Boyd’s pragmatist theory of reference, maybe

In the previous post, I suggested that there might be no unified “pragmatism”. By this I meant that we wouldn’t (as a matter of philosophical method) want to invent the term if it weren’t (as a matter of the history of philosophy) already entrenched and an actors’ category. I’m not sure if I want to take that back, but I do want to talk about something in the neighborhood of “pragmatism” that probably deserves a name.

In the Pragmatism lectures, William James insists that pragmatism makes meaning and truth a matter of what will happen in the future. This is present in his earlier work, too, and it is picked up by his contemporaries. Writing before Pragmatism, Dickinson Miller insists that James’ philosophy is importantly different than traditional British empiricism. Miller writes:

The analysis of the English philosophers, broadly speaking, was not pragmatic, but genetic; they asked of a conception whence it came; Mr. James asks whither it goes. … Both are empirical; but they appeal to past experience, he to ‘future practical experience.’1

So Miller distinguishes two kinds of empiricist, reductionist accounts of meaning: The traditional account identifies the meaning of an idea in terms of the experiences compiled to construct it, whereas the latter identifies the meaning of an idea in terms of the experiences it predicts. Following Miller, call the former approach genetic and the latter pragmatic.

In the 20th-century, reductionist theories of meaning were eclipsed by causal theories of reference. We can apply this distinction to them, too.

Let’s take the usual cartoon Kripke version as our specimen causal theory.2  Our use of the word “water” refers to H2O because of the conditions under which the term was introduced.3 The cartoonish version is that someone in the past held aloft water in their cupped hands and said, “Lo, we shall call this ‘water’.” That baptism connected the word “water” to that kind of stuff, and that kind of stuff turns out to be H2O. So the word referred to H2O because of the circumstances under which the term was introduced. Actual etymology is always more complicated, but (according to the account) has roughly the same structure as the cartoon story. Our terms refer because we are in a tradition of use that goes back to the terms’ origins, such that the origins connect the terms with specific natural kinds. This is obviously genetic.

Richard Boyd allies himself with causal theories of reference, but his account is importantly different. The causal structure that matters isn’t how a term was introduced. Instead

a term refers to a kind just in case there exist causal mechanisms whose tendency is to bring it about, over time, that what is predicated of the term will be approximately true of the kind.4

Here the idea is that the community makes terms refer because they are on a trajectory to make the terms fit the world. Our ancestors could use the term “water” to refer to water because their linguistic community  had a set of practices which made “water” talk match the water stuff. Their linguistic community was also an epistemic community with first proto-scientific and then scientific methods. Our language, in turn, refers because of the future epistemic progress of the our community. This is pragmatic, in Miller’s sense.

So we can make the genetic/pragmatic distinction among causal theories of reference, just as we could for empiricist theories of meaning. The lingering question is whether it’s useful to call forward-looking accounts “pragmatic”. The word carries a lot of other connotations.5


  1. “Professor James on Philosophical Method”,The Philosophical Review, Vol. 8, No. 2 (Mar., 1899)
  2. This is often called the Kripke-Putnam Theory of Reference, but Putnam’s account is different. I won’t dig into it here, but Putnam’s considered opinion incorporates both backward and forward looking factors.
  3. Identifying water as H2O is deeply problematic, but I’m being loose here.
  4. Here paraphrasing just slightly from Boyd’s “Realism, Approximate Truth, and Philosophical Method”, originally published in Minnesota Studies vol 14 and reprinted in Papineau’s Oxford Readings in the Philosophy of Science.
  5. Putnam is an avowed pragmatist, and Boyd is in lots of ways very Putnamian. But I suspect some people who adopt such outlooks are more motivated by the Hegelian and Marxist traditions than by the pragmatist tradition so-called.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.