This post by Michael Moynihan, responding in part to this post by Thane Rosenbaum, asks how “free” free speech should be. The question of discriminating between different forms of speech—based on questions such as “is it knowingly false,” “how likely is it to incite violence,” and “is it political”—is an instantiation of an aggregation problem, exactly the type of problem that motivates the analysis and arguments in the forthcoming book I penned with Maggie Penn, Social Choice and Legitimacy.
But, aside from the question of how one would (or could) construct meaningful and coherent “bounds” on “free” speech, I was led to think about the instrumental nature of speech by the following quote from Moynihan’s post (which includes a quote from Rosenbaum’s post):
“Actually, the United States is an outlier among democracies in granting such generous free speech guarantees. Six European countries, along with Brazil, prohibit the use of Nazi symbols and flags. Many more countries have outlawed Holocaust denial. Indeed, even encouraging racial discrimination in France is a crime. In pluralistic nations like these with clashing cultures and historical tragedies not shared by all, mutual respect and civility helps keep the peace and avoids unnecessary mental trauma.” So one would assume that racial discrimination has been dumped on the ash heap of history in France, considering racist thoughts and symbols have been made illegal. How, then, does one explain that the National Front, whose former leader Jean-Marie Le Pen was found guilty of Holocaust denial, is now the most popular party in the country?
The math of politics point here is both simple and arguably subtle. Basically, speech limitations are not imposed at random, and citizens should draw inferences about the motivations of, and information held by, whoever imposed them.
Consider the classical “marketplace of ideas” justification for strong free speech rights. In a nutshell, this argument says that free speech is socially beneficial because it does minimizes the probability that a “true” (and, by presumption, socially beneficial) argument will be prescreened or forestalled by speech limitations. (Consider, for example, the creationism vs evolution debate.)
My argument here, though in favor of strong speech rights, is slightly different. Specifically, it focuses on constraints imposed by the government. This is an important qualification. In particular, democratic governments are in the end chosen or “produced” through collective action. If “ideas matter” (as the marketplace justification justifiably presumes), then evaluating the policies of the government and its potential successors matter. Then, the transmission of ideas between citizens might lead to changes in/pressures on the government.
Accordingly, if one presumes that governments prefer to maintain power, ceteris paribus, then a policy that discriminates between speech based on content can arguably be informative in its own right.
Here’s a quick sketch: suppose that a government favors some policy that may or may not be socially suboptimal and people have variously informed opinions about the social optimality of that policy.
Suppose that people are prohibited from talking “negatively” about that policy. If people don’t consider the government’s motivation to choose/support such a prohibition, then the prohibition would—for the sake of argument—tamp down dissidence regarding that policy. However, if the citizens think about the government’s motivations—regardless of whether they be policy-based, reelection-focused, or a combination thereof—then the government’s imposition of the prohibition would justifiably lead to the citizens suspecting that not only was the policy in question more likely to be suboptimal, but also that the government does not have the best interests of the electorate at heart. (NO WAY!)
In short, all governments are at least practically dependent upon their citizens’ support. If speech “matters,” then governmental limits on speech—perhaps especially those accompanied by the purest of putative motives—should be viewed with suspicion.
Note that this logic gets even “stronger” once one considers the timing of the limitations: that is, if one thinks about a government considering the (per se) costly imposition of speech limitations that might potentially (in a naive world) mitigate agitation against the government, the fact that the government is willing to incur the costs of imposing such limitations in a particular policy area should make one consider whether the government was alerted to an increased frequency of individuals unhappy with the government in this realm. This “strengthens” the conclusion about the effects of the ban—arguably mirroring the Le Pen example above—-because savvy citizens would infer that the imposition of a limitation on speech on a particular topic is itself indicative of citizen unrest on that issue.
With that quick post, I leave you with this reminder of the most eternal right.
I’ve been reading for a while and I think this is the one I respond to because I have a lot to say.
I read it from bottom to top, so let me first address a more purely theoretical critique unrelated to bigoted French sentiments. Theoretically if the government elites pool (for any reason) their punishment for expression both in states of high anti-policy sentiment and in states of low anti-policy sentiment, a policy of punishing citizen signals on this topic may not imply a large latent sentiment against the policy. Easily, a strategy of complete ambiguation of public sentiments could prevent any downsides resulting from “agitation against the government,” that a signal, whether direct from citizen expression or indirect from potential governmental separating equilibrium actions, would create.
(A good reason governmental elites may pool is that they themselves can’t distinguish different states of public sentiment, id est: they have only one information set. Indeed, if they already know public sentiment it would beg the question of why they need free expression at all.)
Although, in the case of French outlawing of hate speech there seems to be a clear signal of some sort. The general liberal nature of French society makes us presume statements are allowed tacitly rather than disallowed tacitly and only allowed by some form of stipulation (perhaps by precedent). Maybe only in a generally liberal society do stipulations that certain expressions are illegal send signals that support for those positions is large. Stipulation is likely costly simpliciter, and if so a choice to stipulate indicates some counter-balancing cost to the government that balances the cost of stipulation.
Second: If we explain the lack of US holocaust denialism due to the poor signal of support sent by other citizens (who are free from any fear of governmental sanctions which would result from sending such signals), I think we run into some kind of self-referential paradox:
If many people are not expressing their anti-governmental policy sentiment then many people who have anti-governmental policy sentiments will not want to express them (due to a form of peer pressure), Therefore there may exist a large proportion of citizens with anti-governmental policy sentiment even if we do not observe signals of such sentiment.
If we don’t get Holocaust Deniers in the US because they are afraid of the public ostracism that such expressions create (why else is sentiment dependent on signals related to the state of what others believe?), what difference does this fear have with fear of governmental sanctions? Such opportunity costs derived from peer pressures will create similar ambiguities for the true state of public sentiment.
Third: Hate speech is somewhat special in being aggressive. If you will suffer punishment for expression of tolerance from bigots, and will not suffer punishment from tolerant people for expressions of bigotry, bigotry is the only result of a reasonable dynamic process. Establishing a punishment for expression of bigotry is necessary to achieve “tolerance” (which at that point would be imperfect tolerance at best).
That said, it would probably be a dangerous legal precedent to punish bigoted or aggressive expression in US law (if not in general), due to the squishiness of the definition of “bigoted” or “aggressive”.
Fourth: The support for the Tea Party in the US is analogous in many ways to the support of “Front National”, including magnitude. The linked article references a 25% proportion of the French public support that party, and in this recent Gallup poll 22% of respondents consider themselves tea party “supporters” (representing a retreat from the all-time high level of tea party support). Both political entities can be characterized as being largely animated by nativist sentiments, exacerbated by the recent worldwide financial crisis.
http://www.gallup.com/poll/166217/tea-party-favorability-falls-lowest-yet.aspx
Additionally, while Holocaust Denial may not be popular among Tea Party supporters, my guess would be that many tea-partiers support other a-historical racially charged statements more pertinent to US history, e.g. “the War Between the States was a war of Northern aggression and not about slavery.” Even then, these extreme sentiments may represent a minority of both entities’ constituents. While I grant, this is an article from 1993, it indicates a vast majority of French respondents do not believe in Holocaust Denial; so much so that even if all the unaccounted for 6% were Holocaust deniers and National Front supporters, they would not even constitute a majority of National Front supporters. Furthermore, more Americans did not “believe that mass killings of Jews did take place” than French, despite our more open society.
http://www.nytimes.com/1993/12/05/world/french-top-americans-in-poll-on-knowledge-of-holocaust.html