Some Thoughts on Free Speech, Language, and the Rule of Law

Some Thoughts on Free Speech, Language, and the Rule of Law

by Thomas Streeter

(from Robert Jensen and David S. Allen (eds.), Freeing the First Amendment: Critical Perspectives on Freedom of Expression, New York University Press, 1995, pp. 31-53.)

I. Introduction

This chapter discusses the relevance of research and reflection on language to recent critical trends in thinking on free speech. There is a tendency to interpret many of the recent revisionist approaches to free speech as if they were simply calls for exceptions to otherwise clear cut rules and principles, as if, say, pornography or racism are so exceptionally evil that they fall outside the parameters of the kinds of speech that are "obviously" protected under the First Amendment. This misses the fact that the new approaches, with varying degrees of explicitness, involve theoretical and epistemological challenges to the underlying premises of free speech law in general; over the long run, what the new approaches are calling for are not exceptions but a restructuring of free speech law as a whole. The ideas driving this profound rethinking come from a variety of traditions, including various currents of feminism, literary theory, and theories of race and ethnicity. This chapter focuses on just one of those traditions: the complex twentieth century theorizing of language, sometimes called the "linguistic turn" in twentieth century philosophy. Although the linguistic turn is only one aspect of the new thinking about free speech, and although its importance and character is not agreed upon by all those advocating the new thinking, calling attention to it is useful because it nicely highlights some conceptual difficulties of the traditional framework and because it helps differentiate the revisionist criticisms from social determinist and other subtly authoritarian criticisms of free speech.

On the one hand, this chapter argues that the linguistic turn involves some revelations about the nature of language and human communication that do not accord well with the understandings of language implicit in free speech law, particularly with the metaphor of the marketplace of ideas. On the other, it argues that part of what is at stake is the way American culture envisions the rule of law as a whole. In particular, important currents of the understanding of the rule of law suggest the possibility and necessity of constructing rules, procedures, and meanings that transcend or can be abstracted from context, whereas the linguistic turn suggests that this is impossible, that meanings can be determined only in relation to particular contexts. The final part of this chapter, therefore, suggests some avenues for exploring free speech in its historical and social context, as opposed to efforts to abstract it out of context.

II. The Power of Linguistic Style

In the course of a discussion of the campus hate speech controversy, literary critic Henry Louis Gates (speaking from an African American position) provided the following hypothetical examples of potentially "harmful" speech directed at a minority student:

(A) LeVon, if you find yourself struggling in your classes here, you should realize it isn't your fault. It's simply that you're the beneficiary of a disruptive policy of affirmative action that places underqualified, underprepared and often undertalented black students in demanding educational environments like this one. The policy's egalitarian aims may be well-intentioned, but given the fact that aptitude tests place African Americans almost a full standard deviation below the mean, even controlling for socioeconomic disparities, they are also profoundly misguided. The truth is, you probably don't belong here, and your college experience will be a long downhill slide. (B) Out of my face, jungle bunny.[1]

Gates' point was that any reader can tell that the first statement would be much more alienating or "wounding" to a minority student, yet the typical campus hate speech code would only prohibit the second statement. But these two statements raise some interesting questions that Gates did not directly address: why would the second statement seem more wounding, and why would our legal system be more amenable to penalizing the second?

Sociolinguistics offers an answer to the first question: the social phenomenon of linguistic style. It is not the contents of the first statement that give it force; the argument it makes is, at best, dubious and obfuscatory, whereas the second statement at least would communicate the true feelings of the speaker towards the hearer with considerable precision. The first statement's power comes from its style.

It is a well established fact that fluency in any language involves mastery, not just of a single, "correct" version of a language, but of a variety of styles or codes appropriate to specific contexts.[2] Gates' first example is a case of the formal or "elaborated" style of contemporary English, which is highly valued in academic and professional settings. It is characterized by, among other things, Latinate vocabulary ("demanding educational environments" instead of "tough schools") and elaborate syntax. The second is an example of informal or restricted style, characterized by ellipsis (omitting "You get out of my face . . . ") and colloquial constructions.

Linguists also have long insisted that, in an absolute sense, formal style is no more correct or better for communication than informal style. Scientifically speaking, what makes a style appropriate or inappropriate is the social context in which it is used: in an academic setting, the formal character of the first example gives the statement force, but in another context, say, a working class bar, it might only elicit laughter and derision whereas the second statement might have considerable impact. In the appropriate context, therefore, one can use informal style brilliantly and subtly, and conversely, it is quite possible to speak in a thoroughly formal style and yet be inept, offensive, or simply unclear.[3]

What style differences communicate, then, are not specific contents, but social relations between speakers and listeners, i.e., relations of power, hierarchy, solidarity, intimacy, and so forth. In particular, formal language suggests a relation of impersonal authority between speaker and listener, whereas informal language suggests a more intimate (though not necessarily friendly) relationship. You can petrify a child by interjecting into an otherwise informal conversation, "No you may not." The shift to formal style (no ellipsis, "may not" instead of "can't") shows that the speaker is not just making a request, but is asserting his or her powers of authority as an adult over the child listener.

Gates's first example would be more wounding to a minority student, therefore, because, by couching itself in a formal, academic style, it is rhetorically structured as the expression of "impersonal," rational, and thus institutionally sanctioned, sentiments. It thereby invokes the full force of the authority of the university against the student's efforts to succeed in it. Gates's second example, with its informal style, suggests that one individual, the speaker, harbors racist ill will towards the listener. The first example, by contrast, suggests that, not just one individual, but the entire institution of the university in all its impersonal, "rational" majesty, looks upon the student as unfit.

III. The Traditional Framework: Problems with the Marketplace Metaphor

So why is it easier to penalize the second kind of statement than the first, when it is the first that is potentially more damaging (which is not necessarily to suggest that we should penalize the first kind of statement)? Contemporary law in general is insensitive to matters of linguistic style. Hollywood action movies have made a cliche of lampooning the incongruity of reading the highly formal, legalistic Miranda clause during arrests, which are typically emotional encounters between working class cops and criminals, i.e., contexts where informal style would be appropriate.[4] In First Amendment jurisprudence, where language is not only the vehicle but the subject matter of the law, this insensitivity can lead to conceptual confusion. Linguistic style may be a fact of life, but traditional legal liberal ways of thinking about free speech, especially those encapsulated in the metaphor of the "marketplace of ideas," are strangely incapable of addressing it.

A. The Arbitrariness of Language and Legal Indeterminacy

The marketplace metaphor in free speech law involves imagining symbolic and linguistic phenomena as if they were analogous to market exchange, which implies a number of things about language. Most obviously, it implies that language is primarily an exchange, a transference of something (perhaps "information"), from one person to another. Hence, in linguistic exchanges what matters is the contents of the exchange, not the style or form in which it is "packaged," just as in real market exchanges it makes little difference if you pay by check or cash. Yet, as in Gates' example, in language the "package" can be everything. The marketplace metaphor, then, draws our attention away from the importance of just the kind of stylistic differences that sociolinguists say are central to the workings of everyday language.

The marketplace metaphor, furthermore, tends to imply that the good that comes from unconstrained human speech comes from some neutral, universal, mechanical, and leveling process, a linguistic equivalent to the economist's invisible hand out of which will emerge truth, or at least some form of democratic justice. That neutral, mechanical process, furthermore, is contrasted in law with "arbitrary" government interference. And yet, in several ways, linguistics has taught that language itself is arbitrary at its core; in language, the boundary between "natural" processes and arbitrary ones is difficult, some would argue impossible, to discern.

Linguists say that language is "arbitrary" in the sense that meaning emerges, not from anything logically inherent in words or their arrangement, but from the specific conventions and expectations shared by members of a given speech community, conventions and expectations that can and do change dramatically from time to time and place to place. Aside from language in general and perhaps some very deep-level aspects of syntax, there is very little that is universal, neutral, or mechanical about human languages. This insight grew out of the observation that languages differ profoundly from one another, not only in terms of the meanings of specific words, but in terms of basic aspects of the ways those words are arranged: some languages have only two or three words for color, for example, others have nothing English speakers would recognize as verb tenses. But it has also been bolstered by detailed analysis of the workings of language in general. Meanings are fixed neither by logic nor by some natural relation of words to things, but by the contextual and shifting system of interpretation shared by the members of a given speech community.

The arbitrariness of language presents two problems for traditional thinking about freedom of speech. One problem involves legal interpretation, the belief that properly expert judges and lawyers following the proper procedures can arrive at the correct interpretation of a dispute. Often described as the problem of the indeterminacy of law, the purely contextual character of meaning would suggest that legal decisions will always be forced to fall back on contingent, social or political values to decide where the boundaries in the law lie.[5] It is in the character of language, in other words, that a judge will never be able to look at the text of the Bill of Rights and legal precedents to decide whether or not flag burning is protected by the First Amendment; she will always in one way or another be forced to make a choice about whether or not she thinks it should be protected, and will always be faced with the possibility that a reasonable person could plausibly disagree.

Indeterminacy should not be mistaken for the absurd assertion that any word can mean any thing, that there is no stability to meaning whatsoever. As deconstructionist literary critic Barbara Johnson puts it,

[L]anguage cannot itself be entirely reduced to interpretability. This does not mean that language never means, but rather that beyond the apparent meaning, and even beyond the suppressed or hidden meanings (unconscious, poetic, ideological, counterdiscursive), there can always be a residue of functioning--which produces effects--that is not a sign of anything, but merely the outcome of linguistic rules, or even of "the absolute randomness of language." Not that language is always absolutely random, but that we can never be sure that it isn't.[6]

One can believe in indeterminacy, then, and yet say, "it makes sense to me, today," or even "it makes sense to us" that the text of the First Amendment means flag burning should be allowed. But in the face of the arbitrariness of language, it is no longer possible to coherently claim that "it will always mean that to any properly expert and rational person in any time, in any context."

B. The Inevitability of Context and the Impossibility of Formalism

A second problem suggested by the arbitrariness of language involves the impossibility of abstracting from context that is a linchpin of the formalist legal logic which today dominates thinking about freedom of speech. According to some understandings of the rule of law, justice is best served when applied according to indisputable, clear rules of procedure and decisionmaking. Hence the First Amendment protects Nazis marching in Skokie and flag burning, not because anything good is being accomplished in either case, but because the important thing is to uphold the rules impartially and unequivocally. And being impartial and unequivocal typically means that rules are upheld regardless of context.

If one were to suggest, say, that the harm from Nazis marching in a Jewish suburb outweighs the value of protecting their speech because of the history of the Holocaust and the irrational and violent character of Nazi ideology, or that flag burning is such an ineffectual form of political expression and so potentially offensive that nothing would be lost by restricting it, the formalist counterargument is that this would "blur" the boundaries, cross what lawyers call the bright lines, upon which our system of justice rests: the rules are more important than the context.

An important example of formalist reasoning is the Bellotti case, in which the Supreme Court struck down a Massachusetts law limiting corporate campaign donations. The Court reached its decision, not simply by weighing the positive and negative effects of the law, nor by deciding that it was a good thing in this case to grant large corporations the same rights as private individuals. The decision was based on the argument that even considering the source of the campaign donations (the "speech" in question) was inappropriate; every individual has a right to unrestricted political speech, and even asking whether corporate "individuals" are as worthy of protection as ordinary individuals would blur the bright lines upon which the rule of law is based.[7] Another example would be American Booksellers Association, Inc. v. Hudnut, when the court threw out an anti-pornography ordinance. The court argued that, even if pornography has negative effects, the same might be said of other forms of protected speech. From this it concluded that "[i]f the fact that speech plays a role in a process of conditioning were enough to permit governmental regulation, that would be the end of freedom of speech," and thus negative effects do not justify restrictions. As Stanley Fish has pointed out, this is a peculiar logic: faced with facts which call into question the speech/action distinction which underlies the law, the court upholds the law against the facts which would undermine it. But it is a typically formalist logic: the point is to uphold the rule of law, i.e., abstract, neutral principles and procedures; if the coherence of those abstract principles is threatened by facts, you throw out the facts, not the principles.[8]

The problem is that, if the meanings of statements emerge from convention, from social context, then the insistence on excluding context, on divorcing rules and their enforcement from social and political complexities of a situation, is an impossibility. This is not simply an argument that it would be reasonable to sometimes include a little bit of context in legal decisionmaking, that First Amendment law should lean towards a more policy-oriented weighing and balancing of principles and rights in special circumstances such as highly concentrated or technologically inaccessible media. Rather, the argument is that formalist arguments of free speech can not be doing what they claim, that context is present in decisions in spite of claims to the contrary. Decisions that grant protection to marching Nazis and flag burning are not simply decisions that show a preference for bright line rules over context; on the contrary, such decisions are themselves a product of a particular social and historical context, and in turn contribute to the making of particular contexts.

C. The Speech/Conduct Distinction

The collapse of the boundary between "natural" speech and arbitrary interference with it implied by indeterminacy creates a further problem for First Amendment interpretation: the collapse of the distinction between speech and conduct or speech and action. The exercise of free speech, the "free marketplace of ideas," is imagined as a kind of neutral, free and equal exchange, contrasted with unfree or arbitrary coercion. What disappears in the face of the arbitrariness of language is the coherence of that contrast, the faith that there is an important categorical distinction between people talking and arguing and people coercing one another through some kind of action. It is now an axiom of sociolinguistics and many other schools of thought that language use is an important kind of social action, that words do not merely reflect reality or express ideas, they primarily are a way of doing things, a way of acting in the social world. Although J. L. Austin began his classic How to Do Things With Words by describing a limited category of statements that do things--"performatives"--he later enlarged the category and made its boundaries much less clear by acknowledging the frequency of "indirect performatives," i.e., statements that might appear to be merely descriptive but in context can be shown to be in fact doing something.[9] Some have since argued that in a sense all utterances are performatives.

None of which is to suggest that a subtle verbal snub is identical to punching someone in the nose. We do not call trespassing on someone's lawn and shooting them identical, though they are both categorized as violations, as coercive. When Stanley Fish argues that speech in everyday life should not be imagined as if it takes place in "the sterilized and weightless atmosphere of a philosophy seminar,"[10] or when Matsuda et. al argue that words can wound, the argument is not that every slight or insult ought to be treated as if it were assault and battery.[11] What they are criticizing is the belief that there is a fundamental, categorical dichotomy between speech and conduct, that the dichotomy is clear and generalizable enough to form one of the principle structures of our law and democracy.

D. Abstract Individualism vs. the Construction of Subjectivity

All this points to a deeper critique of the marketplace metaphor. The metaphor implies that linguistic exchanges, like market exchanges, take place between individuals who, in the absence of some outside interference, exist merely as individuals, not as persons in particular contexts with particular backgrounds. These are the famous abstract individuals of legal liberalism, the persons referred to as "A" and "B" in law school lectures on contracts: persons bereft, in legal liberalism's ideal world, of gender, class, ethnicity, history. People the world over, the marketplace metaphor suggests, all share the characteristics of being in essence rational, self-interested individuals, inherently active and desirous. Language use, then, is a matter of expressing pre-existing interests; it is a tool used by individuals to buy cheap and sell dear in the marketplace of ideas. Language is something one uses.

But, according to at least some schools of linguistics and language philosophy, language is also something that happens to us, something that "speaks us" as much as we speak it. Language is an inherently collective, social precondition to individuality. Most definitions of language exclude any notion of a language possessed by only one individual; for language to be language it must be shared. People do not choose, after all, their first language; in a sense it chooses people. And the particularities of the language that chooses people, many would say, in turn shapes their consciousness, their sense of what counts as reason, their perceptions of the world and their selves within it, even their desires.[12]

This is not to imply, however, some kind of simple social determinism. Here is where the linguistic turn in philosophy suggests something very different from the common assertion that individual behaviors are "caused" by social structures. For one of the central discoveries of linguistics and language theory is what Barthes called "a paradoxical idea of structure: a system with neither close nor center."[13] Except for analytical purposes, linguistic structure does not exist outside of anyone's use of it. Language is certainly structured, in some sense of that word; linguistic grammar is the central example of structure, although scholars have brought to our attention many higher-level structures like linguistic style. But that structure is not simply some kind of exterior constraint, a Hobbesian limit on individual action; it is not the "structure" of, say, Durkheimian sociology or orthodox Marxism. It is dynamic, changing, and creative. As Chomsky pointed out, one grammatical system is capable of generating an infinite variety of sentences. And grammar is a practical, thoroughly collective human accomplishment, not an exterior system imposed upon individuals by a reified "society." It is enabling as well as constraining: linguistic structure is a precondition of self-expression, not just a limit to it.

Language thus troubles both legal liberalism's happy vision of rational individuals and its dark side, its Hobbesian view of society as the basic constraint on individuals; it calls into question the marketplace metaphor's notions of both individual freedom and social order. The attraction of the marketplace metaphor in law is much the same as the attraction of marketplace theory itself: it posits a realm that is both free of arbitrary constraint, and yet ordered by the certain yet neutral and unequivocal rules of the marketplace. What the fact of linguistic structure calls into question is not merely the "freedom" of linguistic exchange but also its certainty, its divisibility from "arbitrary" external restraints and interference.

E. Summary: The Critique of Free Speech is a Critique of Formalist Understandings of the Rule of Law

When MacKinnon argues that pornography is a form of action, not of speech, or when Matsuda argues that the context of racism and the subjective experiences of minorities in the U.S. ought to be a primary consideration in the creation and interpretation of hate speech laws, in the long run what motivates these scholars is not just a desire for specific exceptions to an otherwise intact First Amendment doctrine.[14] The suggestion is not simply that pornography is so damaging, or that the specific horrors of slavery and its legacy of racism so evil that unusual exceptions to free speech protection are called for (though the evils of rape-culture and racism very well might be the most urgent problems in the U.S. today). Rather, the suggestion, at least implicitly, is that the evils of rape-culture and contemporary racism force us, or should force us, to fundamentally reconsider how American law thinks about freedom, speech, and their regulation.

Furthermore, the critique of the oppositions that underpin free speech law such as speech and action, rules and context, or politics and law, need not be read as a simple denial that any differences exist. It is obviously not the case that there is no difference between slighting someone with a racial epithet and hitting them in the head, or between decisionmaking in courts and decisionmaking in legislatures. The argument is rather that these differences are neither clear nor generalizable enough to coherently underwrite a system of decisionmaking that claims to be able to transcend context and achieve the neutrality that is the goal of law in the first place.

IV. Towards an Analysis of the Context of Free Speech: the Politics of the Rule of Law

Inquiry does not come to an end when one accepts the criticisms of the formalist First Amendment framework, and acknowledges the inevitability of politics and context. Stanley Fish's quip notwithstanding, there is such a thing as free speech. If something is not what we think it is, it does not follow that it does not exist. Free speech is one of the major and most influential political and legal discourses of this century; for better or worse, it has helped make American society, our world, what it is. So the task is to rethink the character of free speech, to specify its historical context and political incidence. This is a large task; here I can only speculate about one aspect of the historical context of free speech, its relation to notions of the rule of law, and one aspect of its political incidence, its relations to social class.

The concept of a neutral, objective system of law that transcends politics is not just an abstraction important to lawyers and judges. (Lawyers and judges, in fact, are often acutely aware of just how political and unstable legal interpretation can sometimes be on a day-to-day basis.) A faith in the neutral rule of law is an important element of American culture, of the popular imagination. Evidence for this can be seen in the way that legal institutions and documents are more often celebrated, more often used to define American democracy, than political institutions and accomplishments. One might think, for example, that in an electoral democracy the most important historical event, the event most widely celebrated, would be the extension of the vote to the majority of the population. Yet most citizens do not know the amendment or the year in which the vote was extended to women, much less the history of the long political struggles that led to the passage of the nineteenth amendment in 1920. On the other hand, the Constitution is regularly celebrated in fora ranging from scholarly conferences to reverential Philip Morris ads, even though that hallowed document underwrote a legal system that upheld slavery for three quarters of a century, excluded women from voting for more than half a century after that, and did not come to rigorously protect political dissent until about fifty years ago. Nonetheless, American culture tends to worship the Constitution and remain ignorant of the history of universal suffrage. The story of the Constitution is a story of law, whereas the story of women's suffrage is a story of protracted political struggle. And in some ways, at least, mainstream American political culture worships the former more than the latter.

A. Free Speech and the Rule of Law: Two legal logics

What is the substance of this worship? What makes law neutral, and how does it support democracy? The short answer might be that if a society makes its decisions according to fixed rules instead of individual or collective whims, individuals will be less able to gain systematic advantage over others. The long answer would involve an extended and controversial discussion of a large chunk of the literature of legal theory and political science. But there is a mid-range answer based in historical observations, which suggests that in the U.S. two patterns of argument or logics have tended to shape legal decisionmaking, particularly in this century. One logic has been called alternately formalist, classical, bright line, rule-based, or simply legal justice; the other, standards-based, revisionist, policy oriented, realist, or substantive justice.[15]

Arguably, the First Amendment has become the centerpiece of the American faith in the rule of law in this century, and not coincidentally, First Amendment law is also highly formalist. Formalism is not simply absolutism, a belief that there should be no exceptions. It is more a way of thinking about what law and legal interpretation are and how they work. (Describing the ACLU's position on the First Amendment as "absolutist" is thus a bit of a red herring.) In at least many of its variations, formalism involves the claim that law is apolitical and neutral because it rests on a rigid, formal model, based on an ideal of axiomatic deduction from rules and unequivocal, "bright line" legal distinctions. The role of law, then, is to locate and uphold clear boundaries--bright lines--between the rights of individuals and between individuals and the state. Legal language and legal expertise are thought valuable precisely because they provide fixed, rigorous meanings unsullied by the political and social winds of the moment. Given a certain set of legal rules and a certain legally defined situation, it is assumed, a properly trained judge or lawyer, within certain boundaries, can use expertise in legal language and reasoning to arrive at, or at least approximate, the correct interpretation, which is generally a matter of pinpointing exactly where the boundaries lie.

Policy oriented decisionmaking, in contrast, tends to be context sensitive, accepting of blurry boundaries, functionalist, and messier. It is also much more common in legal decisionmaking than popular wisdom would suggest. In policy argument, justice is thought to be best served by subtle, well-informed analyses of particular contexts and judicial "balancing" of competing interests and principles; rights and values are treated, not as hard rules distinguished by bright lines, but as general standards that can be differentially implemented according to context. Administrative law, such as that involved in enacting the Federal Communication Commission's public interest standard for broadcasters, is a classic example of policy oriented decisionmaking. Brown v. Board of Education also includes some exemplary policy argument.

Policy-oriented decisionmaking sometimes is justified in terms of head-on attacks on formalism of the type associated with the critiques of free speech just discussed. Both in practice and in theory, the argument goes, the supposedly "bright line" distinctions upon which formalism is based are rarely if ever as bright as imagined. Stanley Fish's polemic, "There is no such thing as free speech," is a recent example of such a critique, but in some ways his position echoes, for example, Felix Cohen's legal realist argument earlier in the century, in "Transcendental Nonsense and the Functional Approach."[16]

It is important, however, that outside the academy policy-oriented legal decisionmaking has been justified less by theoretical criticisms of formalism as a whole and more by a sense that, in certain limited and specialized contexts, policy-oriented decisionmaking is simply practical. Formalism seems to be the place our culture celebrates the ideal of the rule of law; policy argument seems to be the place where most of the detailed legal work of ordering society goes on. Policy argument dominates largely in domains unrelated to communication: the law of corporations, environmental law, urban planning, and so forth. The prominent example of policy logic in communication is probably government licensing of broadcast stations according to the public interest standard. Licensing was originally created because communication by radio waves was understood to be characterized by spectrum scarcity and other complicated and contingent technical matters, such as rapidly evolving technologies and strategic needs of the military. Treating broadcasters differently than newspapers was thus thought to be simply called for by context, not because there was thought to be a formal right or principle at stake such as the public's right to access to communication.

B. The Political Need for Formalism and the Rise of Free Speech

It is sometimes suggested that policy arguments began to replace formalist ones in legal argument somewhere around the turn of the century, and formalism was finally defeated with the end of the Lochner era in 1937. On the level of legal metatheory, there may be truth to this, but it remains the case that in practice both logics remain today. Sometimes the two logics are associated with competing sides in a legal controversy. The argument that television violence ought to be censored because its measurably harmful effects on children outweigh considerations of free speech is a typical policy argument; arguing against such censorship because it would open the door to more serious restrictions of freedom of speech is to lean in a formalist direction. But the two logics are also often mixed in the context of any given argument. Conservatives argue that broadcast licensing violates free speech rights but also is inefficient in the context of new technologies; liberals argue that guarantied citizen access to mass communications would be beneficial for industrial society but also should be treated as a "new First Amendment right."[17]

So it is perhaps the case that what has been changing over the years is not simply a shift from one kind of argument to the other, but a shift in the "mix" of the two, a shift in how the two kinds of argument have been used in which cases. And here the historical literature suggests that, gradually in this century, the focus of formalist argumentation has shifted from the realm of property and contract to free speech. Up through the late nineteenth century, during what Mensch calls the classical era of jurisprudence, property was the central, formal right; in theory property was celebrated as the essence of legal liberalism, and in practice it was used aggressively in a wide variety of areas. Property rights were invoked to justify bans on speaking in public parks, the picketing of factories during union drives, and turn-of-the-century social legislation. Gradually, this formalist application of property fell out of favor, and met its final demise in the 1937 overturn of Lochner, during the New Deal.[18]

Perhaps it is not entirely coincidental that, as formalist notions of property declined, the formalist understanding of free speech rose. In a familiar history, the First Amendment was gradually elevated to its current legal status, both in case law and in the popular imagination. What has triumphed in this period is not a policy-oriented understanding of free speech (in spite of the best efforts of a long line of scholars from Alexander Meiklejohn to Sunstein, but a rigidly formalist one. So today, property rights advocates who would like to see a return to something like the Lochner era interpretations of property, like Richard Epstein, argue that the rules applied to free speech should also be applied to property. Conversely, from somewhere towards the other end of the political spectrum, Cass Sunstein has called for "A New Deal for Free Speech" wherein the 1930s revisions of property law be extended to communication.[19]

Why has formalism in legal discourse shifted from property and contract to free speech? At this point, I can only speculate. It's possible to put a cynical economic interpretation on the shift: Formal interpretations of property were abandoned because they became increasingly impractical in the face of the bureaucratic corporate form of business and other late nineteenth and early twentieth century economic developments. Conversely, the soap box speakers became sanctified in law precisely during the historical period that they ceased being effective. In the nineteenth century, union organizers, pacifists, and other "radicals" all made good use of the soap box--of face-to-face speaking in public places--as a communicative tool, and were regularly arrested for doing so. In this century, however, the key to popular communication has become access to radio, television, and other expensive technology-based mass media, which have rendered the soap box increasingly irrelevant as an organizing tool. A formalist interpretation of the First Amendment grants symbolic protection to soap boxes while in practice protecting media corporations much more effectively than dissidents.

Such an account of the shift, however, risks a functionalist tautology (explaining historical events in terms of the needs they serve for the power bloc) and fails to account for the imaginative power of First Amendment formalism. So a more comprehensive explanation might add two observations. First, from a distance, formalism is satisfying to a legal liberal vision of the rule of law, whereas policy argument can appear as arbitrary, obscure, and haughtily technocratic. College sophomores have little trouble understanding why it might be good for the rule of law to protect Nazis marching in Skokie, but it takes a lot of effort to convince them of the grand principles at stake in, say, the regulation requiring TV stations to charge political candidates the same rate for advertising time they charge their most favored advertiser instead of their standard rates. Second, from up close, from the perspective of those involved in everyday, small legal decisions, formalism is frequently impractical, whereas policy-oriented decisions seem reasonable and pragmatic. Few suburban homeowners would take kindly to the suggestion that their neighbors should be allowed to raise pigs or let their lawns go to weed on the grounds that to do so would be to uphold the sanctity of formal property rights.

It seems to be the case, then, that the American polity seems to want a legal system that can satisfy both the desire for legitimacy provided by formalism and the "practical" effectiveness of policy-oriented decisionmaking. Perhaps, therefore, the formalist interpretation of the First Amendment became popular in part because it came to take property's place as a symbol of legal clarity and formal justice. In both the popular and legal imaginations, the image of the property-holding yeoman farmer was gradually supplanted by the soap box speaker as the central archetype and emblem of legally protected exercise of rights and freedoms in a democratic society.

C. The class context of free speech

1. Labor and Management

The polity, however, is not the public. The community of individuals who appreciate the formalist interpretation of free speech may include a wide range of people, such as lawyers, judges, politicians, journalists, professors, and many others in positions to directly or indirectly influence legal and political consciousness. And it includes a wide range of political positions: liberals at the ACLU seem to have little trouble agreeing with conservatives on the Supreme Court that flag burning is protected speech. But it certainly does not include everyone. The majority of the American public has a hard time seeing the justice of protecting flag burning. And this may not mean simply that the public disdains free speech. The ACLU reports that the majority of the complaints it receives come from workers who feel their speech has been restricted by their bosses--a kind of speech that the Supreme Court and the ACLU agree is not protected.

Elizabeth Mensch has remarked that, although many formerly bright lines have been blurred in twentieth century law, the boundary between capital and labor remains as bright and impermeable as ever.[20] The First Amendment, as it is currently interpreted, protects owners and managers more than individual speakers. It prevents government agencies from interfering with the speech of private agencies delineated by boundaries of ownership and management, not by individual human beings.

As a result, employees have basically no free speech rights with regards to their employers, including employees of media businesses. When a journalist is told by an editor to drop a story because it is politically inflammatory, the journalist can find little comfort in First Amendment law. Network program practices departments engage in systematic and thorough censorship of scripts for television series with all the zeal (if not the same principles) of Communist Party apparatchiks. Under law, there's a sense in which A. J. Liebling's bon mot--that the only freedom of speech in this country is for those who own one--is literally true.

For all that, Liebling's quip is an oversimplification. There are many limits on the power of media owners to influence content, such as the resistance of the community of professional journalists to owner manipulation on both ethical and self-interested grounds. Evidence suggests that, among some groups, there probably is a popular ethic of free speech in the U.S. that extends beyond the powers of owners and managers. When conservative newspaper tycoon Rupert Murdoch bought the left-wing Village Voice and tried to dismiss its editor, for example, the threat of a staff walkout forced him to back down, and he left the paper's editorial content alone thereafter.[21]

2. Social Class and Linguistic Style

Bringing "popular ethics" into the discussion, however, brings us back to the second question suggested by Gates' examples: why does it seem easier to pass rules prohibiting direct racial epithets than elaborate, formal statements? It is well established that linguistic style is associated with social class. Sociolinguist Basil Bernstein demonstrated that children from middle and professional classes tend to do better in school than working class students in part because they speak more often and more fluently in formal style, or what Bernstein calls "elaborated code." Working class students, in contrast, tend to be more comfortable, and are probably more fluent in, informal style, or what Bernstein calls "restricted code."[22]

One style is not better than the other. Rather, each style is an adaptation to specific patterns of life and work. Informal style has the effect of stressing membership within a group; it is useful for interactions among people who are familiar with each other and work with each other on a regular basis, and thus live in "dense" social networks, i.e., high levels of interaction with a limited number of people. It has a high proportion of ellipsis and colloquialisms, not because such language is simpler, but because these take advantage of a higher degree of shared knowledge between speaker and listener. Similarly, it has a higher proportion of personal pronouns (you and they) and tag-questions soliciting agreement of the listener (nice day, isn't it?), because these express a sense of cooperation and solidarity.[23]

Formal style, in contrast, is for people whose social networks are less dense, who regularly deal with strangers and thus communicate in contexts in which ellipsis and colloquialisms are more likely to generate confusion than solidarity. Similarly, formal style's high proportion of subordinate clauses, passive verbs, and adjectives (besides connoting high-mindedness through its echo of Latin grammar) are adaptations to the need to explain details comprehensively when speaker and listener do not share as much background knowledge and cannot easily rely on features of the extra-linguistic context. Interestingly, in spite of the frequency of passive verbs, formal style also contains a higher proportion of pronoun "I." This has the effect of imposing the speaker's individuality on the utterance, of stressing her or his unique nature as a person, as opposed to expressing membership in a group. Some research suggests that formal style leads people to be judged as more intelligent, more educated, and less friendly and less likable than informal style.

It is not the case that working class people use only informal style and middle class people use only formal style. A garage mechanic will probably shift to formal speech when dealing with a customer irate over a bill, and only the most hopelessly pompous college professors use formal style when speaking with their friends and families. But mastery over the different styles is not evenly distributed. Bernstein's work suggests that middle and professional class students' relatively better skills and comfort with formal style functions as a form of what Bourdieu calls "cultural capital," enhancing their life prospects.[24] Given the relation of style to the character of work, moreover, fluency in formal style (though not accent) is probably associated with a person's present occupation, regardless of class background.

What does this have to do with free speech? James Carey has argued that the speech/action distinction in free speech law is an expression of distinctly middle class values and sensibilities. Carey tells the story of a middle class man who enters a working class bar and not long thereafter comes flying out the plate glass window; the man then says with astonishment, "but all I did was use words!" Carey's point is that, to the working class individuals in the bar, words have power. For them, the difference between insulting someone's mother and punching them in the nose is not as obvious or absolute as it is for the middle class person.

Carolyn Marvin has elaborated on these contrasting sets of values in our culture in terms of what she calls "text" and "body":

Text is traditionally the weapon of those whose cultural power and entitlement to participation in social life derive from educational and other textualizing credentials that exempt them from expending their bodies in pursuit of social resources. Those who command the text are most entitled to preserve their bodies and shield them from physical effort and danger. The body, by contrast, is the emblem and resource of those without textual credentials, whose bodies are available to be used up by society and whose power and participation derive from whatever value their bodies have for cultural muscle-work. . . . Text and body, therefore, represent competing concentrations of economic and social power, characteristically different concepts and artifacts, and distinctive logics and frameworks of value . . . Where the body is thought to be concrete, the text is considered abstract. Where the body is seen as particular, the text is described as universalizing.[25]

Those who celebrate the Bill of Rights, Marvin argues, are those who live text-oriented lives; yet as she points out, the Bill of Rights does not wave over the ballpark. The values of formalist free speech, then, are the values of what Marvin calls text: the abstract and universalizing values of people who do not have to "expend their bodies in pursuit of social resources," who lead a life that involves frequent contact with strangers, and who value a formal linguistic style. The point is not that those who value texts think words matter, and that those who value bodies think action matters. The point is that text and body represent different ways of conceptualizing the relation between words and actions. The dualism of speech and conduct, the sense that there is some fundamental, profound difference between talking and acting upon which one can base neutral, apolitical, and universally applicable rules for democracy, is an expression of the world view of people who make a living with words at relatively little expense to their bodies. Although it may be difficult for a text-valuing person such as myself to appreciate the point of view of a non-textual person, perhaps for those who work primarily with their bodies, words are simply an extension of bodily action.[26]

The First Amendment as currently interpreted is envisioned largely in terms of that which middle and professional class people have mastery over, abstract formal expression in speech and writing. This is why it is harder to censure Gates' first example than the second. Within the community of people who share those values, there is something equalizing about free speech. But it should not be surprising that, for people who do not make a living that way, for workers and other people whose bodies are the source of their value to society, formalist protection of free speech may not make sense, and might even appear as simply another way that people with privileges (such as academics writing about free speech) exercise their power over people who don't.

V. Conclusion

The analyses and arguments of this chapter do not offer resolutions to all of the many important debates among non-formalist theorists of freedom of speech, such as those between Gates and Matsuda et al. over campus hate speech codes. But it does do two things. First, it tries to clarify some of the underlying principles and issues at stake today in debates over free speech, particularly the inevitability of context and the problems this poses for traditional formalist understandings of the rule of law. Second, it points in the direction of a rethinking of free speech based in context, and suggests two (among many possible) avenues to pursue: the historical shift of formalism from property to free speech and to matters of language and social class in both legal discourse and in nonlegal situations. Clearly, these examples of context-based analysis are intended only to be suggestive. But what they suggest, it is hoped, is that this kind of inquiry, if expanded into rich and subtle contextual analyses, might indeed help resolve some debates and contribute to a more fully democratic, substantive interpretation of the role of free speech in law and culture.

[1]. Henry Louis Gates, "Let Them Talk," The New Republic, Sept. 20 & 27, 1993, pp. 37-49: p. 45.

[2]. "Style" is the generally accepted sociolinguistic term for language varieties that can be classified on a continuum for formal to informal. The word "code" is used by Basil Bernstein, Class, Codes And Control, 2d edition (Boston: Routledge & K. Paul, 1974).

[3]. William Labov, "The Logic of Nonstandard English," in Giglioli (ed.) Language and Social Context (Penguin, 1972), pp. 179-216.

[4]. For a sociolinguistically informed analysis of the role of linguistic style during arrest and interrogation see, Janet E. Ainsworth, "In a Different Register: The Pragmatics of Powerlessness in Police Interrogation," Yale Law Journal, 103 (November, 1993): 259-322.

[5]. Mark Kelman, A Guide to Critical Legal Studies (Cambridge, Mass.: Harvard University Press, 1987), p. 12 and passim.

[6]. Barbara Johnson, A World of Difference (Baltimore: Johns Hopkins Univ. Press, 1987), p. 6.

[7]. First National Bank of Boston v Bellotti, 435 US 765, 776 (1978)

[8]. 771 F.2d 323 (7th Cir. 1985), aff'd, <=8> 475 U.S. 1601 (1986), p. 329; quoted in Stanley Fish, "Fraught With Death: Skepticism, Progressivism, and the First Amendment," University of Colorado Law Review, 64 Fall 1993: 1061-1086, p. 1065.

[9]. See Ainsworth, "In a Different Register," note 15: "Austin initially adopts the intuitively appealing assumption that constative utterances, unlike performatives, are true or false. Having set up these opposing categories of performative and constative utterances, Austin ultimately deconstructs this dichotomy" with his analysis of indirect performatives.

[10]. Fish, "Fraught With Death," p. 1061.

[11]. Mari J. Matsuda, Charles R. Lawrence III, Richard Delgado, and Kimberle Williams Crenshaw, Words that Wound: Critical Race Theory, Assaultive Speech, and the First Amendment (Boulder, Colorado: Westview Press, 1993).

[12]. The classic and extreme version of this notion is the "Sapir-Whorf hypothesis" named after linguists Edward Sapir and Benjamin Whorf. For a post-structuralist variation of it, see Rosalind Coward and John Ellis, Language and Materialism: Developments in Semiology and the Theory of the Subject (London: Routledge and Kegan Paul, 1977).

[13]. Roland Barthes, Image, Music, Text (New York: Hill and Wang, 1977), p. 159.

[14]. Catharine A. MacKinnon, Only Words (Cambridge, Mass: Harvard University Press, 1993).

[15]. Elizabeth Mensch divides legal thought into classical and realist or revisionist forms. Duncan Kennedy talks of the distinction between rules and standards. Roberto Unger speaks of "legal justice" and "substantive justice." See Elizabeth Mensch, "The History of Mainstream Legal Thought" in David Kairys, ed., The Politics of Law: A Progressive Critique (New York: Pantheon, 1982), pp. 18-39; Duncan Kennedy, "Form and Substance in Private Law Adjudication," Harvard Law Review, 89 (1976): 1685, pp. 1687-89; see also Roberto M. Unger, Knowledge and Politics (New York: The Free Press, 1975), p. 91.

[16]. Stanley Fish, "There's No Such Thing As Free Speech And It's a Good Thing Too," Boston Review, Feb. 1992, p. 3; Felix Cohen, "Transcendental Nonsense and the Functional Approach," Columbia Law Review 35 (1935): 809.

[17]. For example, Jerome A. Barron, Freedom Of The Press For Whom? The Right Of Access To Mass Media (Bloomington: Indiana University Press 1973).

[18]. Jennifer Nedelsky, Private Property and the Limits of American Constitutionalism: The Madisonian Framework and Its Legacy (Chicago: University of Chicago Press, 1990).

[19]. Cass R. Sunstein, "Free Speech Now," The University of Chicago Law Review, 59 (Winter 1992): 255; Richard A. Epstein, "Property, Speech, and the Politics of Distrust," The University of Chicago law review 59 (Winter 1992): p. 41.

[20]. Mensch, "The History of Mainstream Legal Thought," p. 26.

[21]. Alex S. Jones, "At Village Voice, A Clashing Of Visions," The New York Times, June 28, 1985, Section B; p. 5, Column 1.

[22]. Bernstein, Class, Codes And Control.

[23]. This survey of Bernstein's work relies heavily on Peter Trudgill, Sociolinguistics: An Introduction to Language and Society (London: Penguin Books, 1983, revised edition), pp. 132-140.

[24]. Pierre Bourdieu, Distinction: A Social Critique of the Judgment of Taste, trans. R. Nice (London: Routledge & Kegan Paul, 1984).

[25]. Carolyn Marvin, "Theorizing the Flagbody: Symbolic Dimensions of the Flag Desecration Debate, or Why the Bill of Rights Does Not Fly in the Ballpark," Critical Studies in Mass Communication, 8, (June, 1991): pp. 120-121.

[26]. Social class is of course a complex construct, and is used here suggestively, not comprehensively or precisely. Marvin points out that the values of "body" in fact extend to and in many ways are exemplified by military personnel, a group which overlaps with but is not limited to working class individuals.