containing sexually explicit information included Virginia's Departments of Corrections, Social Services, Juvenile Justice, and Mental Health, as well as the Office of the Attorney General and the Library of Virginia. See Urofsky, 995 F. Supp. at 638, 642.
(8.) VA. CODE ANN. [sections] 2.1-805.
(9.) Id. [sections] 2.1-804.
(10.) The statute apparently covered communications on standard office telephones that were part of a computerized network. See Urofsky, 995 F. Supp. at 635 n. 1.
(11.) See id. at 635.
(12.) See id. at 639.
(13.) 391 U.S. 563 (1968).
(14.) Urofsky, 995 F. Supp. at 636 (quoting Pickering, 391 U.S. at 568).
(15.) Id. at 637.
(16.) Id. at 639. As a method of furthering workplace efficiency, the court found the statute overinclusive because it interfered with "countless work-related endeavors ... dealing with sexuality" and underinclusive because it targeted only sexual distractions. Id. at 640. As a method of preventing sexually hostile work environments, the court found the statute overinclusive because it restricted speech on matters such as sadomasochistic abuse and human rights violations and underinclusive because, inter alia, it targeted only electronic communications. See id.
(17.) See id. at 643. For example, state policy already prohibited Virginia employees from making "[i]nappropriate or unauthorized" use of state computers, and numerous content-neutral federal laws penalize activities that may create or contribute to a hostile work environment. See id.
(18.) Id. at 639 (quotations and citations omitted).
(19.) Id. (quotations and citations omitted).
(20.) See, e.g., United States v. National Treasury Employees Union, 513 U.S. 454 (1995); United States Civil Serv. Comm'n v. National Ass'n of Letter Carriers, 413 U.S. 548 (1973).
(21.) See, e.g., Rankin v. McPherson, 483 U.S. 378 (1987); Picketing v. Board of Educ., 391 U.S. 563 (1968).
(22.) See, e.g., Rutan v. Republican Party, 497 U.S. 62 (1990); Branti v. Finkel, 445 U.S. 507 (1980); Elrod v. Bums, 427 U.S. 347 (1976).
(23.) See, e.g., United States v. Robel, 389 U.S. 258 (1967); Elfbrandt v. Russell, 384 U.S. 11 (1966); Garner v. Board of Pub. Works, 341 U.S. 716 (1951).
(24.) See Pickering, 391 U.S. at 563.
(25.) See id. at 568. In Rankin, the Court used this same balancing test to analyze the claim of a county employee who had been fired for an on-the-job comment. In so doing, the Court clarified that this approach was proper regardless of whether the speech at issue occurred within or outside of the workplace. See 483 U.S. at 383.
(26.) 513 U.S. 454 (1995).
(27.) See id. at 468.
(28.) Urofsky, 995 F. Supp. at 637 (quoting NTEU, 513 U.S. at 468).
(29.) Indeed, proper analysis of the Virginia statute as a restriction of expressive activities would require application of the O'Brien standard, which the Supreme Court has developed for evaluating nonspeech restrictions that incidentally affect free speech. See United States v. O'Brien, 391 U.S. 367 (1968). The Urofsky court may have avoided this form of analysis because it realized, correctly, that the First Amendment issues at stake could not be captured fully within it.
(30.) See, e.g., Greer v. Spock, 424 U.S. 828, 832 (1976) (upholding the military's refusal to permit a presidential candidate to enter an army base to hold a meeting "to discuss election issues with service personnel and their dependents"); General Media Communications v. Cohen, 131 F.3d 273 (2d Cir. 1997), cert. denied, 118 S. Ct. 2367 (1998) (upholding a statute that prohibited the sale or rental of sexually explicit material on military property).
(31.) There are several paths to the conclusion that "freedom of speech necessarily protects the right to receive." Virginia State Bd. of Pharmacy v. Virginia Citizens Consumer Council, Inc., 425 U.S. 748,757 (4976) (quotations omitted). For instance, the receiver's right can be predicated upon the free speech right of the speaker; the right to deliver a message lacks substance if the government may prevent it from being heard. Cf. Martin v. City of Struthers, 319 U.S. 141, 143 (1943) ("This freedom embraces the right to distribute literature, and necessarily protects the right to receive it."). Alternatively, the receiver's right can be predicated upon his or her own right of free speech; one cannot participate in "the uninhibited, robust, and wide-open debate and discussion that are contemplated by the First Amendment" if one cannot access information to debate and discuss. Lamont v. Postmaster Gen., 381 U.S. 301,307 (1965) (citation omitted).
(32.) Virginia State Bd. of Pharmacy, 425 U.S. at 756.
(33.) See, e.g., Kleindeinst v. Mandel, 408 U.S. 753,762 (1972) ("`It is now well established that the Constitution protects the right to receive information and ideas.'") (quoting Stanley v. Georgia, 394 U.S. 557, 564 (1969)).
(34.) Turner Broad. Sys. v. FCC, 117 S. Ct. 1174, 1204 (1997) [Turner II] (Breyer, J., concurring) (quoting Turner Broad. Sys. v. FCC, 512 U.S. 622, 663 (1994) [Turner I]); accord Arkansas Educ. Television Comm'n v. Forbes, 118 S. Ct. 1633, 1640 (1998); Denver Area Educ. Telecomms. Consortium v. FCC, 518 U.S. 727,753-60 (1996); Red Lion Broad. Co. v. FCC, 395 U.S. 367, 390 (1969).
(35.) This is not to say that restrictions on public employees' right to hear should be scrutinized in the same manner as analogous restrictions imposed on ordinary citizens. Quite the contrary: The claim that the government as employer may impose more restrictions than the government as sovereign, see, e.g., Pickering v. Board of Educ., 391 U.S. 563, 568 (1968), retains its validity in the right-to-hear context. Thus, while strict scrutiny would be appropriate if the government imposed content-based restrictions on the citizenry's ability to access information, see Turner II, 117 S. Ct. at 1208 (O'Connor, J., dissenting), a more flexible approach is required in government employment cases, see infra note 36 and accompanying text.
(36.) There was some argument in Urofsky as to whether the balancing test was indeed appropriate, given the content-discriminatory nature of the Virginia statute. In most circumstances, content-based restrictions must withstand strict scrutiny in order to survive. See, e.g., Simon & Schuster, Inc. v. Members of the N.Y. State Crime Victims Bd., 502 U.S. 105, 116-18 (1991). However, the Urofsky court decided to apply a less stringent standard of review because it believed that under Board of County Comm'rs v. Umbehr, 518 U.S. 668 (1996), "the government may take actions as an employer that are forbidden to it as a sovereign." Urofsky, 995 F. Supp. at 638 (citing Umbehr, 518 U.S. at 678). But as the court itself realized, "the application of Umbehr to the [Virginia statute] is limited" because Umbehr involved an adverse action against an individual speaker, not "a content-based prior restraint affecting thousands of government employees." Id. It should also be noted that, whereas Umbehr involved the government in its role as a contractor, Urofsky involved the government in its role as an educator, and content-based speech restrictions are particularly pernicious in the education context. See Tinker v. Des Moines Indep. Community Sch. Dist., 393 U.S. 503, 506, 514 (1969); Shelton v. Tucker, 364 U.S. 479, 487 (1960). Thus, there is at least a plausible argument that the correct standard for analyzing the Virginia statute is strict scrutiny.
(37.) Urofsky, 995 F. Supp. at 638 (citing Reno v. ACLU, 117 S. Ct. 2329, 2335 (1997)).
(38.) See Urofsky, 995 F. Supp. at 639-42.
(39.) See supra notes 9-10 and accompanying text.
(40.) Urofsky, 995 F. Supp. at 639.
(41.) See, e.g., Andre R. Barry, Balancing Away the Freedom of Speech: Turner Broadcasting System v. FCC, 21 HARV. J.L. & PUB. POL'Y 272 (1997) (objecting to this doctrinal trend); Cass R. Sunstein, The First Amendment in Cyberspace, 104 YALE L.J. 1757, 1798-99 (1995); Owen M. Fiss, The Censorship of Television (Nov. 3, 1998) (unpublished manuscript, on file with author). Whereas the right to receive was originally explored as it applied to such traditional forms of speech as literature distribution, see Martin v. City of Struthers, 319 U.S. 141, 143 (1943); Lovell v. City of Griffin, 303 U.S. 444, 452 (1938), and interpersonal communication, see Thomas v. Collins, 323 U.S. 516 (1945), these forms of speech have in many respects been superseded by television, radio, and-computer technology as methods of information exchange, cf. Owen M. Fiss, Free Speech and Social Structure, 71 IOWA L. REV. 1405, 1408-13 (1986) (making a similar observation); Sunstein, supra, at 1792-94 (same). Arguably, the greater the role a technology plays in the public discourse, the more its regulation implicates First Amendment concerns.
(42.) The Virginia statute is just one example of this recent trend. Other examples include the Child Online Protection Act, Pub. L. No. 105-775 (Oct. 22, 1998), the Telecommunications Act of 1996, 110 Stat. 56 (to be codified in scattered sections of 47 U.S.C.), the Cable Television Consumer Protection and Competition Act of 1992, 106 Stat. 1460 (codified in scattered sections of 47 U.S.C.), and the Television Program Improvement Act of 1990, 47 U.S.C. [sections] 303c (1994).…