Truth Subjectivism and Giving Implausible Views their Due

Hi world! Some blog housekeeping beneath the Green Raven for anyone who cares. But for now I am just going to jump right in - a book I think is very good on the whole has done a thing I don't like, so I am going to complain about it. Chase Wrenn has an introductory book on Truth (link beneath the raven - I actually really like and recommend it!) that at one point discusses "relativism". This is defined (on pg.13) as the view that

"[W]hat makes a claim true for someone is that they believe it, and what makes it false for someone is that they disbelieve it." 

Now for reasons that will become apparent I don't think this is best described as "relativism" and I will refer to it instead by "subjectivism", but it does fall within a family of views I have discussed on the blog before which have something of the flavour. I think it is fair to say that it is an implausible view of truth, but I think this fact lets philosophers get away with attacks on it that do not grant its proponents the conceptual resources such a person has available to them. I am not here to defend subjectivism, but I do wish to show that some of the criticisms Wrenn offers - and he does so in a way that I think is a fair representation of many people's objections to subjectivism - just don't hit the mark.

Consider the first objection Wrenn offers:
"[I]f all it takes for a claim to be true for a person is for her to believe it, then whatever she believes is true for her. She can't be wrong. But people are mistaken about things all the time."
After which some examples of typically mistaken beliefs are given. But is it really the case that a subjectivist cannot make sense of people being mistaken, of some beliefs being false? Well, consider the situation from the point of view of a subjectivist: they have their present beliefs, and when they look out at others they note that they have lots of beliefs which contrast with their current view. Those beliefs others hold are hence false from the point of view of our subjectivist, since (per Wrenn's definition) to be false for them just is to be disbelieved by them, and others hold beliefs which they disbelieve. What is more, when they remember their past self, they doubtless remember believing some things they now do not believe and believe the negation of, so they also know of themselves that they believed false things. They can even simulate for themselves a change in perspective (why shouldn't they be able to do hypothetical reasoning, after all), so they can understand a perspective from which their present beliefs turn out to be false, though it is not one they would endorse.

In short, it seems they can look around and make perfect sense of a great many beliefs being false, including their own beliefs, and can understand the idea that stuff they believe now may turn out to be false. And this they get from the fairly simple operation of noting actual or hypothetical contradictions between their beliefs and other beliefs they know to be out there. She has no trouble making sense of the idea that she can be wrong -- she remembers changing her mind and can imagine doing it again -- nor that people are mistaken about things all the time -- people disagree with her!

What they can't do, of course, is say that their own current beliefs, the perspective they now operate from, is false or contains falsehoods. But that is just to say: they are self-consistent. Which given the reaction philosophers typically have when I say I am a dialetheist sounds like it should count in their favour!

Later on Wrenn says "We cannot make things true just by believing them". Now in context it might be that Wrenn just means this to be a rephrasing of the central argument above. But you could view this as a different sort of argument - namely, subjectivism seems to posit an implausible quasi-causal or metaphysical link between what I believe and how the world is. A subjectivist seems to be committed to saying something like: I can bring it about that it is it true that there are green ravens by coming to believe there are green ravens. And hence if I grant that "if S is true then s" (where "S" names the sentence s) it seems that I can bring it about there are green ravens by believing there are green ravens. There thus appears to be a chain: I believe there are green ravens > it is true that there are green ravens > there are green ravens, which begins with my mental act and ends with exotic birds! Since (alas) it does not seem I can do that, isn't the subjectivist in trouble?

Again, consider this from within the point of view of the subjectivist. They will have beliefs about the casual structure of the world. It almost certainly rules out such mental-acts-to-ravens links. It's the presumption that they do not believe this which allows the objection to purportedly be dialectically effective, after all. Whereas if you are talking to a believer in The Secret then you're just going to get enthusiastic nods at this point. (But what about the subjectivist who does believe in the secret you say? Well their problem isn't truth theory, say I.) So the subjectivist is almost certainly not going to agree to the claim that willing it can make it true for them - that would contradict known truths for them!

What to say instead? After all, you might worry that even if subjectivists happen not to believe such bizarre causal claims they are committed to them by their position, or they are entailed by what they do believe, or some such. Well, there are two very standard moves here which the subjectivist should be allowed to make. One is to stress a difference between constitution vs causation (of course this is no easy thing!) and insist that while they think a belief's being true is constituted by their believing it they are not saying their belief causes it to be true, nor that its being true causes the state of affairs it describes to obtain.  The other is to make some typical deflationist moves (e.g.) and argue that they can explain norms of belief (such as don't believe there are green ravens if there are no such!) without having to appeal to concepts of truth or falsity, and hence consistently with their idiosyncratic view of truth.

At the end of all this philosophising, then, they would be able to coherently say from their own perspective - it is not true that my believing something causes it to be so, and I have good reason not to believe in the absurdities which you worried I might (think I could) will into being. That, at least, would seem to me to greatly defuse this worry.

But subjectivists never really get to develop their ideas in enough detail (well, with one exception, I am very much inspired in my reasoning in this post by reading the essays on truth by Wiredu in here, which at least bear some affinity to the subjectivist's position). And this gets at my real complaint here - so rather than respond to the other objections Wrenn launches I am going to just say what I think the underlying problem for these and many other objections to subjectivism are.

Because people take subjectivism to be such a wildly implausible view they make no effort to see the world from a subjectivist's point of view. Even calling the view "relativism" betrays that - because it imagines a point of view beyond any particular person's perspective, surveys all of them at once, and says "truth for the individual is just what is believed by them, and so truth per se is relative to each individual's beliefs and all beliefs are true for their believers". From this last conjunct objections are drawn out. But from a committed subjectivist's point of view that cannot be right - their point of view is the one from which truth is evaluated,  and the other points of view are not on a par with that but simply errors. The objection imagines a kind of God's-eye-perspective on truth and launches their attack from there, but the kind of person who is attracted to subjectivism (or for that matter relativism) is almost certainly the kind of person who is suspicious of the idea of such a God's eye perspective. Seen from within, these objections simply lose their force, they don't take seriously what the subjectivist is trying to do or say as a philosopher of truth.

Of course one can always make the move against the subjectivist that many philosophers now make against the sceptic. One can say that after all our job is not to find arguments that would persuade the subjectivist, our job is just to explain why they are wrong in an internally coherent manner. We wish not to help the subjectivist see their error but just ensure others do not fall into it. Well, good luck and God speed to those whose philosophical mission is explaining why they are right to their own satisfaction. But if you want philosophy to help you enter into the lifeworld of others quite different from yourself, this cannot be enough.

The Green Raven - "caw caw check out the amazing artist who produced me here also there is no real problem with
Hempel's paradox Janina Hosiasson-Lindenbaum essentially solved it in the 40s cawwwww cawwwwwwwwww"

I realise it has been an uncharacteristically long gap between posts. I usually try and do one a month, but life happened while I was busy making other plans. I am in any case busy writing a follow up to this blog post for my long suffering sponsor Sam Fletcher (sorry!) but in the mean time this blog hasn't been updated in too long and it turns out this is the post that needed a follow up.... and was easier for me to very quickly dash off and get back to all the many tasks keeping me busy at the moment! (Sam isn't the only person owed stuff by me.) Because, due to events around that post I have somehow managed to memetically associate myself with thinking about truth. For this reason I have wound up being asked by many people to recommend them an introductory text on truth - my current faves for the role are this (the subject of today's post, for just a good and moderately general overview), this (for scientists just looking to quickly get up to speed with what philosophers are saying, which is many of those asking me) and this (for people who might need a bit more persuading on the why care about this at all front, so for instance for intro undergrads doing a required course).

One of the things that happened though is that I became very embarrassed about having a blog post appear when I owe so much work to other people. Rather than having that happen again I am just going to make a couple of notes: (a) I write blog posts very quickly, because they are usually on issues that I am mulling over in one way or another quite frequently so the actual thinking required at the point of writing the blog post is largely just phrasing. So please don't take this as evidence I am dedicating a bunch of time to this when I should be doing <whatever I owe you>.  (b) Life is short and then we die, facing who-knows-what torments afterwards. Some even say we'll have to do it all again, though I scarce believe a good God allows such evil. One more post, one less <thing I owe you>, what matters it before such tragedy?


  1. I suppose the main problem I have with relativism or subjectivism about truth is that, as a deflationist, I don't know what it asserts, and I don't quite see what theory allows it to assert something I understand.

    Start with simple deflationism. What would it mean for me to say that "Snow is white" is true for me but not for you. ""Snow is white" is true for me" just says that snow is white. What does ""Snow is white" is not true for you" mean? Does it mean that snow is not white? That would be awkward. (I'm possibly ok with dialethism in some contexts, but not about snow being white.) Does it mean that according to you, snow is not white? That's pretty boring. That just says that not everyone believes the same things. Does it mean that if I understood "snow is white" the way you understand things, I would say "Snow is not white"? Maybe, but if the former is a truly semantic claim, then as I deflationist, I reject the possibility of "understanding it the way you do". If it just means something pragmatic, like "If I understood 'snow is white' in a way that was similar enough to yours to be useful I would say "snow is not white," then I guess we have a thesis, but I'm not sure its more than a thesis about a sort of interesting non-overlap of our ideolects.

    As you know, I think the only alternative to deflationism that isn't just evading the topic is something like a world-as-model semantics that's in the neighborhood of the CToT. So, what does it mean to say that the model picked out by "snow is white" both is and isn't the right model of the world. One obvious way in which this is true is that sometimes our models of the world have a de se component. So "the baseball is breaking to the left" is extremely important to me adjusting my swing and getting a hit. And the model that makes it true has to be centered around me. And its super interesting! And I think de se stuff is less well understood than it should be. But it feels like a bit of a parlor trick to me to call this alethic relativism. Its a reason to reject world-as-model semantics for at least world-as-centered model semantics. But then any theory of truth that wanted to piggy back on this would have to (it seems to me, if it were being honest) strip out the centered part of the semantics to say what correspondence amounts to. If this cant be done, so much the worse for correspondence. So modus tollens, not modus ponens.

    1. Thanks for the reply, Eric! My sense is that a subjectivist should not try and put their position in anything like the way Wrenn does here, because in some sense that will always involve trying to take the God's eye perspective they are trying to reject. Instead, to articulate their position from the inside, they should try and spell out how they intend to make use of truth-ascriptions - when so done I expect this will actually look a lot like some form of deflationism, but in a way that makes it clear that truth claims and reporting on what they believe claims are really doing much the same work for them. So think part Wiredu and part early-Wittgenstein's showing-not-saying distinction. At the least, this is what I one day hope to defend!

    2. Been thinking about this some more. Here's a core commitment that I think a deflationist ought to have.

      (D) Everything about the T-predicate is exhausted by endorsement of some T-up and T-down rules like T("Snow is white)->Snow is white and Snow is white->T("snow is white")

      To be an alethic relativist or subjectivist, I need at least two T-predicates, call them Tme and Tyou.

      I'm obviously going to stick with the old inference rules for Tme. So

      Tme("snow is white)->"snow is white" etc.

      What could be the inference rule for Tyou? I guess it would have to be something like:

      Tyou("Snow is white") & I-were-you (in some sense) -> Snow is white.

      that is Tyou("Snow is white") would imply that [If I were you, Snow would be white]

      maybe this can be made to work. I dunno.

    3. there are " marks that shouldn't be there in one line.
      it should be Tme("snow is white")->snow is white

    4. Aye, but this seems like I just need some means of marking the perspective (model, world... etc - something that marks whether its my belief set or your belief set we're talking from) I am evaluating the claim relative to. But I think standard and deflationist consistent logical machinery will allow one to achieve that, will it not? Or do you think the machinery here is consistent with a deflationist perspective on truth?

    5. I was trying to say "yes, it is consistent". I probably shouldn't have used "snow is white" since that made is sound a bit ridiculous, but maybe others are less silly:

      Tyou("the ball is breaking to the left")-> If I were you, the ball would be breaking to the left.

      Tyou("there's a norm against such and such") -> If I were you, there would be a norm against such and such.

      Falseyou("all people are created equal")->If I were you, no all men would be created equal.

    6. Ah we're more in agreement than I thought!

      But ok I don't think the natural language gloss should be in terms of "If I were you" because that suggests what is happening is one is reasoning from a metaphysically difficult counterfactual, and one which I would not agree to even if I could be assured it makes sense. This because I do not think now, from my present perspective, that the moral equality of persons depends on the fact that I happen to be me; so I would not agree on the counter factual reading.

      The fact, though, that I cannot think of a better natural language gloss on what the subjectivist wants to say is part of why I have not written this paper though!

    7. I don't love it either. But I don't think of it as a "natural language gloss". I think of it as something more fundamental to truth. I want to preserve the idea that the truth predicate is just a logical device. That's what I'm doing here. And I can't think of anything better than what I have. At the end of the day, I'm not sure how you are going to avoid the idea that if T("S") is only Tme("S") then its going to be that S depends on the fact that I am me. At least on pain of not being a deflationist (and on pain of giving up the deflationist commitment that (in twitterease) meaning is fake. That's because when I say Tyou("S"), "S" is going to be in my ideolect. So Tyou("S") *must* combine some facts about me, you, and S as *I* understand it.


Post a Comment

Popular posts from this blog

How I Am A Marxist

On Not Believing In One's Work

Arguments in Philosophy