Joey and I talked a lot yesterday about the revelatory nature of philosophical reasoning. Sometimes, connections just come to you—and these are the most profound and interesting. Joey had this happen to him in Paradoxes yesterday, and little did I think the same thing would happen to me exactly twelve hours later. I’ll also preface this blog post by saying it’s one of those ones that Joey thinks most people don’t like—I’m going to work you through my own thought process in depth, instead of quickly reaching a conclusion that can be applied in many disciplines. However, I still think this is interesting.
Throughout my discussions with Joey (and Brian, when he visited) about pragmatism, I have tried to defend three different arguments in favor of a correspondence theory of truth. I have come to believe that the first two are dead ends, but I always had hope for the third. After what I’ve realized today, I think that the third may be correct.
The first defense was an argument suggested by Brian. It goes roughly as follows (and yes, the gravity example was lifted directly from his formulation of the argument):
1A. The pragmatist says that when I say “the law of gravity is true in Newtonian physics”, I mean that it is useful to believe in the law of gravity.
1B. It is useful to believe in the law of gravity.
1C. The only reason that it is useful to believe in the law of gravity is because there is a fact of the matter about the fact that I will fall down if I jump off a building.
1D. Therefore, a commitment to believing in things because they are useful requires a further commitment to the fact that there must be some truth-functional basis for their usefulness.
I believe that this argument doesn’t work, because the conclusion, 1D, doesn’t follow. The point of theories of truth is to explain what I mean by asserting something. A believer of the correspondence theory of truth says that I mean that that thing is true. A pragmatist says that I mean that it is useful to make this assertion under certain conditions. The fact that there necessarily exists some objective world which provides a mapping against which all things are useful or not does not deny the pragmatist’s claim. In “Consequences of Pragmatism,” Rorty explains that “For the pragmatist, true sentences are not true because they correspond to reality, and so there is no need to worry what sort of reality, if any, a given sentence corresponds to—no need to worry about what “makes” it true”. But argument 1 is precisely a claim about what makes pragmatic language “true”. I think that this just begs the question. Argument 1 assumes the pragmatist is committing to defending the explanations for what I mean in asserting a proposition. By hypothesis, the pragmatist refuses this commitment. I don’t think we can force him to make this commitment on pain of circularity.
The second defense was quite different. My argument was just that the world best justified by a Rortian pragmatism is not a useful world, or at least one that I would want to live in. I still think that his suggested world (especially as articulated in Part 5 of “Consequences of Pragmatism”, which I understand is quite different from his suggestions in other papers) would be a terribly depressing and nihilistic world in which to live. Obviously, though, the fact that I wouldn’t want to live there isn’t an argument against pragmatism without further argument. In any case, I think this style of objection to the usefulness of a pragmatic world doesn’t work for two reasons.
First, I don’t think a pragmatist has to defend a “post-Philosophical (with a capital P)” world. I think we can just conceive of pragmatism as a specific philosophical ideology that bridges the gap between philosophy of language, epistemology, and metaphysics. It doesn’t necessarily have to make a further normative claim on how our world should be—it can just be a claim on how we should conceive of the world we live in (and even then, that might be more than it is claiming).
Second, this is more of an issue with Rorty’s specific account of usefulness than a denial of pragmatism in general. A different interpretation of usefulness (for instance, as directed towards the preservation of the human species, to provide an example that I don’t subscribe to but which at least serves as a contrast) would produce a different final world, and that world might be one I’d prefer. I’ll get back to this a bit later.
The third defense was the one I always felt was the most promising. My argument was pretty simple: Given two theories, we should choose the one with the most explanatory powers. I felt that pragmatism had less explanatory power than the correspondence theory of truth. This is because you can explain pragmatism in terms of the correspondence theory of truth. You could just say that it is true that in all actual cases of language, we use words pragmatically. However, you cannot explain the correspondence theory of truth in terms of pragmatism. You can’t say that it’s useful to use words in a truth-functional way, because pragmatism denies that there is any essence to “using words in a truth-functional manner”.
Jump ahead two months. I was spurred by some questions put to me by another ex-debater (Julian Switala) to start reading about dream-skepticism, and I came across “Can I Know That I Am Not Dreaming” by David and Jean Beer Blumenfeld. This article spells out Descartes’ argument for why we might always be dreaming, and therefore we cannot know to be true any belief which is based on our experience. This moves into the discussion of Descartes’ evil demon.
At the very end of the article, they described Michael Slote’s answer to the proposition that an evil demon determines all of our experiences. Michael Slote argues that some hypotheses are “inquiry-limiting” if an agent’s accepting them “ensures the impossibility of his coming to have rationally justified or warranted belief (consistent with his other beliefs) in more and more true explanations of various aspects of or facts about the phenomena in question.”
Compare two propositions: (A) An external world is ordered in a way so as to create our experiences, or (B) an evil demon controls what we encounter and perceive as an external world. The evil demon proposition is necessarily inquiry-limiting, because all of our sense experience is supplied by a deceiver, so all explanations based in our experience are suspect. In contrast, the external world hypothesis is not obviously inquiry-limiting.
Why should we prefer non-inquiry-limiting hypotheses? Slote’s somewhat surprising answer is that we should reject inquiry-limiting hypotheses because they conflict with the end of science.
“[Science] seeks to give explanations of events, processes, etc., but also to give, as far as possible, explanations of all the various aspects of the very things it posits in its explanations. For it is a goal of scientific enterprise to gain deeper and deeper and more and more explanations of whatever things there are in the world, wherever possible.”
Science demands that we follow the Principle of Unlimited Inquiry:
“(a) that it is scientifically unreasonable for someone to accept what (he sees or has reason to believe) is for him at that time an inquiry-limiting explanation of a certain phenomenon, other things being equal; and (b) that there is reason for such a person to reject such an explanation in favour of an acceptable non-inquiry-limiting explanation of the phenomenon in question, if he can find one.”
I think that this is basically the same style of argument I made against pragmatism. I was really excited to discover this passage and see the similarities between Slote’s argument and my own. So, based on this reading, let me try to formalize the argument I made a bit:
3A. Pragmatism and the correspondence theory of truth are explanatory theories—they purport to explain what we mean by asserting something.
3B. We have no access to an epistemic means to discriminate between two explanatory theories about the meanings of assertions.
3C. There is a possible practical way to discriminate between them: Given two theories, we should not prefer one that is an inquiry-limiting hypothesis.
3D. Pragmatism is an inquiry-limiting hypothesis:
It can’t explain truth-functional sentences (my original example)
It rejects the idea that truth-functionality has content as a characteristic of sentences.
It rejects entire parts of philosophy as “disinteresting”.
It forces us to attribute the way things are entirely to contingency.
3D’: The correspondence theory of truth is not obviously inquiry-limiting.
3E: Therefore, we should reject pragmatism.
3D is not the critical premise. I think that most pragmatists view this as a virtue of pragmatism, rather than a problem with it. Rather, 3C is what most pragmatists would try to deny. I find Slote’s defense of 3C to be distinctly unsatisfying. It seems to beg the question (to consider a reductio, why should the function of science define our choice in theory, rather than say, the function of truth-seeking?). Moreover, my faith in science is not absolute. (You can probably expect a blog post sometime by Joey about this…)
With that being said, I think that there is a legitimate way to defend 3C. Two tracks suggest themselves. I think that both work. First, ultimately, what are pragmatism and the correspondence theory of truth? They are ideas that try to explain what we mean with words. It’s obvious to me that there’s some sort of content to all statements that I consciously make within the syntactic rules of language. (Even if you don’t agree with Searle, at least bear with me that syntactically correct statements contain semantic content, even if they aren’t semantically coherent.) A theory of meaning should be able to explain this fact to me. One which arbitrarily imposes limits on what I can mean by my words would fail this test. The pragmatist might object that he avoids this concern, because he presents an error theory for why my intuitions about really meaning to talk about truth statements are false. But I feel that the very notion of salvaging pragmatism by presenting an error theory means that pragmatism has already lost. How could we have an error truth without truth?
Second, I was inspired in this by Rorty’s own professed love for 1984. I found one of the most disgusting features of this futuristic world the idea that newspeak could eliminate patterns of thought from the population by eliminating them from the language. I think that the (perhaps pragmatic) need to avoid any conclusion like this is what speaks against inquiry-limiting hypotheses. I think this goes to the heart of my own fears about pragmatism. What is to stop us from going past the point of no return? Maybe pragmatics dictates that we get rid of whole parts of philosophy, maybe we will think it’s useful to get rid of parts of language. When are we ever in the proper epistemic position to be confident in our judgments of what is useful, if we take usefulness to be the underlying principle behind all propositions?
The great irony of this is that Michael Slote seems to be a pragmatist himself. He subscribes to the view that defines “’p is epistemically reasonable in believing q’ as ‘if p were a rational being, and if his concerns were purely intellectual, it would be reasonable (i.e., a good thing) for him to believe q.’” I doubt that he would welcome my reinterpretation of his views.