I'm attracted to a view of well-being with (roughly) the following structure: x is intrinsically good for y if and only if y believes that x is intrinsically good for y (under the right conditions, such as a coherent belief set). In other words, there is nothing more to something's being good for x than for x to believe that it is good for her. The closest cousin of this view is a desire-satisfaction view, which holds (again, very roughly, and leaving out all sorts of bells and whistles) that x is intrinsically good for y if and only if y desires x (under the right conditions, such as full-information, say). I think the belief formulation can plausibly respond to several worries that plague desire accounts (such as the so-called "paradox of desire" and the problem of welfare-irrelevant desires, etc.). But the view to which I'm attracted (call it, for lack of anything better, "belief-ism") faces an immediate and substantial hurdle: semantic circularity. It looks as though the analysis of "intrinsically good" or "prudentially valuable", etc., is given in terms of the concept it purports to be analyzing. Below the fold, I want to try out one possible response to this worry.
Strictly speaking, the way I stated the circularity problem above is not quite right. After all, on a belief-ist analysis, something is prudentially valuable if and only if it is believed to be so by the person for whom it is prudentially valuable. Hence the analysis of "prudential value" is not given in terms of "prudential value," but rather in terms of beliefs about prudential value. But this doesn't clear up the problem. We need to know what makes a belief about prudential value a belief about prudential value. Plausibly, this will be given by the semantic analysis of prudential value. Hence the circle reappears: the analysis of prudential value is given in terms of beliefs about prudential value, which are analyzed in terms of prudential value, which are analyzed in terms of beliefs about prudential value, etc., etc. David Brink writes: "It is true that, on this view, we analyze X, not in terms of X, but in terms of beliefs about X. But if we accept the not unreasonable assumption that any story about what makes a belief a belief about X must ultimately advert to X, then it appears that this sort of analysis is ultimately circular."
One possibility is to accept a form of primitivism. What it means to be a belief about prudential value is just primitive. (Sharon Street tries out a response like this in the normative domain in a nice essay "Constructivism about Reasons," which I recommend.) This might work, but it is unsatisfying. Let's call this Plan B. Furthermore, one surely can't give an analysis of "prudential value" in terms of some belief-independent or objective property. This would seem to defeat belief-ism, insofar as such a move would appear to imply that x is good for y if and only if x possesses the relevant belief-independent or objective property.
So what to do? I've been thinking about this problem for a long time, and only recently did I realize that the answer was staring me straight in the face. The basic idea is this. Belief-ism is a theory about the truth conditions of claims about well-being. What makes it true that x is prudentially valuable for y is that y believes that x is prudentially valuable for y. So why, then, do we believe that the semantic analysis of "prudentially valuable" must be given in terms of what is believed to be prudentially valuable? Simple: because our best theory of truth tells us that in order for a sentence to be true, the terms must successfully refer–very roughly, the meaning of the sentence must "correspond" to the world. Any other semantic analysis, and you're not going to end up with truth conditions that are compatible with belief-ism. Solution: reject the "correspondence" theory of truth for talk of prudential value (I put that in quotes because many theories beyond the simple correspondence theory will accept that truth conditions are determined by semantics) in favor of a "coherence" theory of truth for talk of prudential value. Such a view yields belief-ism: roughly speaking, “x is prudentially valuable for y" is true if and only if “x is prudentially valuable for y" is part of y's coherent set of beliefs about prudential value (or, assuming that y's belief set is not coherent, y's set of beliefs about prudential value rendered coherent). (I give a fuller specification of a coherence theory of truth in the moral domain here.) Furthermore, it does not require that the semantics of "prudential value" or "well-being" or "intrinsically good" have anything to do at all with the truth conditions of these claims.
Hence, my solution runs as follows: give any non-circular semantic analysis you want of "prudential value". I prefer a semantic analysis that refers to non-natural properties, say "property q". So when I believe that cherry pie is intrinsically good for me, I believe that cherry pie possesses property q. But the existence of non-existence or presence or non-presence of property q has nothing at all to do with whether cherry pie is intrinsically good for me, i.e., whether "cherry pie is intrinsically good for me" is true. The truth of these sentences is determined instead by their being coherently believed.
Anyway, that's my approach. I'm suspicious that what I'm arguing commits me to p and not-p somewhere down the line, but I'm not sure that I see where or how it does so yet. Also it would be helpful to be schooled those who are better versed in the philosophy of language; I'm sure I've violated a bunch of standard conventions.
Dale,
that’s interesting. Few points.
1. I’m not sure what it would be for something to be intrinsically good for someone. Does ‘intrinsically’ qualify something being good for the person in some way? What’s the contrast class? Instrumentally good for me? What is that?
2. I’m not sure there is a problem. Or, if there were it would be too a problem much more generally. There is a similar debate in response-dependent properties generally.
So think of views of the type:
X is red iff certain kinds of judges judge X to be red in certain circumstances.
There was an objection to these accounts that they are circular but it’s not clear that they are. You could think that only the intension of the concept of redness features on the right hand side. This intension could be for instance whatever inferencial dispositions the competent judges need to have in order to be making redness judgments. But, you could think that their judgments determine the extension via the biconditional. The account would be circular if the extension of the term was already assumed or given on the left side but this needn’t be the case. This is the story Crispin Wright gives anyway.
It seems like you could tell a similar story about goodness for. You might say that to have that concept requires certain inferential dispositions for instance from judging that something is good for you to desiring those things. You might then say that whenever competent people use the concept with this conceptual role, their judgments fix the extension of the term. In this way, your account of the extension of the term does not assume what you use to give the account. In this way the account would not be vicuously circular.
I guess the point is that you needn’t give the account as an account of the concept, you can give it as well as an account of the extension of the term. You can still say that it is a part of the concept that it has an extension that is fixed in this way. I also don’t think you need coherence theory of truth here – any sort of minimalism will do just as well.
3. The account is obviously false.
Hi Jussi –
Obvious falsity aside, I think you’re right that a conceptual role semantics might help me out here. But I do want to frame my view as a competitor with the desiderative account. But if one of the crucial inferential dispositions involves desiring what one judges to be good, it is not clear that prudential value judgments have the explanatory force I’d like them to have. One of the ways I want to motivate my view is by denying that people must desire what they judge to be good–so anyway, I’m just insisting here, I think, on a rejection of judgment internalism. But then I’m not sure if there’s any other alternative account of the proper inferential dispositions that could be filled in. Bottom line: I recognize that that’s a possibility, but I feel like it’s in tension with my view. But I could perhaps be convinced otherwise.
Whoops. Re: intrinsic v instrumental. By “intrinsic” I don’t mean “supervenient on intrinsic properties” but rather “non-instrumental,” or “valuable as an end”. So x is intrinsically good for me if it is valuable for me as an end. Another way to say this, I think, is that x improves my well-being. X is prudentially valuable for me, etc., etc. I get the feeling I haven’t really addressed your questions.
Dale,
everyone will have to give an account of what it takes to be using the concept of being good for rather than some other concept (say the concept of being blue). Otherwise, there is no point in ascribing the concept to people at all rather than some other concept. In addition, it has to take something to use the given concept correctly – that is supposed to be constitutive of being a concept in the first place. That there is some connection to desires is just one proposal. There are others that would be connected to some sort of Aristotelian notion of flourishing. But, I take it that you want to deny these options. You do have to say something though about what makes this the concept it is. Otherwise, not even the coherentism story about truth conditions will save you.
ah. You mean finally valuable. You want to give this as an account of well-being? So if I believe that something improves my well-being it so does?
Also, how about things being good for lawn-mowers, cars, and machine-guns? These things don’t have beliefs about what is good for them. So, you will need another account for these. Why not use that account for people too? What evidence is there that being good for is ambiguous in this way?
Hi Dale,
You write:
Some questions:
(1) Don’t you have your x’s and y’s all mixed up? I take it that you meant to say: “y is prudentially valuable for x” is true if and only if “y is prudentially valuable for x” is part of x’s coherent set of beliefs about prudential value (or, assuming that x’s belief set is not coherent, x’s set of beliefs about prudential value rendered coherent).
(2) How do we differentiate x’s beliefs about prudential value from x’s other beliefs (i.e., those that are not about prudential value)? Why is the truth of the proposition “y is prudentially valuable for x” a function of its coherence with some subset of x’s beliefs and not all of x’s beliefs?
(3) People’s beliefs change over time. Hence, x’s coherent set of beliefs about prudential value could change over time. So what do you want to say if, at t1, “y is prudentially valuable for x at t1” is part of x’s coherent set of beliefs about prudential value but, at t2, “y is prudentially valuable for x at t1” is not part of x’s coherent set of beliefs about prudential value?
(3) Suppose that on x’s coherent set of beliefs about prudential value it would, at t10, be intrinsically bad for x to not exist at t10. Suppose that x irrevocably ceases to exist at t9. Does your theory imply that something can be intrinsically bad for a x at a time at which x does not exist? Isn’t this absurd?
Dale,
Interesting. I too am inclined to think you can separate questions about the concept of WB from a substantive theory of WB (cf. Darwall’s approach in his book).
Substantively: must you have the concept of WB to *have* WB? Babies, cats, Yanomami…
Can you believe x is good for you yet have no desire for it, take no pleasure or satisfaction in it…? Perhaps I believe a hair-shirt would be good for me…
Offhand, beliefism seems like it may sever the connection between WB and pleasure, on the one hand, and success re. the things you *care* about (insofar as you don’t conceive of it as a benefit to you), on the other. I wonder if subjectivism, stripped of affect and conation, can retain its appeal.
Dan
Jussi –
Re: Concepts. You’re right that I need an account of the concept. I thought I gave one in terms of a non-natural property, property “q”–whatever that is. Is there some problem with this suggestion? Maybe I’m not seeing it.
Re: Lawn-mowers. I’m not sure I see the force of the objection. There might be some pressure to treat all “good for” locutions in the same way, but there is substantial pressure against it. Any account of goodness for a lawn-mower is unlikely to be accepted as part of an account of what is good for a person, or something with mental states. One might appeal to “flourishing” or something like that, but without starting a first-order intuition battle, I just think such accounts are wrongheaded.
Doug –
1. Dang! You’re right. I’ll go back and switch that.
2. Two questions here. How do we differentiate? By the semantics. Say, for instance, that the semantic content of prudential goodness is given in terms of property q. Then beliefs about prudential goodness are just those beliefs about property q. Why not all beliefs? This is getting into the coherence theory I favor, but I would want to allow access to all other beliefs that are relevant in terms of the coherence of the evaluative beliefs. For instance, the beliefs “sour things are bad for me” and “lemons are bad for me” might require “lemons are sour” to explain their coherence. But your proposal here could be interpreted in two ways: 1. we group all of our beliefs together into one system and subscribe to a coherence theory of truth across the board. I want to reject this. My view is a form of pluralism about truth. Other domains will have different truth predicates, such as the domain of scientific theory. 2. We group all our beliefs together into one system and evaluate only the evaluative beliefs in terms of their coherence with the entire system. In principle I’d be open to that. One caveat, though: I regard “coherence” as defined in terms of warrantability. But I’m skeptical that non-evaluative claims warrant evaluative claims. No good arguments, just skeptical.
(3) and (4): If I’m reading you correctly, these are standard problems for a desiderative account, as well. This is just to note that I’m not sure there’s anything specially problematic about belief-ism here, as opposed to a desire view. But it’s possible I’ve misunderstood, of course.
3. This is a large issue. I’d say that the coherence of someone’s belief set is evaluated at a time. So, at t1, it is true that y is prudentially valuable for x. At t2, it’s false. I don’t see this as particularly problematic, especially given that the desiderative view has to say something similar.
4. I’m not sure about the scope of your first sentence. You might be saying that, let’s say at t8, x believes it will be intrinsically bad for x not to exist at t10. Alternatively, you might be saying that given x’s belief set at t8, were he to live to t10, he would have had, at t10, the belief that it is bad for x not to exist at t10. I’m not sure I understand the force of the second, so I’ll assume the first. The answer is that insofar as x-at-t8 believes that it is bad for x to be nonexistent at t10, it is bad for x to be nonexistent at t10. I don’t see this as at all problematic or absurd. The same is true of various posthumous desires or posthumous value judgments. Suppose I believe it would be good for me to be recognized as a master Hammond Organ player, but only after I die. Suppose I am so recognized. I maintain this makes me better-off. So I guess I’m not seeing why it’s absurd.
It is true that, on this view, we analyze X, not in terms of X, but in terms of beliefs about X. But if we accept the not unreasonable assumption that any story about what makes a belief a belief about X must ultimately advert to X, then it appears that this sort of analysis is ultimately circular
Hi Dale,
I wonder about the Brink quote. Suppose I modify your analysis in the following way. You say this,
x is intrinsically good for y if and only if y believes that x is intrinsically good for y.
Now suppose I alter it as follows,
x is intrinsically good for y if and only if y just says that x is intrinsically good for y.
Is there some semantic problem here? Does one object that y can’t say that x is intrinsically good unless y already knows x is intrinsically good? Of course not. I can say that pea pods are intrinsically good without having the slightest clue, independently of my simplying saying so, whether they are. It is my saying so that is (part of what is) making them intrinsically good, so how could I antecedently know?
But why doesn’t the same go for belief? Must I antecedently know that x is intrinsically valuable prior to believing that it is? No, in fact, that’s not possible on this analysis. My believing it is so is (at least part of) what is making it so. So, it seems to me that I need not be able to advert to anything that is intrinsically valuable prior to my forming my beliefs about it.
Dale,
I don’t think the place-holder non-natural property q does the trick. You still require an account of what it takes to have a competence that is required to be talking about this property rather than of some other non-natural or natural property. What does it take to be talking about q rather than about the weather?
Of course, different things are good for lawn-mowers and people but this doesn’t entail that we mean something different with saying that a thing is good for a law-mower compared to what we mean that it is good for a person. And, even if your account is right about what it is for something to be good for a person, you still need an account what it is for things to be good for non-persons. When you give that account, the question arises why doesn’t it apply to people?
I’m also starting to wonder that on your account claims about goodness for are not made true by what they are about. They are about q but made true by coherent beliefs. I think this will be a problem – there seems to be strong link between truth and reference.
But if we accept the not unreasonable assumption that any story about what makes a belief a belief about X must ultimately advert to X, then it appears that this sort of analysis is ultimately circular
One other quick point. The belief in your analysis is not a belief about intrinsic goodness. In the de re case, it is a belief about x that x is intrinically good. In the de dicto case it is a belief that x is intrinsically good. In neither case is the belief about intrinsic goodness.
Whew. There’s a lot here. This is all very helpful! Hopefully I can say something coherent.
Dan –
These are precisely the substantive points where my view is vulnerable. But with regard to the first one–must you believe about well-being to have welfare–I’m not sure that the desiderative approach is *that much* better-off. Depending on what you mean by desire, many things that we think might have well-being don’t have desires. Second, my view is counterfactual. One needn’t actually have any range of beliefs for them to be present in one’s belief set made coherent. It’s possible that you need some “seed” beliefs, but that’s not clear. I’m musing here, but it seems to me that necessarily, were I to possess a coherent belief set about welfare I would have beliefs about welfare–i.e., I would have the concept of welfare. Furthermore, I favor a full-information, full-experience requirement. This might help with babies because as babies start to experience various lives, they will come to form beliefs about the good, and will eventually end up with a coherent belief. We probably don’t have much to say about what they would look like, but in principle they could vary among babies–it’s probably the case that the extent to which I view x as being good rather than z depends on various genetic facts about me, at least to some extent. I’m not at all sure if this works, I’m just thinking out loud, really. I’d have to think more about it. This move surely won’t help with cats, I expect. This is an interesting pressure that I’m not fully prepared to resolve.
Re: appeal of subjectivism. I think *some* subjectivisms derive appeal from the connection between welfare and conation/affect. But not all. I regard the chief virtue of subjectivism as expressing the authority of an agent over her good. This authority can get expressed in different ways, but it seems to me that someone who desires to x, but who is firmly committed, given a coherent conception of the good, to the worthlessness of x for her is not benefited by x. I think the hairshirt example is like this. I think the authority of the agent over her welfare goes deeper than mere desire or conation: it goes to her conception of the good, conceived as her judgments about well-being. (Maybe “deeper” begs the question here, but I do think I can stand back and evaluate my desires and affective responses from another evaluative perspective, that of my beliefs about the good. I just think this alternative perspective should be the authoritative one.)
Mike –
I don’t think there’s a problem for your restatement, but I do think there’s an important disanalogy. When I just say that x is intrinsically good, that is different than asserting that x is intrinsically good. But assertion has the same semantic problem: if I assert that x is prudentially valuable for me, what am I asserting? Presumably, the analysis, on a straightforward interpretation, must appeal to the assertion that x is prudentially valuable for me. But this appears to be circular. The issue isn’t knowledge. Rather, it’s: what makes a belief a belief about x, where x is “prudential value”. There has to be some semantic analysis. But if the semantic analysis appeals to “prudential value”, the analysis is circular.
You’re right that the belief is not about intrinsic goodness. But I’m not sure this helps. I believe that x is prudentially valuable for me. What, then, do I believe about x? On the straightforward analysis, I would have to believe that I believe that x is prudentially valuable for me. But now we’re caught in the circle again. I feel like maybe I’m missing the upshot. Have I responded to the worry?
Jussi –
You write:
“You still require an account of what it takes to have a competence that is required to be talking about this property rather than of some other non-natural or natural property. What does it take to be talking about q rather than about the weather?”
Can you say a little more about the trap I’m about to walk into? I’m not sure what it is that you’re asking for. Why can’t I give the same sort of analysis that someone might give about talk of the weather? It’s possible I’m just being bullheaded here.
You also write:
“I’m also starting to wonder that on your account claims about goodness for are not made true by what they are about. They are about q but made true by coherent beliefs. I think this will be a problem – there seems to be strong link between truth and reference.”
In fact I deny this claim explicitly. On my view there is no link between truth and reference when it comes to the domain of well-being. Judgments about prudential value are about q and made true by coherent beliefs–this is the key move in retaining the belief-ism while avoiding the circularity problem. So in essence I deny that there is any such strong link.
Dale,
I don’t think that the desire theory is subject to the same worries that I expressed with regard to belief-ism. Here’s one simple version of the desire theory:
DT: y is intrinsically bad for S at t if and only if y is the state of S’s desiring, at t, that p when in fact p is false.
DT does not imply, as apparently belief-ism does, that the proposition “y is intrinsically bad for S at t” can be true at one time and false at another. I tend to think that such a proposition is either true or false, not true at one time and false at another.
DT does not imply, as apparently belief-ism does, that y can be intrinsically bad for S at a time at which S does not exist. Suppose that, at t2, S desires that he will be recognized as a master Hammond Organ player after his death (say, t10). Suppose that upon S’s death, at t9, you ensure that he is not so recognized by taking credit for his work. Your actions are, then, make it true that he was in a harmed condition at t2 while alive. At t2, he desired that he would, at t10, be recognized as a master Hammond Organ player. It was and is false that he would, at t10, be recognized as a master Hammond Organ player. Thus, according to DT, he was in a harmed condition while he was alive, not while he was dead and non-existent. So, unlike belief-ism, DT does not absurdly imply that y can be intrinsically bad for S at time that S does not exist. I know of no one who thinks that S can be in a state that is intrinsically bad for S at a time in which S does not exist. How can the non-existent be in any state, let alone one that’s intrinsically bad for him or her?
Hi Dale,
One more thing:
Suppose that y is some very painful sensation (say, the sensation of being dipped in boiling oil) and suppose that S has never experienced y before. Suppose that S believes, as a result of someone’s false testimony, that y is intrinsically good for S. Suppose that this belief is part of S’s coherent set of beliefs about prudential value. Belief-ism entails that “y is prudentially valuable for S” is true. Isn’t this absurd?
Hi Dale,
One more thing:
Suppose that y is some very painful sensation (say, the sensation of being dipped in boiling oil) and suppose that S has never experienced y before. Suppose that S believes, as a result of someone’s false testimony, that y is intrinsically good for S. Suppose that this belief is part of S’s coherent set of beliefs about prudential value. Belief-ism entails that “y is prudentially valuable for S” is true. Isn’t this absurd?
Dale, you say first this,
The issue isn’t knowledge. Rather, it’s: what makes a belief a belief about x, where x is “prudential value”. There has to be some semantic analysis. But if the semantic analysis appeals to “prudential value”, the analysis is circular.
But then you say this,
You’re right that the belief is not about intrinsic goodness. But I’m not sure this helps.
If the issue arises because what you believe is about intrinsic goodness, then it is a non-issue here. What you believe is not about intrinsic goodness at all.
But set that aside and set aside your proposed analysis (for the moment). All of these contexts–asserting p, believing p, saying p, etc.–are referentially opaque and non-truth-functional. So, the truth-conditions for (P) have nothing to do with there actually obtaining the state of affairs that x is intrinsically good (or even with such a state of affairs being so much as possible).
P. S believes that x is intrinsically good.
Obviously, (P) can be true even if (1) is true,
1. It is impossible x is intrinsically good.
What makes (P) true has nothing to do with whether (1) is true or false. So to specify the truth-conditions for the proposition expressed by ‘S believes that x is intrinsically good’ you do not need to specify the truth-conditions for the proposition expressed by ‘x is intrinsically good’. The semantics of ‘x is intrinsically good’ are entirely unrelated to the semantics of ‘x is intrinsically good’. How could a semantic issue arise here, since the truth-conditions are fully unrelated?
The semantics of ‘x is intrinsically good’ are entirely unrelated to the semantics of ‘x is intrinsically good’.
That is,
The semantics of ‘x is intrinsically good’ are entirely unrelated to the semantics of ‘S believes that x is intrinsically good’.
Doug –
Ah. I did misunderstand. Ok. There are a number of things to say here. One runs as follows. I can make the same claim about posthumous desires. y coherently believes that x-at-t10 is good for her. So x-at-t10 is good for her. No implication of that view holds that it is good for her at t10, when she doesn’t exist. Why am I committed to non-existent things being in states?
Re: true, then false. Why is this problematic? Let’s say I coherently believe that being a philosopher is good for me during the time t1-t10. So, during t1-t10, being a philosopher is good for me: the times of my life t1-t10 achieve at least one good thing, viz., being a philosopher. At t11 I switch. Now it is false that being a philosopher is good for me. So if I am a philosopher from t11-t20, the state of being a philosopher is no longer good, as it was from t1-t10. I don’t see why this is a problem.
Furthermore, the same can hold for time-indexed goods. Let’s say I coherently believe that it is good for me that I x-at-t11, but only from t1-t10. So if I x-at-t11, this will be good for me from t1-t10 (“I x-at-t11” is true timelessly). It will not be good for me that I x-at-t11 at t11 onwards.
Now that I’m re-reading your comment, perhaps you meant to say this: I might have a belief of the following form: x is intrinsically good for me at t, where this is interpreted not to mean that x-at-t is intrinsically good for me, but rather to mean that the only time at which I am benefited by x is t. At all other times x is neutral for me. Is that the idea? Alternatively, I might, at t11 come to believe something like this: “x was NEVER good for me”, despite believing, at t1-t10, that it was. This might sound like a problematic example of true-false switching.
The answer to this is a little more complex. In any event, I’m not going to argue for any of the following claims, but I’m merely going to tell you what I believe about the proper specification of a coherence theory. I think what I want to say here is that that sort of a sentence is not susceptible to the coherence theory of truth. Given my analysis, whether x was good for S at time t depends on S’s set of beliefs at time t. (More precisely: they depend on S’s beliefs about property q at time t.) Hence if my analysis is true, the question of whether x is/was good for S at some particular time is not a question of property q at t, but rather of what did S believe about property q at t? But this is a perfectly descriptive claim, which permits of a semantic theory of truth. This gets a little more complex when it comes to inferences: “x is good for me now” seems to imply “x is good for me”: how can it be so if “good for me now” has a different truth predicate than “good for me”? I try to discuss this in the article–the idea is that a disjunctive truth predicate operates in a “metasystem”. Anyway, I try to argue that making this move can get me everything we want in terms of inference, etc., etc. (Furthermore, the inference from x is good for me now” to “x is good for me” is truth-preserving on my view. “x is good for me now” implies that x is part of my coherent set of beliefs about property q. Hence, given the biconditional in the original post, “x is good for me” is true.)
Re: boiling oil. My view does not have this implication. As I noted above, I include a full-info/full-experience constraint on coherent belief sets. (One way to put this more precisely is that the warrantedness of implication relations in the coherent belief set are judged against a background of full descriptive information. There are other ways to specify it, but this one is parsimonious.) I excluded this bit from the post so as not to clutter the thing up. Given this, the supposition that y is intrinsically good would be incoherent for this person, because, given a true belief about the experience she will encounter, she will not believe it is intrinsically good.
Mike –
D’oh! You’re right. The first sentence shouldn’t say “about prudential value” it should say “about x’s prudential value for me”, or something like that.
I’m going to think about your other comments while I walk home and get back to you later this evening. Thanks everyone so far for a nice discussion.
Dale,
Your view has the following consequence. When you believe that cherry pie has property q, and it does not, your belief might still be true. This seems like a bad feature of your view. It is not quite the p and not-p that you feared, but it’s awfully close.
You can escape contradiction by denying all the usual disquotation rules and schemas. Is that your intention? It seems very costly to me.
On belief-ism, S coherently believes that x-at-t10 is bad for S at t10 implies that x-at-t10 is bad for S at t10. But S does not exist at t10 per my stipulation. How can something be bad for S at a time that S does not exist?
Belief-ism implies that “Being a philosopher at t11 is bad for S at t11” is true if S coherently believes-at-t2 that “Being a philosopher at t11 is bad for S at t11.” And Belief-ism implies that “Being a philosopher at t11 is bad for S at t11” is false if S does not coherently believe-at-t4 that “Being a philosopher at t11 is bad for S at t11.” How can “Being a philosopher at t11 is bad for S at t11” be true at t2 and false at t4?
I don’t know what “I x-at-t11” means.
Now that I’m re-reading your comment, perhaps you meant to say this: I might have a belief of the following form: x is intrinsically good for me at t, where this is interpreted not to mean that x-at-t is intrinsically good for me, but rather to mean that the only time at which I am benefited by x is t. At all other times x is neutral for me. Is that the idea?
By “x is intrinsically good for S at t,” I mean that “t is a time at which x is intrinsically good for S.” I don’t know why you put an ‘only’ in there. Nor do I know why you think that “t is a time at which x is intrinsically good for S” would imply that at all other times x would be neither intrinsically good or bad for S.
So now that I read further through your comment, it seems that you didn’t formulate belief-ism very precisely in your post. Am I now to understand that belief-ism is the view that x is intrinsically good for S at t iff S coherently believes at t that x is intrinsically good for S at t?
Is it possible to have false beliefs when one’s beliefs are fully informed and coherent? Could S coherently believe (under the right conditions) that the experience of being dipped in boiling oil at t would be intrinsically good for S at t? If ‘yes’, then your view does have this absurd implication. If ‘no’, then it seems to me that your bi-conditional is true but your mistaken about which side has explanatory priority.
One other quick thing:
Doug’s second question is, I think, a difficult one, because it is much harder than you might think to work out exactly what it takes for a sentence to be about T, for some topic T. This is why Mike is insisting that the belief that something is intrinsically good for you is not a belief about intrinsic goodness. I would have thought that intuitively it is indeed about intrinsic goodness, but quite possibly the notion of a sentence being about a topic is vague or ambiguous. (See David Lewis’s ”Statements Partly about Observation” for a characteristically intense introduction; Lloyd Humberstone has a couple of rather more difficult papers on the same topic if you get really interested.)
Hi Dale,
Interesting proposal. I’m wondering why a person’s own idealized beliefs (as opposed to those of others) would play this special role in determining his or her well-being. Perhaps I’ve just missed it in the comments, but can you give an intuitive “pitch” for this sort of view?
Jamie –
I take it that this is what a coherence theory of truth entails. Insofar as I accept such a theory for moral claims, I accept what it entails. I’m not sure why I should be particularly worried about the T-sentences–giving them up seems no more unintuitive than the rejection of a semantic conception of truth.
Doug –
“On belief-ism, S coherently believes that x-at-t10 is bad for S at t10 implies that x-at-t10 is bad for S at t10. But S does not exist at t10 per my stipulation. How can something be bad for S at a time that S does not exist?”
My view does not imply this. As I said, sentences of the form “x is good for S at t” are not evaluated given the coherence theory, but rather an alternative theory. However, I think I’ve traced the source of our misunderstanding, and it was my mistaken answer to your first question number 3. I read your question differently than it was intended. Reading it again now, I should have said just what I did here, rather than confusing the issue by addressing a different question, viz., about whether “x is intrinsically good for S” can be true for S at different times. Which, of course, it can. I goofed.
“Belief-ism implies that “Being a philosopher at t11 is bad for S at t11” is true if S coherently believes-at-t2 that “Being a philosopher at t11 is bad for S at t11.” And Belief-ism implies that “Being a philosopher at t11 is bad for S at t11” is false if S does not coherently believe-at-t4 that “Being a philosopher at t11 is bad for S at t11.” How can “Being a philosopher at t11 is bad for S at t11″ be true at t2 and false at t4?”
Same response.
“I don’t know what “I x-at-t11″ means.”
I meant “x” to be a verb, like “am a philosopher”. Sorry for the confusion.
“By “x is intrinsically good for S at t,” I mean that “t is a time at which x is intrinsically good for S.” I don’t know why you put an ‘only’ in there. Nor do I know why you think that “t is a time at which x is intrinsically good for S” would imply that at all other times x would be neither intrinsically good or bad for S.”
When I say “I might have a belief of the following form” I mean just that: I might have this sort of a belief. The question then was, if I were to have this sort of a belief, would that fall under the phenomenon you’re describing? The phrase “you meant to say this” was the problem here, I think; what I mean to say was: “is this the kind of thing you mean”?
“So now that I read further through your comment, it seems that you didn’t formulate belief-ism very precisely in your post. Am I now to understand that belief-ism is the view that x is intrinsically good for S at t iff S coherently believes at t that x is intrinsically good for S at t?”
As I said twice in the original post, the description of the view was very rough, meant simply to get the circularity worry and potential solution on the table. But this is not the right statement of the view. Rather, it is more like: “x is intrinsically good for S at t iff S coherently believes at t that x is intrinsically good for S”. Remember that beliefs about what benefits someone at a time are not evaluated on the coherence theory.
“Is it possible to have false beliefs when one’s beliefs are fully informed and coherent?”
Presumably you mean, “is it possible to have false beliefs of the form “x is intrinsically good for me” when one’s beliefs of the form “x is intrinsically good for me” are fully informed and coherent?” Given the coherence theory, the answer must be “no”.
“Could S coherently believe (under the right conditions) that the experience of being dipped in boiling oil at t would be intrinsically good for S at t? If ‘yes’, then your view does have this absurd implication.”
I’m unsure of which absurd implication you’re referring to. Is it metaphysically possible that someone will believe that being dipped into boiling oil would be intrinsically good for her? Yes. Under these conditions would doing so be good for a person? Yes. Is this absurd? Maybe, but I don’t see how this isn’t any more problematic than on a desiderative view or any other strong subjectivist view for that matter. (Grass counters, lint collectors, etc., etc., etc.) Actually, I think the coherentist view has about as plausible a story to tell about these cases as is possible for a subjectivist view to tell. (Weird or oddball activities are much more plausibly good for me if they are supported by a coherent conception of the good and warranted by other beliefs about value, than when it’s some bare desire.) In any event, you haven’t done anything to motivate it’s being specially absurd. Your first statement of the objection, however, appeared to be making use of a different absurd consequence, viz., that on the basis of false testimony could someone come to coherently believe that swimming in boiling oil is good. And here the answer is clearly “no”. I thought that was the objection. Here you deleted the reference to false testimony, so I suppose you accept that solution to the previous objection?
Jamie (and Mike) –
I was not intending to provide some rigorous theory about the semantics of “x is intrinsically valuable for me” here. In fact, you might say, my view is in a way the anti-semantic theory; you can fill in whatever semantics you think is plausible, and belief-ism remains intact. So I certainly don’t have to, and wouldn’t want to, disagree with Lewis. I don’t have to shy away from the difficulty. But that’s for those people who are firmly committed to some semantics of moral claims or other to figure out. I’m interested in only a non-circular placeholder.
Furthermore, I don’t see why we don’t have a circularity problem here. It is true that the truth conditions of “I believe ‘x'” are different than the truth conditions of “x”. But, as Doug notes, we still need to know how to demarcate different sorts of beliefs. How do we know that the belief about what is intrinsically valuable for me is a belief about what is intrinsically valuable for me? Semantically. And if the semantics refers to the beliefs, the semantics is circular.
Steve –
Thanks for the question. The funny thing is that I didn’t really try to give an intuitive pitch for the view! Ack! All I’m really interested in is hashing out the circularity worry. Anyway, how should I pitch the view? I’m really not interested in pitching it to people who don’t at least feel the pull of the desire-satisfaction type view, or other subjectivist views. My view is supposed to be one side of an intramural disagreement between subjectivists. I think there’s something deeply interesting about a person who believes x is good, but does not desire (for its own sake) x. For agents like this, it seems to me intuitive to say that it’s really the belief that governs the good, rather than desire. That’s not much of a pitch, I know, but anyway, that’s the most basic guiding idea behind my view.
Jamie –
I think I might have interpreted your previous point uncharitably. So, for instance, you might think that the disquotational conventions, etc., etc., are extremely plausible in themselves, and it would be a real drag philosophically if we couldn’t accept them. Then given that my theory can’t, we should reject the belief-ism that gets us into this mess in the first place.
This is a powerful objection, I think. But I wonder if it doesn’t put the cart before the horse. Seems to me that the various disquotational schemas, etc., should not be divorced from the plausibility of a roughly semantic theory of truth in a given domain. So, basically my response to your sort of worry is to try to make belief-ism look as plausible as possible in the domain of well-being. This will include defending the claim that there really is nothing more to something’s being good for me than that I desire it. If I can do that (big “if”), then it seems to me that the intuitive force of the schemas you mention might be blunted. Alternatively, you could agree with all that and still insist on some sort of primitivism. But then we’d have to weigh the relative merits of primitivism versus the view I’ve set out. And it’s at this point where I think my intuitions give out. But you’re right to say that there’s a serious worry to be found here.
Dale,
about this:
“”You still require an account of what it takes to have a competence that is required to be talking about this property rather than of some other non-natural or natural property. What does it take to be talking about q rather than about the weather?”
Can you say a little more about the trap I’m about to walk into? I’m not sure what it is that you’re asking for. Why can’t I give the same sort of analysis that someone might give about talk of the weather? It’s possible I’m just being bullheaded here.”
This is not a trap. I’m trying to be helpful here because I think there is a view in the same ballpark that avoids the problems you are strugling with. Indeed the analysis should be of the same sort. What does it take to be talking about weather?
Well, presumably there has to be some link to behaviour and observations – that you say ‘it is raining’ when you see rain falling down and when you pick up the umbrella on your way out.
This is just the sort of story we would want for the good for talk – some dispositions about the things from which one infers that it is good for and the things one infers and does as a result of judging that it is good for one. This needn’t be much but it should be substantial enough to distinguish this concept from others.
Once you had that story, beliefism would come easily. You could say that it is constitutive of the meaning of the term that nothing else is required for having the property of being good for than that people who have this inferential dispositions judge that the object is good for you. This would give a nice contrast to other properties where more is needed for having the relevant property than that competent judges judge the object to have the property.
Isn’t it odd to say that your belief that X is q is about whether X has q or not but it can still be true even if X does not have q? I don’t understand that really.
Also, there will be problems about pluralism about truth if you think that truth is coherence in ethical domain and something else elsewhere. One of them is mixed inferences. Consider inferences of the form:
P1. Eating vegetables is good for me.
P2. I’m eating vegetables.
C. Therefore, I’m doing something that is good for me.
This is a valid inference. By valid, we would want to me that it preserves *a property*, namely, truth of the statements from premises to the conclusion. But, if truth of P1 is coherence and truth of P2 some other semantic notion, then there is no property which is preserved. This means that you’d have to give an alternative account of validity or be a coherentist universally.
Hi Dale,
Yes, you accurately traced the source of my confusion: your misleading answer to my question (3). Thanks for clearing that up.
By the way, when you stated your position in the form of “…iff…” and said that it was rough, I assumed that some part of it was vague, not that it was false. Thus I assumed that you took the bi-conditional to be true but that it was rough only in that you didn’t fully specify what the right conditions for coherently believing were. I didn’t think that you were offering a bi-conditional that is false no matter how the right conditions are specified.
So, now, your response to the suggestion that belief-ism has a number of counter-intuitive implications has been that the desire theory has these counter-intuitive implications as well. I don’t understand how this constitutes much of response. Am I missing something? Do you think that an adequate response to the objection that a view has a certain counter-intuitive implication is made merely by out that another view has the same counter-intuitive implication?
In any case, I don’t see why I need to motivate the position that belief-ism is “specially absurd.” Why isn’t it enough to show that it is absurd, although no more absurd than other false and absurd views? Perhaps, though, you have a better response but just want to focus solely on the circularity objection for now. If so, fair enough.
Dale,
Hmm. Well, I did say that violating the schemas is a cost, not that it’s a refutation. But I do think it’s a very high cost.
I think what I said might be slightly misleading. It’s not that we look at the schemas and think, “Wow, those are really intuitive.” It’s that the instances seem like obvious tautologies. So losing the particular instances is what is so costly.
Suppose George believes that eating cherry pie has nonnatural property q. This belief coheres perfectly with George’s other beliefs, but as a matter of fact eating cherry pie doesn’t have q. I tell you, “George believes that eating cherry pie has q,” and you say, “Yes, and what he believes is true, but eating cherry pie doesn’t have q.” To my mind, this shows you must mean something other than ‘true’. But, as you say, maybe you can marshal enough other support for the coherence view that you can bear the cost.
I don’t understand the comment (8:25) addressed to me (and Mike). I mean, I don’t see that it has anything to do with anything that I said. I only mentioned Lewis in connection with the difficulty of spelling out what it is for a sentence to be about a given topic, and I only pointed it out because I thought there had been some confusion in the earlier comments. I agree with you that there is a circularity problem in explaining or defining is intrinsically good in terms of is believed to be intrinsically good, but that’s a separate question. (There is also a very good Humberstone paper about that one, as it happens.)
Hi all –
Thanks for the great discussion so far, and for putting up with my tardiness on the uptake!
Anyway, here goes:
Jussi –
I see now the question you’re asking a little better. I don’t think I have a well worked-out answer for you. But offhand, you appear to be construing the conceptual role that any term plays rather weakly, in terms, for instance, of what most people would do when they use term “x”–I take it that this is the force of the weather example. If so, I think I could accept your claim that there is some conceptual role involving desire or motivation. Generally, when people use “good for me” they have some disposition to seek it. But then your claim that: “You could say that it is constitutive of the meaning of the term that nothing else is required for having the property of being good for than that people who have this inferential dispositions judge that the object is good for you” sounds like a stronger connection than I would want to make between the semantics of the term and the conceptual role. After all, on my view, something could still be good for someone merely on the grounds that they judge it to be so, without possessing the relevant inferential disposition. This, of course, doesn’t respond to what I now take to be your previous worry: “but what does it take to be competent at using ‘good for'”? I think this is an interesting challenge that I clearly haven’t thought enough about. So perhaps there is a solution here in what you say, but I might have to think more about it.
On the other point: what you are illustrating is simply what it means to divorce the truth conditions of claims from their semantics–I took this to be an essential element of the story. Sentences of this kind are true not just in case the world is as they say it is, but whether they cohere with each other. Anyway, that’s the basic idea. Is there something more to your not understanding the proposal?
Re: mixed inferences. I try to give a more in-depth response in the “Coherence Theory” article, but a quickie response is this. I don’t have to give you an alternative definition of validity. This argument is valid in the standard, say, descriptive/scientific domain. It just so happens that the first premise and conclusion are false. But that doesn’t mean that the argument isn’t valid. Soundness is a little more complex, obviously, but I certainly don’t have to stray from our ordinary understanding to get a good understanding of validity.
Next up, Doug and Jamie. My laptop’s out of juice!
Thanks again.
Dale,
sorry I don’t get the replies. You say that:
“After all, on my view, something could still be good for someone merely on the grounds that they judge it to be so, without possessing the relevant inferential disposition”
So, someone can lack the relevant inferential dispositions, thus lack the concept of being good for, but they can still make judgments about goodness for. I’m not sure I understand.
I don’t get the reply to the mixed inferences either (even though I’m starting to see the things you are denying about truth and reference). I agree that it isn’t relevant whether the premises are true. The point is if they were true, this property would be preserved in the conclusion. But, this explanation doesn’t seem to work as the premises couldn’t even share a property of truth as truth would mean different things for these sentences.
Back in black. Here goes:
Doug –
When I said the view was supposed to have a rough specification, I meant “sufficient, without unnecessary complication, to understand the reply to the circularity worry”. I don’t see that I’ve failed to do this. Of course, I wouldn’t accept the simple specification of it at the end of the day, bells and whistles need to be added–bells and whistles of many types, including counterfactual conditions, time-relativity, etc., etc. I just wanted to leave that out to get the circularity response on the table.
I sure hope I don’t think that’s an adequate response! If I did, I would have bigger problems than simply believing a silly view about well-being. (Would any reasonable person believe that such a response is adequate to the question you’re asking?) I would love to have a wide-ranging discussion of the various merits of strong subjectivism versus weak subjectivism or objectivism in the domain of well-being. It strikes me that such a discussion would be very worthwhile. But I wasn’t interpreting your objection to be an objection to strong subjectivism generally (which it now appears that it is), but rather to belief-ism in particular–the way you stated the first version of the objection (the false testimony business) seemed to indicate that. My response: desire theories have this problem, too, is meant to indicate just that: there is no problem particular to belief-ism here, but rather a wider problem with subjectivism in general. Furthermore, it seemed to me that the truth or falsity of subjectivism in well-being is neither here nor there with regard to whether belief-ism can solve the circularity worry or even whether belief-ism is the best version of strong subjectivism. So I guess I had this in the back of my head when I was interpreting your question, perhaps mistakenly.
So, finally, to at least try to respond to the objection! I regard the resonance or non-alienation constraints on well-being as simply dispositive. I don’t see how it is at all plausible to say that some thing x can be good for me unless I take some sort of positive attitude toward x. This is an argument for subjectivism, but not strong subjectivism–there could still be a further constraint on well-being, such that “x” must be objectively valuable for itself, or pleasurable, or whatever. (Incidentally, I’d also like to argue that belief-ism provides the best foundation for a weak subjectivism, too, even if coherent belief is a necessary and not sufficient condition.) So why strong subjectivism? I guess this might just be pounding the table, but I don’t see how someone could desire everything they get, or firmly, strongly, and coherently believe that everything they achieve is worthwhile, and not live what would be an ideal life for them. Anyway this is just to state intuitions, but that’s what they are for what it’s worth. But I’d certainly welcome a new thread on subj v obj.
For my purposes, however, I’m interested only in making the comparative argument: belief-ism provides a better understanding of subjectivism in well-being than desire-type views. If I can convince desire theorists of that, I’m happy.
Jamie –
I don’t have much to add–I think you’ve properly diagnosed an problem, and indicated my preferred strategy to solve it. The “tautologicality” of the individual T-sentences, it seems to me piggybacks on the crucial presumption that meaning is a crucial element in truth. My burden is to argue that this presumption should be reexamined, by showing that the resulting view is plausible when it comes to the domain of well-being, at least for those who are generally inclined toward some form of subjectivism. Whether the overall argument is successful is another matter, but I’m heartened to know that this is the task ahead!
Re: Lewis. That’s what I get for responding after 11pm. Now that I re-read our discussion, I’m not sure I understand what *I* was trying to say. Anyway, your suggestion now seems quite helpful to me, and worth looking into in more detail–I guess that’s all I’d want to say about it now.
Jussi –
Oh. So you do want to say that, even in cases of weather, all persons, in order to possess the concept of, say, “rain” must be disposed to say “it is raining” when believing it is raining, walking out the door with the umbrella, etc.? If that’s right, then I was misunderstanding your original suggestion. I’m thinking out loud now, but could I say that in order to possess the concept of good for, you need only possess the disposition to say “x is good for me” when you believe that x possesses property q? Offhand that strikes me as parallel with the weather case. But don’t quote me on that! (Or on any of this, really.)
Dale,
Is it possible for S to take intrinsic attitudinal pain at t in the fact that she is currently experiencing the sensation of being dipped in boiling oil while also coherently believing at t that this sensation is intrinsically good for her?
If so, then belief-ism implies that intrinsic attitudinal pain can be intrinsically good for S. And if, as I assume, intrinsic attitudinal hedonism is a form of subjectivism, then the objection that I’m raising applies to belief-ism but not to all forms of subjectivism. It seems, after all, that intrinsic attitudinal pain and intrinsic attitudinal pleasure are attitudes and, thus, that on IAH x can be intrinsically good for S only if S takes some a positive attitude toward x.
Jussi –
Mixed inferences:
“I agree that it isn’t relevant whether the premises are true. The point is if they were true, this property would be preserved in the conclusion. But, this explanation doesn’t seem to work as the premises couldn’t even share a property of truth as truth would mean different things for these sentences.”
Sure it does! The sentence: “x is good for me” means “x has property q”. There is no property q, so, if interpreted according to a semantic theory of truth, the sentence is false. But false sentences can still be used in evaluating the validity of arguments. If truth preservation is semantic truth preservation, we get the following outcome:
1. eating veg is good for me. (F) (interpreted to mean “eating veg has property q.”)
2. i am eating veg. (T)
3. I am doing something that is good for me. (F) (interpreted to mean “i am doing something that displays property q.”)
Validity is preserved. Of course soundness isn’t. So soundness is more complex, but I certainly don’t see why I have any problem with validity.
Hi Doug –
I think we might be working with different accounts of subjectivism. I’d class IAH as a form of objectivism. Why? Because the state of me taking pleasure in something is good for me regardless of I have any positive attitude toward that state. (In other words, IAH doesn’t say that x is good for me if I take IAP in x, but rather that the state of my taking IAP is good.) You might formulate a version of subjectivism in IAH terms, as you note. For instance, you might say that x is good for me if and only if I take intrinsic attitudinal pleasure in it. But that, I take it, would be a different view and would also, it seems to me, be vulnerable to your counterexample. In other words, I can take intrinsic attitudinal pleasure in being dipped in boiling oil. (At least, I’m pretty sure I can, given Feldman’s analysis.) Maybe you’d find this less implausible.
There are issues that come up on my distinction. (For instance, I’m committed to rejecting the Supervenience Principle of intrinsic value, as I noted in a response above.) I like my way of dividing up the pie for a number of reasons: it’s more consonant with the desiderative tradition overall, and because it adequately captures the distinction between those views that give agents authority over their own good, and those views that do not. IAH, I take it, does not–whatever you desire, think, believe, or have any pro-attitude towards, intrinsic attitudinal pleasure is good for you. Anyway, this is probably just taxonomical.
Question: assume that I take intrinsic attitudinal pleasure in being dipped in boiling oil, would you say that I would be better-off if so dipped? I can think of different ways of going here.
Hi Dale,
Question: assume that I take intrinsic attitudinal pleasure in being dipped in boiling oil, would you say that I would be better-off if so dipped? I’m not sure. But I find the idea that the sensation of being dipped in boiling oil is good for S if S takes IAP in the fact that she’s having this sensation a lot more plausible than the idea that the sensation of being dipped in boiling oil is good for S if S coherently believes that it is good for her.
Am I correct in thinking that the view that I called DT above is a form of objectivism? It holds that the state of S’s desiring that p when p is true is good for S whether or not S has any positive attitude toward this state.
Dale,
now I’m getting confused. So, claims about being good-for are assessed by two standards of truth depending on what suits your purpose? Coherence makes the first premise true, but it is still false because there is no property q.
Yeah. On my view that’s objectivism. You might think that’s implausible, but here’s an example to motivate the way I classify it.
I positively despise the song “Ball and Chain” by Social Distortion. I am strongly averse to hearing it ever again. But DT as you state it is committed to claiming that the state in which I desire to hear it, and I hear it, is good for me. But not only do I never desire to hear the song again, I certainly don’t desire to desire to hear it and to hear it (call this state HBC). I am strongly averse to HBC; I would consider a desire to hear this song a remarkable degradation in my aesthetic sensibilities, and thus regard HBC with horror. But DT must hold that HBC is good for me, now, even if I am strongly averse to it. That to me sounds like objectivism.
It’s interesting that you should mention this, because I’m working on this very topic right now: why DT can’t really capture what we want in a desire view. This would also be fun to discuss!
Jussi –
I hope that’s not what I mean. It seems to me that my solution to the validity question is quite straightforward. “Argument x is valid” strikes me as a descriptive claim: does this collection of sentences, arranged in this way, possess the necessary modal property required for validity? Hence because that is a descriptive question, it gets evaluated on the truth predicate appropriate to descriptive systems. On this theory, the sentences come out as I illustrated. And, as it turns out, that collection of sentences, arranged in that way, does have that property, hence the argument is valid. (Even if some sentences are false.)
Soundness is different, because here you’re not evaluating the properties of an argument itself, but also the truth of individual sentences, which requires access to multiple systems and truth predicates at once. Again, this is more complex, but I don’t see that my solution to the validity issue is somehow ad hoc. Rather, it’s fully motivated given the sort of claim “argument x is valid” is.
Dale,
The idea that the truth property preserved in validity is not the one possessed by the premises in a sound argument, that makes me a little seasick.
Suppose an argument is sound. Can we be sure that its conclusion is true? Ordinarily we can, since the property possessed by the premise(s) (in virtue of which the argument is sound) is the property that we know will be transmitted across valid arguments to the conclusion. But you think those are different properties.
Do you think that sound arguments all have true conclusions?
Dale: I may be being dense here, but I don’t understand your idea of severing truth from semantic value. According to your view, all of the following three claims can be true at once, right? And they could all be truly asserted?
1) “Le gateau est bon pour moi” means that cake has non-natural property q.
2) Le gateau est bon pour moi.
3) ~ Cake has non natural property q.
Let me put my worry this way: if my French teacher were to assert all of these things at once, it wouldn’t be his French I doubted, it would be his English. He doesn’t seem to know what “means” means.
Jamie –
Hmm. You’re noting an interesting problem here. Let me just follow my intuitions and see if they lead me somewhere nasty.
Soundness is difficult because no one system (descriptive, evaluative, etc.) can adequately judge the truth of sentences in different systems. So what you need is a sort of “metasystem” from which you can talk about all the other systems. The metasystem also has a new truth-ish predicate: Truth*. A sentence is true* if and only if it is true when evaluated on the truth predicated in its home system (you could, in principle, give a disjunctive def. of truth*). Sound mixed inferences preserve truth*. However, all valid inferences will preserve truth* as well as truth. This is not analytically true, of course, but I’d want to treat this as an adequacy condition on any theory of truth in any domain. In other words, whatever truth predicate you insist on, that property must be preserved by valid inferences. The coherence theory captures this by its definition of what is required for a system to be coherent, and how that’s evaluated (viz., it’s evaluated by the system of descriptive sentences–just as “‘x’ is valid” is a descriptive claim, so is “system of beliefs y is coherent”–along with that system’s truth predicate and account of validity. By the way, this is also why the definition of “coherence” isn’t circular.)
Anyway, I *think* that’s what I want to say. It’s been a few years now since I thought about all of the technical nooks and crannies of this view. Somehow I feel as though I’ve already got one foot off the plank.
Wouldn’t full information include the fact that no x is ‘q’ and thus preclude any rational agent from ever both being fully informed and believing that any x is good for them?
Hi Simon –
Man, that looks bad. How to respond? My initial reaction is to say that how bad this looks is a function of how ingrained the T-sentences are in our ordinary thought, a helpful point that Jamie brought up earlier.
So yes, they could all be asserted as true. But because
4. “Le gateau est bon pour moi” is true if and only if cake has non-natural property q.
is false, these sentences are not inconsistent.
But that’s not to say it doesn’t look bad. Maybe an immanent/transcendent distinction might help here. Sentence 2 is really in a different idiom–one that permits of an entirely different truth predicate than the others. And so it looks strange when you state things this way, but that’s only because you’re mixing two sentences that are transcendent of that discourse with one that is immanent. We do this sometimes when referring to fictional contexts. You’ll have to forgive the hedging–these are hard issues that I’m unsure of how to work through on the fly.
Simon and David –
I’m just now realizing that my answer to Simon is going to be pertinent to my answer to David. But I’m also now realizing that my answer to Simon causes serious problems in other areas. Maybe I’ll try to respond again a bit later today.
Simon (and David) –
Here’s why I’m thinking my previous response is inadequate. In response to a question Doug asked earlier, I said that sentences are demarcated by their semantic content. So sentences about, say, property q are sentences about what is good for me, that then get evaluated by the coherence theory. But here’s a sentence about property q if any is:
3) ~ Cake has non natural property q.
So that turns out to be an evaluative claim on my analysis (it looks like), and hence incoherence with (2), and hence either (2) or (3) has to go. So that seems to be a decent answer to your original problem, but then it seems like there’s another problem. How is it that I say, which I’ve been wanting to say, that, metaphysically speaking, there is no non-natural property q? How could I say that, if sentences about property q are evaluative sentences, and not descriptive? (I think the best I could do is to say that “‘cake has non-natural property q’ is false if evaluated according to the semantic theory.” That sentence is true. But I’m not sure that’s satisfactory, although it might be as good as I can do in this case.) This is a much harder problem that I’m not sure I see a ready answer to. This is pertinent to David’s question also because if sentences about property q are evaluative not descriptive, then they won’t be part of the information one has in evaluating whether one’s system is coherent. But this is just to say that I’m not sure that I’ve solved the demarcation problem as easily as I thought I had.
Anyway, it seems as though the comments have slowed, so I’d just like to say thanks to everyone who has lent me their brain! This has been quite helpful for me.
Dale,
I apologize for coming to this discussion late. But I have an objection to the account. You say toward the end,
“So when I believe that cherry pie is intrinsically good for me, I believe that cherry pie possesses property q.”
And your thesis is that “’x is intrinsically good for S’ is true if and only if ‘x is intrinsically good for S’ is part of S’s coherent set of beliefs about intrinsic goodness.” [edited, not unfairly I hope].
Now what happens when you believe your thesis about intrinsic goodness? That is, what do we say about you, Dale Dorsey? What you believe is that cherry pie is intrinsically good for you just in case that belief is part of your coherent set of beliefs about intrinsic goodness. So your own “property q” is one which raises the circularity problem.
In other words, there is no way to both (i) eliminate the circularity problem and (ii) have true beliefs about intrinsic goodness. Eliminating the circularity requires the believer to have a false (because non-circular) analysis of intrinsic goodness. And you, the theorist of intrinsic goodness, can’t be in this position.
Hi Heath –
There’s a crucial distinction here, and it’s related to my denial of the T-sentences.
“What you believe is that cherry pie is intrinsically good for you just in case that belief is part of your coherent set of beliefs about intrinsic goodness.”
This is false. What I believe is:
“’cherry pie is intrinsically good for me’ is true if and only if ‘cherry pie is intrinsically good for me’ is part of my coherent set of beliefs about intrinsic goodness.”
And that’s the difference. Because I deny “x is intrinsically good for S” iff x is intrinsically good for S, you cannot infer from the fact that I believe that “‘cherry pie is intrinsically good for me’ is true if and only if ‘cherry pie is intrinsically good for me’ is part of my coherent set of beliefs about intrinsic goodness” that I also believe: “cherry pie is good for me iff I coherently believe cherry pie is intrinsically good for me.”
That might sound weird, but it’s a natural outgrowth of this form of alethic pluralism, I think.
Because I deny “x is intrinsically good for S” iff x is intrinsically good for S, you cannot infer from the fact that I believe that “‘cherry pie is intrinsically good for me’ is true if and only if ‘cherry pie is intrinsically good for me’ is part of my coherent set of beliefs about intrinsic goodness” that I also believe: “cherry pie is good for me iff I coherently believe cherry pie is intrinsically good for me.”
This seems to risk making the view empty. Does your view tell us when some x is good for us or merely when the sentence ‘x is good for me’ is true? If your view is only a view about the truth of sentences, but the truth of sentence ‘Z’ doesn’t guarantee Z, then what have we really learned? I know now that my coherent belief that cherry pie is good for me guarantees that ‘cherry pie is good for me’ is true, but I don’t know whether or not cherry pie is good for me. And, surely, this latter is what we want to know.
Hi David –
Thanks for your comment. Anyway, here’s a shot in the dark:
I feel like there’s a metalanguage/object language distinction that’s being forgotten here (possibly by me). I took Heath’s first sentence to be expressing a claim of the metalanguage. But his claim is not quite right–my substitution is the proper metalanguage claim, along with the denial of the T-sentences, also in the metalanguage. The theory of truth and the semantic claims are also couched in the metalanguage–it tells me about when sentences in the object language are true.
At the object level, I don’t see why I can’t know that cherry pie is good for me. It’s true, it’s believed, and that belief is justified (at least, let’s assume).
I should add that if we’re in the object language, I can certainly hold that x is good if and only if I coherently believe x is good. On my view this is interpreted to be an explanatory claim, not a semantic one–what explains why x has property q is that I coherently believe it does.
“what explains why x has property q is that I coherently believe it does.”
I’m confused; I thought that you denied that x has property q but asserted that ‘x is good for me’ is still true iff I coherently believe it. But now it sounds as though ‘x has property q’ is true iff I coherently believe that x is good for me. Is that right? I think I may just be out of my depths here, my apologies.
Hi David –
I tried to post a response, but for some reason it didn’t take. (And if you’re out of your depths, that’s a serious problem. I was proceeding on the assumption that at least one of us knew what we were talking about!) In any event, the short version is that I think you’re pointing to the fact that I need to make crystal clear when someone is using “x has property q” in the prudential object language, and when “x is using property q” in the metaphysical object language, or “descriptive” object language. It’s not exactly clear how I’m going to do this, but I think the questions you’re raising (and Simon and Heath are raising) are clues that I need to do it.
Thanks again!
Hi Dale,
On a different line, how do you think your position with respect to coherence is distinct from, or related to (maybe even builds upon), Sumner’s stance on authenticity (see e.g. *Welfare, Happiness, and Ethics* pp. 160-175)?