Free Speech and the Marketplace of Ideas

Saving a Bad Metaphor

Very short summary: In this essay, I discuss the relationship between free speech and the metaphor of the marketplace of ideas. I explain why the metaphor is problematic and does not support the libertarian conception of free speech. However, I also argue that besides the marketplace metaphor, issues related to free speech can be tackled with a minimal set of economic concepts and principles initially used to study market processes and behavior.

---

Introductory note: My wife and I spent a few days in London last week. We had a fantastic time, enjoying two plays (“Romeo and Juliette” at Shakespeare’s Globe and a play with Ewan McGregor at West End!), a visit at the Parliament and at Westminster Abbey, sightseeing and, of course, bookstores. There are many great ones but if I had to give my top 3 based on the book selections and atmosphere, that would be Waterstones (the one on Gower St), Hartchards Picadilly, and the London Review Bookshop. I’m glad that I’ll be back in London in July for the PPE Society conference at King’s College. Today’s essay is (very) loosely connected to England, with some photos as an extra!

---

Behind a significant amount of hypocrisy, recent debates over free speech in North America and Europe hint at an old and fundamental question: how can we discover truth or, more precisely, what is the best way to organize society so that we favor the discovery and spreading of true ideas? In this context, the metaphor of the “marketplace of ideas” (or “market for ideas”) keeps on being invoked, especially by the supporters of the most libertarian conception of free speech.[1]

The popular view states that the marketplace of ideas metaphor has roots in England, though its modern interpretation and use are decidedly North American. The metaphor alludes to a claim made by the poet John Milton that the surest and most effective way to discover truth is to let ideas be freely expressed and circulated. The second chapter of John Stuart Mill’s On Liberty is where we find the most eloquent early developed statement of this view. Mill argues that if “all mankind minus one, were of one opinion, and only one person were of the contrary opinion, mankind would be no more justified in silencing that one person, than he, if he had the power, would be justified in silencing mankind.”[2] Crucially, Mill defends this claim based on the social benefits of letting people free to express their opinions. First, “the opinion which it is attempted to suppress by authority may possibly be true. Those who desire to suppress it, of course, deny its truth; but they are not infallible.”[3] As important, true or not, beliefs (and the people who hold them) must be permanently challenged: “However unwillingly a person who has a strong opinion may admit the possibility that his opinion may be false, he ought to be moved by the consideration that however true it may be, if it is not fully, frequently, and fearlessly discussed, it will be held as a dead dogma, not a living truth.”[4]

With Chernow’s biography of Alexander Hamilton at Waterstones (I’ll go to the musical in NYC this month)

A careful reading of Mill’s text reveals that he never uses the marketplace metaphor, nor even alludes to mechanisms or principles that may be remotely related to market mechanisms. However, Mill does suggest that true ideas have a distinctive advantage over false ones:[5]

“The real advantage which truth has, consists in this, that when opinion is true, it may be extinguished once, twice, or many times, but in the course of ages there will generally be found persons to rediscover it, until some one of its reappearances falls on a time when from favourable circumstances it escapes persecution until it has made such head as to withstand all subsequent attempts to suppress it.”

This very specific claim appears to underpin most of the American reinterpretation of Milton’s and Mill’s views in terms of the marketplace metaphor. Historical accounts of the metaphor trace its origins in US jurisprudence on free speech.[6] A 1919 Supreme Court Judgment is often mentioned, as the dissenting opinion of Justice Oliver Wendell Holmes Jr. explicitly refers to the “competition of the market” as the “best test of truth.” A 1969 judgment states that “the purpose of the First Amendment [is] to preserve an uninhibited marketplace of ideas in which truth will ultimately prevail.”[7]

A quick search on Google n-gram indicates that the number of mentions of the expression “marketplace of ideas” has dramatically increased since the 1960s:

This reflects the fact that the defense of free speech, as characterized notably by the First Amendment of the U.S. Constitution, is now largely related to the marketplace metaphor. What I would call the “folk view” of the relationship between free speech and the marketplace metaphor runs as follows:

In a society where free speech prevails, ideas are put in competition against each other. Individuals, acting as “producers” and “consumers” of ideas, are incentivized to adopt and act based on true ideas, as doing so will lead to better outcomes for them than if they rely on false ideas. Over the long run, the market process must eliminate the latter, keeping only the former alive.

London’s nicest bookstore (very subjective opinion!)

This is a sophisticated formulation of the folk view, admittedly far more sophisticated than what can be found in its actual expression in the media or even by most academics. But it is useful for two reasons. First, it reveals its Millian roots (the presumption that truth will prevail over the long run). Second, it discloses the mechanisms that are supposed to account for the goodness of free speech. Also, the sophisticated formulation of the folk view has a corollary:

Restrictions on free speech entail a coercive interference with the market process. Such interference may at best accelerate the process of adoption of true ideas. However, because a single individual or organization doesn’t have privileged access to truth, the most likely result of coercion is to force the adoption of false ideas, either as an unintentional outcome or as part of a plan of manipulation or indoctrination.

This corollary makes explicit the connection between the libertarian conception of free speech and the marketplace metaphor. It combines two related but distinct accounts of the market process. The first is the Hayekian argument that the market process takes advantage of the dispersion of local knowledge. Even more than on markets for goods, our views regarding what is true are shaped by localized, idiosyncratic, and tacit knowledge. A free market for ideas is then the best way to aggregate and process this form of knowledge, a task that no central authority (or a single human brain) is able to do. The second is the Friedmanian (though more rigorously exposed by Armen Alchian) argument that market competition triggers an evolutionary mechanism of selection where the best strategies/products/producers progressively displace less good ones. With respect to the marketplace of ideas, this grossly emulates Mill’s reasoning that true ideas have more chances to survive over the long run.[8]

The most admirable politician of the 20th century (subjective opinion too!)

These two accounts and the libertarian conception of free speech they lead to don’t imply that every interference with free speech is wrong. Indeed, on the libertarian view, that private companies like Facebook or Twitter regulate speech on their platforms is not problematic, insofar as this takes place within the competitive market process. In this case, users can just migrate to a different platform. The objection is with respect to interference coming from the state because, here, no exit option exists. By regulating speech, the state blocks the market process and the competition that selects for the best ideas.[9]

Critics of the libertarian conception of free speech have either targeted the libertarian corollary of the marketplace metaphor or the metaphor itself. Both have been largely destructive. A good example of the former strategy is Ronald Coase’s article “The Market for Goods and the Market for Ideas.”[10] Coase argues that insofar as there is a market for ideas, government regulation is more justified than in markets for goods. The reasons are familiar, related to market failures and imperfect competition. Ideas are non-rival “goods” with a low degree of exclusion, which means that their spontaneous production through market mechanisms is unlikely to be efficient and will cause externalities. Asymmetric information is pervasive, and “consumers” have few incentives to invest resources in monitoring information producers. More generally, transaction costs are important, which severely limit the ability of the market process to identify the best ideas. Finally (though not discussed by Coase), we may also note that cognitive biases are likely to impede the selection mechanism of the market.

A more radical critique is to reject the marketplace metaphor, pure and simple, as is the case with Lisa Herzog in her book, Citizen Knowledge.[11] Herzog makes the point that ideas can only be described as goods in a metaphorical and, ultimately, misleading sense. Ideas are not traded or exchanged in the economic sense. Moreover, they are not separable, as they come into more or less coherent bundles. The way we “choose” ideas and the relationship of these choices with our preferences is also quite different from what happens with goods. I’ve formulated the marketplace metaphor above to make this point transparent. What is assumed in the metaphor is that (i) individuals produce and consume ideas to achieve better personal outcomes and (ii) better personal outcomes are more likely to be achieved by producing and consuming true ideas. But neither is true in many, indeed most, of the cases. People can stick to “bad” ideas because very often, this has no direct consequence on them. Even when it does, the effects are very often too indirect and remote in time to provide proper incentives. Because there is no difficulty finding examples where it is definitely in the interests of some agents to produce and spread false beliefs, we have no reason to think that the pseudo-market process will favor true ideas over the long run. At this stage, this is no longer a matter of science, not even of ideology, but of faith.

In Shakespeare’s Globe

The upshot is that other metaphors or analogies than the marketplace are commendable. For instance, Herzog suggests the metaphors of the battlefield (that she argues is a more accurate description of Mill’s view) or of the sport contest, noting that in the latter case, the competitive game cannot proceed without an agreement on the rules and the underlying ethics. This is useful, though not necessarily a major improvement over the marketplace metaphor. The general problem with metaphors is that they are loose and impressionistic. It is therefore easy to be misled by vague intuitions that what is true in one case must be true in another, superficially similar one. What we need, rather than metaphors, is a model.

In this perspective, if the marketplace metaphor is misleading, I would suggest that a genuine economic model of free speech can be relevant. What do I mean by this? Consider the following points:

· Ideas are usually produced, adopted, and used in a decentralized way.

· Ideas are produced, adopted, and used in response to incentives (or lack thereof), which may be monetary, but also symbolic, expressive, or moral.

· It is impossible to predict the outcome of such a decentralized process, or whether the ideas that emerge are “true” (as we don’t know what is true). However, up to a point, we can understand the relationship between the rules that regulate speech (either designed or evolved) and the likely outcomes.

· We know that externalities will be a significant issue because speech can directly hurt or indirectly influence harmful (or beneficial) behavior.

These four points are not based on analogical or metaphorical reasoning. They only follow from the recognition of some of the characteristics of social interactions, where more or less rational agents can communicate ideas by uttering words. Arguably, they don’t describe a marketplace in any meaningful sense but they do correspond to a social setting that is polycentric, where the behavior of agents can be assumed to be minimally rational (i.e., not incoherent) such that it responds to rules (and thus incentives) in a way that can be partially predicted. The rules that apply to this social setting must be designed (or evolved) to solve unavoidable externality problems, which include the fact that individuals are likely to believe falsehoods, but not only.

We can briefly reflect on potential applications of this generic model. First, we may conjecture, for instance, that competition between information producers increases the cost of, and thus reduces attempts at persuasion and indoctrination. Second, a good example of a polycentric social system where ideas are produced and circulated is science. Scientific activities are typically conducted in a highly decentralized way, with a mix of cooperation and competition. The behavior of scientists is not erratic, though it responds to a great diversity of incentives. Scientific activities follow a great diversity of rules aiming at incentivizing and monitoring scientists and favoring the selection of true ideas. We collectively have an understanding that some rules are more likely to help us achieve this objective than others. Finally, externalities are an issue in science, especially when they take the form of falsification of the results.

As a last example, consider the case of politics as the domain of life where political ideas compete to justify coercion. My fellow Substacker Lionel Page has recently published a fascinating post on the relationship between political ideas and interests that can be interpreted through the lens of my generic model. Go read Lionel’s post, but let me quote its last two paragraphs:

Ideologies are shaped by economic and social structures, but an entirely cynical view is not warranted. Ideologies should not just be conceived as artificial ideas masking the reality of naked interests. Instead, they function as general frameworks articulated in the game of morals to adjudicate how resources should be allocated in the game of life. Ideologies that are conceptually inconsistent or rely on non-credible factual claims tend to be weaker contenders in that game. Therefore, political debates neither follow a purely idealistic logic (in the game of morals), nor are they simply determined by material interests (in the game of life).

Ideological debates matter because political ideas have power and can help change the world. But they do not do so because ideas that are “correct” or “right” from a purely abstract point of view win and convince people. Instead, political ideas become successful largely to the extent that they are able to articulate the interests of large coalitions with a conceptual framework based on principles that are consistent and facts that are credible.

In politics, ideas are in competition, and there are mechanisms that tend to select for ideologies based on their consistency and ability to articulate the interests of coalitions. However, this is not the same as to contend, naively, that the “best” or “true” ideas will be the only ones to survive.

My point is that we have economic concepts, traditionally used to study market processes and behavior, that are relevant to understanding how ideas emerge and spread, as these examples illustrate. The economic model is not intended to support a particular stance with respect to free speech, contrary to the marketplace metaphor. The view that true ideas will necessarily prevail on the so-called market for ideas reflects intellectual laziness. The economic model suggests that there is a relationship between rules that regulate a polycentric system and the outcomes that spontaneously emerge from the interactions between minimally rational agents, and that this relationship can be rigorously studied with typical economic concepts and assumptions.

No wonder why my luggage was so heavy!

---

[1] As the beginning of the paragraph indicates, there is a lot of hypocrisy in this debate, especially from the self-proclaimed libertarian camp. Admonishing, even with justification, European countries for banning some forms of speech while bullying American universities on the ground that they let another kind of speech develop is not defending free speech.

[2] John Stuart Mill, On Liberty, Utilitarianism and Other Essays, ed. Mark Philp and Frederick Rosen, Second edition (Oxford: Oxford University Press, 2015), p. 19.

[3] Ibid., p. 19.

[4] Ibid., p. 35.

[5] Ibid., p. 30.

[6] Joseph Blocher, “Institutions in the Marketplace of Ideas,” Duke Law Journal 57, no. 4 (February 1, 2008): 821–89. Lisa Herzog, Citizen Knowledge: Markets, Experts, and the Infrastructure of Democracy (Oxford University Press, 2023).

[7] Herzog, Citizen Knowledge.

[8] F. A. Hayek, “The Use of Knowledge in Society,” The American Economic Review 35, no. 4 (1945): 519–30. Armen A. Alchian, “Uncertainty, Evolution, and Economic Theory,” Journal of Political Economy 58, no. 3 (June 1950): 211–21.

[9] A good illustration of this view can be found here.

[10] R. H. Coase, “The Market for Goods and the Market for Ideas,” The American Economic Review 64, no. 2 (1974): 384–91.

[11] Herzog, Citizen Knowledge. See especially Chapter 5. I have discussed Herzog’s book here.