Some Unsuccessful Approaches to Track Three

Both judicial opinions and the academic commentary thereon offer a variety of ways of dealing with Track Three laws. Unfortunately, all of them, in at­tempting to make distinctions within Track Three, end up looking ad hoc and juryrigged.

At the other extreme, Robert Kamenshine would bar the government from taking any partisan political position, and Stephen Arons would completely eliminate public schools.^[178] And in between the extremes of Schauer’s hands- off position and the KamenshineZArons position that would give full force to the Barnette principle of evaluative neutrality^[179] and invalidate most of Track Three, the commentators have staked out a variety of intermediate positions. Benzanson and Buss would distinguish among Track Three laws based on whether there is a danger of government’s monopolizing a speech marketplace or a danger that government subsidies will distort a message or otherwise de­ceive its audience.^[180] Mark Yudof cites fears of government indoctrination and domination of the marketplace of ideas, and he advocates a structural approach to handle such dangers, including decentralization and professionalization of Track Three actions.^[181] Howard Wasserman is concerned with government dom­ination of the marketplace of ideas, government’s drowning out of other speak­ers, government indoctrination, and government ventriloquism (disguisedly speaking through private speakers).^[182] Steven Shiffrin would “balance”;^[183] Leslie Jacobs would forbid partisanship in government’s aiding private speakers but not in government’s own speech;^[184] whereas Edward Ziegler would forbid only government’s backing or opposing particular candidates or initiatives.^[185] And Martin Redish and Kevin Finnerty are concerned with indoctrination in public education.^[186]

Although I cannot discuss in detail all of the distinctions that various com­mentators and judges have attempted to draw within Track Three, in what fol­lows I shall analyze what I believe are the three principal contenders for Track Three sorting devices: the concept of the public forum; the distinction between viewpoint discrimination and other forms of content and speaker discrimination; and the notion of message distortion.

A. Public Fora Versus Nonpublic Fora

I have already mentioned that the courts of the United States sometime distin­guish between public fora and nonpublic fora.^[187] In public fora, the constitu­tional rule is supposed to preclude government’s making distinctions based on the content of the messages being conveyed. In nonpublic fora, the government may favor certain speakers and messages.^[188] Whether a forum is “public” or “nonpublic,” however, turns on whether government intends to create a public forum or only a nonpublic one.^[189] Yet if governmental intent is the touchstone for determining what kind of forum is in issue, it is difficult to see why gov­ernmental partisanship in allowing access to the forum would not conclusively demonstrate an intent to create a nonpublic forum - that is, a forum for the benefit of specific speakers and messages. The result would then be that gov­ernmental partisanship would establish a nonpublic forum which would in turn legitimate the very partisanship that established it. In any event, public forum analysis has proven to be completely unilluminating - a way of characterizing conclusions about the legitimacy of Track Three laws rather than a basis for demonstrating the soundness of those conclusions.

B. Distinguishing Among Subject Matter, Viewpoint,

and Speaker Discrimination

One approach to Track Three laws that is sometimes put forward in the litera­ture is to distinguish between those content distinctions that turn on the subject matter of the message and those that turn on the particular viewpoint expressed by the message. (Track Three discriminations among speakers might similarly be analyzed in terms of whether government, by favoring some speakers over others, intended to advance or suppress certain subject matter or to advance or suppress certain viewpoints.) If government is discriminating against subject matter - that is, certain topics, such as religion, politics, or art - the discrimina­tion will ordinarily be treated as consistent with freedom of expression.

There are several problems with this approach to Track Three laws. First, and of least concern, the approach is inconsistent with the case law (though the case law is itself internally inconsistent). For example, the United States Supreme Court has struck down as violative of free speech Track Three laws that discriminated on the basis of subject matter. In Carey v. Brown^[190] and Police Department v. Mosely,^[191] for example, the Court struck down laws that gave picketers involved in labor disputes the right to picket at places where all other picketers were excluded. The laws did not base eligibility for the exemption on any particular viewpoint that the labor picketers were expressing. And in Southeastern Promotions, Ltd. v. Conrad^[192] the Court held unconstitutional a policy of a city-run theater to put on only family-suitable plays, again without regard to any specific message the play might convey. On the other hand, the Court approved value-inculcation in public schools;^[193] and indeed, the public schools teach specific viewpoints throughout the curriculum.

The second and much more serious problem with distinguishing subject matter discrimination from viewpoint discrimination is that the line between the two types of discrimination is an impossible one to draw and to defend. Is a ban on discussions of abortion in public schools a subject matter ban or a view­point ban? Were the exemptions for labor picketing in Carey and Mosley really viewpoint-neutral? Are “family-suitable” plays a viewpoint-neutral category? Subject matter and viewpoint bleed into one another.

Consider in this regard the question of whether the St. Paul ordinance at issue in R.A.V.

Justice John Paul Stevens, however, saw the ordinance as making only a subject matter discrimination, not a viewpoint one.

Contrary to the suggestion of the majority, the St. Paul ordinance does not regulate expression based on viewpoint. The St. Paul ordinance is evenhanded. In a battle between advocates of tolerance and advocates of intolerance, the ordinance does not prevent either side from hurling fighting words at the other on the basis of their conflicting ideas, but it does bar both sides from hurling such words on the basis of the target’s “race, color, creed, religion or gender.” To extend the Court’s pugilistic metaphor, the St. Paul ordinance simply bans punches “below the belt” - by either party. It does not, therefore, favor one side of any debate.^{[196] [197]}

The same battle over characterizing a content discrimination as subject­matter-based or as viewpoint-based arose in Rosenberger v. Rector and Visitors of University of Virginia,⁵⁴ Lamb’s Chapel v. Moriches Union Free School District,^[198] and Good News Club v. Milford Central School.^[199]

In Rosenberger, the question was whether a University of Virginia policy authorizing payment of the printing costs of a variety of student publications but prohibiting payment for any student publication manifesting “a particular belief in or about a deity or an ultimate reality” was viewpoint discrimination, as the excluded student group claimed, or was merely a subject matter restriction (excluding religion), as the University of Virginia claimed.

In Lamb’s Chapel, the Court found a school district rule prohibiting the after­school use of school property for religious purposes to constitute viewpoint dis­crimination when applied to exclude a group wishing to show religious-oriented fllms on family and childrearing matters. The Court rejected the lower court’s holding that the district’s rule amounted only to subject matter discrimination, reasoning that the rule permitted all views on family issues except those reflect­ing religious perspectives.⁵⁶ And in GoodNews Club, the Court again found a similar school district rule to be unconstitutional viewpoint discrimination as applied to a Christian group engaged in instructional and devotional activities. The dissent characterized the group’s activities as proselytizing and worship, not merely the teaching of morals from a religious perspective, and argued that schools should be able to exclude such activities in a nondiscriminatory man- ner.⁵⁷ Justice Scaliarespondedbypointing out that because political, social, and cultural organizations were free to inculcate beliefs and proselytize, prohibiting religious groups from doing so constituted viewpoint discrimination.⁵⁸

Despite the difficulties in distinguishing subject-matter from viewpoint dis­crimination within the broader category of content discrimination, some com­mentators remain attracted to this distinction as a way of negotiating the Track Three subsidy-of-private-speakers domain. Martin Redish and Daryl Kessler are notable examples.⁵⁹ Redish and Kessler label subsidies to private speakers and expenditures of government funds to purchase library books, put on plays, and so forth as “auxiliary subsidies.”⁶⁰ They then divide the realm of auxiliary subsidies into “categorical” and “viewpoint” subsidies and subsidies of “judg­mental necessity.”⁶¹ Categorical subsidies are “viewpoint neutral choices to fund particular categories, subjects, or classes of speech.”⁶² Redish and Kessler give as examples the funding of a study on the effects of smoking and the purchase of history books or books of American fiction by a public library.^[200]

54 Supra note 19, at 831.

55 Id. at 895.

56 Supra note 52, at 394.

57 Supra note 53, at 2114 (Stevens, J., dissenting).

58 Id. at 2110.

59 Martin H. Redish and Daryl I. Kessler, “Government Subsidies and Free Expression,” 80 Minn.

L. Rev. 543 (1996).

60 Id. at 567.

61 Id.

62 _Id.

Categorical subsidies, which Redish and Kessler would generally permit, are to be distinguished from subsidies of speakers based on their viewpoint, which Redish and Kessler would generally forbid. They reason that subsidies of cat­egories of speech, though content-based, are unlike content-based restrictions on speech in that the former increase the total amount of speech. And they are unlike viewpoint-based subsidies in that they present little danger of skewing public debate; they do not provide one side with more resources, nor do they disguise from the audience the government’s role in inducing the speech.^[201]

On the danger of masking impermissible viewpoint discrimination through categorical subsidies, Redish and Kessler write:

One must concede that government may on occasion seek to disguise what are in re­ality viewpoint-based subsidies behind the mask of permissible categorical subsidies. This problem has long plagued the doctrinal distinction between viewpoint-based and content-neutral regulations of expression. Although this danger is impossible to avoid completely, at least two tactics can mitigate the likelihood of its arising. Initially, courts must prohibit the government from defining categorical subsidies in a viewpoint laden manner, such that the very contours of the category effectively exclude viewpoints with which government disagrees. For example, government cannot define the category as “the evils of abortion,” thereby effectively excluding any expression that advocates free­dom of choice. A reviewing court generally should be in a position to resolve this issue on the four corners of the governmentally established category, without the need to resort to a separate factual inquiry.

In certain situations, however, courts must still undertake a separate factual inquiry. Such a case would arise when, although on its face the categorical description unambigu­ously excludes viewpoint distinctions, an unsuccessful applicant for the subsidy asserts that in reality the government based its denial on the applicant’s underlying viewpoint.^[202]

The final type of speech subsidy that Redish and Kessler discuss is the sub­sidy of “judgmental necessity.” Such a subsidy occurs when the government subsidizes a particular category of speech but, because of finite funds, must discriminate among speakers within the category. Thus, if the National Endow­ment for the Arts funds “modern art,” it still must select for grants only some of the modern artists who apply. Or if the National Endowment for the Humanities funds historical research on the Revolutionary War, it still must choose among competing researchers based on some criteria. Those criteria for selection will undoubtedly include evaluative criteria that reflect one viewpoint among sev­eral. Redish and Kessler would permit use of such evaluative criteria to choose within a viewpoint-neutral category as long as the criteria are “substantially related” to the government’s purposes in funding the general category. Thus, government could choose for funding the “best” modern artists or American historians, but it could not choose modern artists or American historians based on their politics.^[203] Likewise, in selecting American history texts for use in the public schools, school boards can pick the texts based on accuracy; and that means that if the board believes Christopher Columbus was a racist, it may select history books based on whether they portray Columbus as such.^[204]

In deciding to purchase accurate textbooks in the first instance, the government makes a viewpoint-neutral choice to fund a particular study or class of speech, a constitutionally valid categorical subsidy. Within this category of speech, a school board must of course select one textbook for use in its schools. Under our approach, the board’s choice falls within the “judgmental necessity” exception, as long as it makes that choice on the basis of criteria “substantially related” to the predescribed purpose of the program pursuant to which textbooks are funded. Thejob of the school board, presumably, is to select the textbook that will provide students with the most complete and accurate description and understanding of history. If a school board believes that Christopher Columbus was a racist, then it is likely that, all other things being equal, this school board will choose a textbook that describes Columbus as a racist over one that ignores his treatment of Native Americans. Insofar as subjective viewpoint is inextricably linked to the board’s judgment of textbook quality, it is difficult to argue that a school board cannot make the decision based on the textbooks’ treatment of Columbus.^[205]

Despite their heroic efforts to justify the distinction between subject matter and viewpoint discrimination, I do not believe Redish and Kessler have suc­ceeded, nor do I think success is possible. Subject matter and viewpoint do not define natural kinds. There is no principle relevant to freedom of expression as a human right that tracks such a distinction among Track Three laws. Not only will any postulated boundary between subject matter and viewpoint discrimi­nation be contestable - R.A.V., Rosenberger, Lamb’s Chapel, and Good News Club surely illustrate that - but it will be unprincipled as well. All Track Three laws involve content discrimination, and all content discrimination violates the Barnette principle of evaluative neutrality. And that is as true of content dis­crimination based on subject matter as it is of content discrimination based on viewpoint.

Consider: If the National Endowment for the Arts subsidizes “modern art,” is modern art a category, or is it a viewpoint? The government is surely expressing views: that modern art is “art”; that it is as or more meritorious in terms of art’s functions than other categories of art; that art should be funded; and so forth. And not only is each of these an expression of a viewpoint, but each of these is controversial. The opposing viewpoints - that modern art is not “art,” that it is decadent or otherwise nonmeritorious, or that art should not be publicly funded - are viewpoints that the government is discriminating against in its subsidies. And the same is true whenever the National Endowment for the Humanities decides, contrary to Henry Ford, that history is not bunk but is instead worthy of public subsidy.

Moreover, because funds for subsidies are always scarce, government will have to make what Redish and Kessler call subsidies of judgmental necessity within the subject matter categories, which essentially involves drawing the subject matter categories more narrowly. If the National Endowment for the Humanities cannot fund all historical research, it must limit its funding to “good” or “the best” historical research, which it might define as research demonstrating “the oppressive nature of Eurocentric thought” or “the anti­Native racism of Columbus.” Redish and Kessler admit the evaluative nature of such judgments and concede their necessity. What is true, however, is that the lines defining the categories themselves perforce exhibit the same evaluative judgments, including quite contestable ones.

The same problem arises when the government subsidizes not subject mat­ter, but speakers. For example, in Arkansas Education Television Commission v. Forbes,^[206] the Supreme Court dealt with a candidates’ debate on publicly- funded television that excluded minor candidates, defined as those having little support in the polls. The Court upheld the exclusion because it was not based on the speaker’s viewpoint.^[207] Surely, however, restricting public funding to candi­dates far ahead in the polls expresses a message, and a quite contestable one, about whose views it is important for the public to hear. (Arguably, the very public exposure that third-party candidates seek through public subsidies may be both necessary and sufficient to generate the showing in the polls required to get those public subsidies. If so, an equally plausible case could be made for discriminating against, not in favor of, the major party candidates.) All of which is not to deny that there are good reasons to restrict candidates’ debates to those ex ante likely to win the elections. It is only to deny that those reasons are somehow neutral or nonevaluative and do not express a viewpoint.

C. Guarding Against the Distortion of Views

Another approach to Track Three that is sometimes put forward is built on a concern that the audience not be deceived regarding the authorship of speech. Thus, if a doctor conveys family planning information to a patient but, because of conditions attached to the governmental funding of the clinic, fails to mention the availability or desirability of abortion, the patient may interpret the infor­mation differently depending upon whether she is aware of the government’s role in the doctor’s speech. If she is unaware of that role, she may conclude that what the doctor conveys about family planning is all that the doctor knows or endorses, which may not be the case. Thus, governmental funding may distort the message that the audience receives from the speaker.

Another type of message distortion that one might worry about in connection with Track Three laws is that caused by inducing speakers to tailor their mes­sages in order to receive governmental grants and privileges. The problem here is not so much speaker insincerity but rather that speakers will have their beliefs subtly altered by the allure of receiving funding. In other words, speakers will be speaking sincerely but at the same time will be exhibiting bad faith or false consciousness.

Commentators and justices who worry about these effects of Track Three laws usually distinguish between those cases where government itself is clearly the speaker and those in which the government is funding or exempting from regulation speakers with favored messages, finding much more danger of mes­sage distortion in the latter than in the former. But approaching Track Three laws from the vantage point of a worry about message distortion is, I believe, unpromising.

First, because governments “speak” only through agents, agents who are required to follow a government script to hold public employment will be induced, even if only subtly, to alter their beliefs to fit their scripts. A biology teacher who rejects Darwinism in favor of Creationism will likely encounter cognitive dissonance if he is forced to teach only Darwinism to his students and to mark Creationist answers as incorrect. So any Track Three law can represent an inducement to speakers to alter their beliefs. The problem is not confined to government subsidies of private speakers.

Second, even where there is transparency regarding government’s role, and thus no danger of message distortion, disquiet with Track Three laws will likely remain. Consider a public school curriculum that teaches the superiority of the Republican Party. Because it is a public school, government’s role in shaping the message is obvious to the audience. And it is the government speaking, not the government subsidizing private speakers. Nonetheless, those who worry about Track Three laws tend to put this example in the “clear violation of freedom of expression” column.

I doubt, therefore, that we will get much mileage out of concern for how Track Three laws might distort messages, either from the audience’s standpoint or from the speaker’s. Message distortion sounds like a bad thing, and perhaps it is. Perhaps it violates a right not to be deceived, a right that is independent of the right of freedom of expression.^[208] But message distortion is a concern that has either too much or too little traction within Track Three.

IV.