The Broken Marketplace

Conversations about free speech often invoke the marketplace of ideas: that the answer to “bad speech” is “more speech” because vigorous public dialogue winnows good ideas from bad ones, allowing people to distinguish what is empirically correct or logically consistent or morally justified from what isn’t.

Of course, in most contexts, “free” speech is an unfortunate misnomer. It doesn’t mean “having no restrictions whatsoever”. It means something closer to “appropriately regulated”. There’s a long tradition of restricting various forms of speech: for instance, speech that incites violence or will cause immediate harm (e.g., shouting “Fire” in a crowded theater) or speech that serves no valuable political or social purpose (e.g., spam, obscenity). The point of “free speech” is to enable productive discussions – political, social, scientific, and moral ones.

Free speech, then, is seen as an important enabling condition for the development of ideas and taking “correct” collective action – electing the better candidate; applying the better policy; developing the more useful technology. But it’s hardly the only important factor. In logical terms, free speech is a necessary condition for the “marketplace of ideas” to elevate good ideas over bad ones – it’s far from a sufficient condition.

Consider all the things free speech regulation does NOT do:

• Speech regulation does nothing about the practical barriers that make some arguments (and evidence) more accessible than others. Knowledge and information often becomes "bottled up" within certain communities or resources. The obvious example is with language: the overwhelming amount of scientific research is conducted in English, creating barriers to understanding regardless of how "freely accessible" that research is. A more practical example is simple economics. Producing and distributing knowledge costs resources. Scihub, open access journals, and other projects expand access to information, but I can't imagine a world that could completely eliminate such barriers.
• All speech takes place in networks where the speakers have differential access to “listeners” and different levels of influence with those listeners. This is certainly true online, but in-person gatherings do not magically eliminate existing structures of influence.
• Free speech laws say nothing about whether people understand – or have the capacity or incentive to understand – the speech they are listening to. This includes understanding the language, culture, and necessary background inforation to interpret the speech itself.
• Even granting the capacity and motivation to understand arguments, humans do not necessarily process information effectively. The long literature on cognitive biases and heuristics suggests that under some circumstances humans can be excellent information processors and, in other circumstances, they can be abysmal ones.
• Finally, speakers (and listeners) have to have skill at arguing. It is not just the first unrestricted utterance that concerns us; it’s how people dissect that utterance, interpret it, respond to it, re-purpose it – and what our collective goals are for arguing in the first place. The many examples of arguments that we run into as children and adults, I would argue, are not really arguments at all, but facsimiles of arguments. Presidential debates, courtroom dramas, and debate competitions, for instance, tend to be about "winning" (in some form) and not actually developing the arguments at hand.

The point of the marketplace of ideas isn't just that the best ideas will rise to the top with proper speech regulation. It's that discussions, arguments, and collaborations develop ideas and that we are collectively better off when we create social, cultural, and technological structures to enable such speech. But our development on that front has been...less than satisfactory.

Next week we'll begin exploring some of these points in more depth. See you then.