Okay, circling around on this to maybe say something more constructive now that I’ve thought about it a bit.
Part of the problem is that your central thesis is not very clear on first read, so I had to think about it a bit to really get what your “big idea” or ideas are that motivate this post. I realize you say right up top that you believe a strong version of verificationism is correct, but to me that’s not really getting at the core of what you’re thinking, that’s just something worked out by other people you can point at and say “something in that direction seems right”.
(FWIW, even some of the people who came up with logical positivism and related ideas like verificationism eventually worked themselves into a corner and realized there was no way out and the whole thing fell apart. The arguments for why it doesn’t work eventually get pretty subtle if you really press the issue, and I doubt I could do them justice, so I’ll stay higher level and may not have the time and energy to address every objection that you could bring up, but basically there’s 50+ years of literature trying to make ideas like this work and then finding there were inevitably problems.)
So, as best I can tell, the insight you had driving this was something like “oh, there’s no way to fully ground most statements, thus those statements are meaningless”. I’ll respond based on this assumption.
Now, up to a point you are right. Because there is an epistemological gap between the ontological and the ontic, statements about the ontic are “meaningless” in that they are not fully grounded. This is due to epistemic circularity, a modern formulation of the problem of the criterion. Thus the only statements that can be true in a classical sense are statements about statements, i.e. you can only get “truth” from inside an ontology, and there is no matter of truth to be assessed about statements of any other kind. This, alas, is not so great, because it takes all the power out of our everyday experience of truth.
One possible response is to hold this classical notion of truth fixed and say all those statements that can’t be grounded are false. The alternative is to say we screwed up on the classical notion of truth, and it’s not the category we thought it was.
Which every path you take, they converge to the same place, because if you reject things as not having truth value, now you have to introduce some new concept to talk about what everyone in everyday speech thinks of as truth, and you’ll be forced to go around annoying everyone with your jargon but whatever. The alternative is to accept that the classical notion of truth is malformed and not the category it was thought to be, and to rehabilitate truth in epistemology to match what people actually mean by it.
As I say, in the limit they converge to the same place, but in different language (cf. the situation with moral realism and anti-realism). I’ll speak now from the second perspective because it’s mine, but there’s one from the other side.
So then, if we were wrong that truth is about statements that can be proven true and grounded in reality, then what is truth? I take from your talk of “constrained expectations” you already see the gist of it: truth can be about predicting experiences, as in what is true is that which is part of a web of causality that generates our perceptions. This is messy, of course, because we’re embedded in the world and have to perceive it from inside it, but it gives us a way to make sense of what we really mean when we talk about truth, and to see that classical notions of truth were poor but reasonable first approximations of this, made with the intuitive assumption that there was ever some (classical) truth to know that was of a special kind. And on this view, truth of statements about the ontic are not meaningless; in fact, they are the only kind of statements you can make, because the ontological is ontic, but the ontic is not ontological.
Thus, also, why you have a comment section full of people arguing with you, because of course the natural notion of truth we have is meaningful, and it is only on a particular narrow view of it that it is not, which is right within itself but leaves out much and confusingly repurposes words to mean things they don’t normally mean.
Depends. In a certain vague sense, they are both okay pointers to what I think is the fundamental thing they are about, the two truths doctrine. In another sense, no, because the map and territory metaphor suggests a correspondence theory of truth, whereas ontological and ontic about mental categories and being or existence, respectively, and historically tied to a different approaches to truth, namely those associated with transcendental idealism. And if you don’t take my stance that they’re both different aspects of the same way of understanding reality that are contextualized in different ways and thus both wrong at some limit but in different ways, then there is an ocean of difference between them.
My central thesis is “beliefs that don’t constrain expectation are meaningless”, or more specifically to avoid some of the pitfalls, “beliefs about external reality are meaningless.”
For what it’s worth, I came up with the view first, then someone suggested verificationism fit my views, I saw that it was a form of logical positivism, recalled the Sequence posts on that, and reread them.
>basically there’s 50+ years of literature trying to make ideas like this work and then finding there were inevitably problems
I spent some time looking through critiques of verificationism. They all seemed to be disputing claims I don’t agree with. The analytic/synthetic distinction is irrelevant to my argument, but much of the critiques I found centered on that. I’m not saying that beliefs about mathematics are meaningless, so Godel isn’t relevant. It seems to me that all the problems are related to specific features of logical positivism proposals that I see no reason to accept.
I think people use truth/reality to mean several different things. As I said in the post, I’m ok with some of those uses. But there are definitely uses that appear meaningless and impossible to rescue, from my perspective—yet people will adamantly defend those uses. If by reality you just mean your best map that predicts experiences, that’s fine. But people will absolutely defend a notion of some “objective reality” that “really exists”, and there’s just no meaningful account of that that I can think of. So no, I’m not repurposing words to mean something new—the actual meaning of the word in common usage is incoherent.
I don’t know what your stance on moral realism is, but I’m confident there are places where I could make a post asserting moral anti-realism or error theory and get a bunch of smart people in the comments arguing with me. That doesn’t imply that moral realism is meaningful, or that you need to invent a new concept to represent what people mean when they talk about morals. Incidentally, different moral realists have different accounts of the nature of such beliefs, and I assume the general populace that believes in morals would have different understandings of what that means, if asked.
If you think it’s so obvious that the concept of external reality is coherent, then can you give an account of it? I don’t need a super detailed technical account, but at least some description of what it would mean for something to exist or not, without referencing predictions, just in the abstract.
I gave arguments for that in the post. Instead of responding to those, you insisted there’s some hypothetical value system that makes the concept of truth useful, and didn’t really argue for its coherency at all.
The burden of proof is on people asserting the positive, not entirely on those trying to prove the negative. Pointing out that nobody has given an account of meaning, and 50+ comments later still nobody has done so, is very suggestive that nobody actually has one.
A binary attribute is coherent if it’s possible for the attribute to hold as well as possible for the attribute not to hold. External reality is incoherent, because it never holds and it never doesn’t hold—there’s nothing that it would mean for it to hold or not hold.
there’s some hypothetical value system that makes the concept of truth useful
Theres a value system according to which truth is valuable and that is the one where truth itself is a terminal value.
If you want to argue that usefulness is the only true value, go ahead
nobody has given an account of meaning,
Shared meaning is what allows communication to take place. Therefore, communication taking place is evidence of shared meaning.
Do you think that verificationism is filling a vacuum ? That’s it’s the only theory of meaning anyone ever came up with? There are multiple fields that deal with the subject of meaning. There are multiple theories of meaning, not zero.
A binary attribute is coherent if it’s possible for the attribute to hold as well as possible for the attribute not to hold. External reality is incoherent, because it never holds and it never doesn’t hold—
Again: the existence of external reality can be supported by abductive reasoning: it’s the best explanation for why science works at all.
You’re entitled to put whatever you want in your value system, but if it references incoherent things then you just can’t achieve it. Personally I would much prefer that 1+1=blrgh, but sadly not only does it equal 2, blrgh is meaningless. If I kept insisting that blrgh was meaningful and played a central role in my value system, what would you say? Also, I think truth can be defined without referencing external reality, just observations.
Shared meaning is what allows communication to take place. Therefore, communication taking place is evidence of shared meaning.
Evidence, but not definitive evidence, as I pointed out elsewhere. It’s possible for people to be mistaken about what has meaning, and I argue they are so mistaken when talking about external reality.
Do you think that verificationism is filling a vacuum ? That’s it’s the only theory of meaning anyone ever came up with? There are multiple fields that deal with the subject of meaning.
I’m talking specifically about the meaning of external reality. I looked up critiques of verificationism and didn’t see any accounts of such meaning. If there’s a particular account of the meaning of external reality that you’d like me to look at, let me know.
I responded to the explain argument elsewhere. Regardless, it’s unclear how an incoherent claim can serve as an explanation of anything.
Incoherent != Impossible. It’s not possible for me to get a unicorn, but it’s coherent for me to want one. It’s incoherent for me to want to both have a unicorn and not have a unicorn at the same time. It’s incoherent for me to want the “true map” to have an XML tag saying “this really exists”, because the concept of a “true map” is meaningless, and the map with the tag and the map without yield the same predictions and so neither is more correct.
Its where your sense data come from. You need the idea of an external world to define “sense organs”.
Disagree. If you’re defining it as the source of sense data, that’s roughly equivalent to the definition in The Simple Truth essay I linked to and responded to in OP.
My sense data can be predicted well by my map. Why do I need to define a concept of external reality to make sense of that?
Why? How does realism explain why science works? What is the exact argument, premises, and conclusion?
>There’s a definition of “external reality”: it’s where your sense data come from.
This definition is incoherent, and I addressed it directly in OP. It’s also circular—it assumes your sense data comes from somewhere, which is precisely what I’m claiming is incoherent. Giving a definition that assumes the matter in debate is begging the question.
>There is a purpose served by the posit of an external reality: explaining how science works .
I still don’t know how your proposed explanation works. I think any such explanation can be replaced with an equally valid explanation that doesn’t assume incoherent concepts such as realism. And even if you managed to show that, all you’d be showing is that the concept is useful, which neither implies coherency nor truth.
>There is a meaning of “true model”, based on correspondence.
This again relies on the assumption of realism, which is circular.
I’ve already given a definition of incoherent. A belief is coherent if it constrains expectations. A definition is incoherent if it relies on incoherent beliefs.
Are you going to hold yourself to same standard?
You’re making three claims: one, that realism is coherent; two, that realism explains how science works; three, that one can’t explain science as easily without realism.
My claim is that realism is incoherent, and that whatever argument you have for your second claim, I can find one just as good without using realism.
I only asked you to provide details on the second claim. If you do, I will hold myself to the same standards when arguing against the third. I can’t give a precise argument to explain science without realism now, because I don’t know what you want in an explanation until I’ve seen your argument.
That’s what “sense data” means.
So reality is defined in terms of sense data, and sense data is defined in terms of reality. No, that doesn’t work.
Okay, circling around on this to maybe say something more constructive now that I’ve thought about it a bit.
Part of the problem is that your central thesis is not very clear on first read, so I had to think about it a bit to really get what your “big idea” or ideas are that motivate this post. I realize you say right up top that you believe a strong version of verificationism is correct, but to me that’s not really getting at the core of what you’re thinking, that’s just something worked out by other people you can point at and say “something in that direction seems right”.
(FWIW, even some of the people who came up with logical positivism and related ideas like verificationism eventually worked themselves into a corner and realized there was no way out and the whole thing fell apart. The arguments for why it doesn’t work eventually get pretty subtle if you really press the issue, and I doubt I could do them justice, so I’ll stay higher level and may not have the time and energy to address every objection that you could bring up, but basically there’s 50+ years of literature trying to make ideas like this work and then finding there were inevitably problems.)
So, as best I can tell, the insight you had driving this was something like “oh, there’s no way to fully ground most statements, thus those statements are meaningless”. I’ll respond based on this assumption.
Now, up to a point you are right. Because there is an epistemological gap between the ontological and the ontic, statements about the ontic are “meaningless” in that they are not fully grounded. This is due to epistemic circularity, a modern formulation of the problem of the criterion. Thus the only statements that can be true in a classical sense are statements about statements, i.e. you can only get “truth” from inside an ontology, and there is no matter of truth to be assessed about statements of any other kind. This, alas, is not so great, because it takes all the power out of our everyday experience of truth.
One possible response is to hold this classical notion of truth fixed and say all those statements that can’t be grounded are false. The alternative is to say we screwed up on the classical notion of truth, and it’s not the category we thought it was.
Which every path you take, they converge to the same place, because if you reject things as not having truth value, now you have to introduce some new concept to talk about what everyone in everyday speech thinks of as truth, and you’ll be forced to go around annoying everyone with your jargon but whatever. The alternative is to accept that the classical notion of truth is malformed and not the category it was thought to be, and to rehabilitate truth in epistemology to match what people actually mean by it.
As I say, in the limit they converge to the same place, but in different language (cf. the situation with moral realism and anti-realism). I’ll speak now from the second perspective because it’s mine, but there’s one from the other side.
So then, if we were wrong that truth is about statements that can be proven true and grounded in reality, then what is truth? I take from your talk of “constrained expectations” you already see the gist of it: truth can be about predicting experiences, as in what is true is that which is part of a web of causality that generates our perceptions. This is messy, of course, because we’re embedded in the world and have to perceive it from inside it, but it gives us a way to make sense of what we really mean when we talk about truth, and to see that classical notions of truth were poor but reasonable first approximations of this, made with the intuitive assumption that there was ever some (classical) truth to know that was of a special kind. And on this view, truth of statements about the ontic are not meaningless; in fact, they are the only kind of statements you can make, because the ontological is ontic, but the ontic is not ontological.
Thus, also, why you have a comment section full of people arguing with you, because of course the natural notion of truth we have is meaningful, and it is only on a particular narrow view of it that it is not, which is right within itself but leaves out much and confusingly repurposes words to mean things they don’t normally mean.
What do you mean by ontic vs. ontological?
Standard rationalist terminology would be roughly territory and map, respectively.
Is there any difference?
Depends. In a certain vague sense, they are both okay pointers to what I think is the fundamental thing they are about, the two truths doctrine. In another sense, no, because the map and territory metaphor suggests a correspondence theory of truth, whereas ontological and ontic about mental categories and being or existence, respectively, and historically tied to a different approaches to truth, namely those associated with transcendental idealism. And if you don’t take my stance that they’re both different aspects of the same way of understanding reality that are contextualized in different ways and thus both wrong at some limit but in different ways, then there is an ocean of difference between them.
My central thesis is “beliefs that don’t constrain expectation are meaningless”, or more specifically to avoid some of the pitfalls, “beliefs about external reality are meaningless.”
For what it’s worth, I came up with the view first, then someone suggested verificationism fit my views, I saw that it was a form of logical positivism, recalled the Sequence posts on that, and reread them.
>basically there’s 50+ years of literature trying to make ideas like this work and then finding there were inevitably problems
I spent some time looking through critiques of verificationism. They all seemed to be disputing claims I don’t agree with. The analytic/synthetic distinction is irrelevant to my argument, but much of the critiques I found centered on that. I’m not saying that beliefs about mathematics are meaningless, so Godel isn’t relevant. It seems to me that all the problems are related to specific features of logical positivism proposals that I see no reason to accept.
I think people use truth/reality to mean several different things. As I said in the post, I’m ok with some of those uses. But there are definitely uses that appear meaningless and impossible to rescue, from my perspective—yet people will adamantly defend those uses. If by reality you just mean your best map that predicts experiences, that’s fine. But people will absolutely defend a notion of some “objective reality” that “really exists”, and there’s just no meaningful account of that that I can think of. So no, I’m not repurposing words to mean something new—the actual meaning of the word in common usage is incoherent.
I don’t know what your stance on moral realism is, but I’m confident there are places where I could make a post asserting moral anti-realism or error theory and get a bunch of smart people in the comments arguing with me. That doesn’t imply that moral realism is meaningful, or that you need to invent a new concept to represent what people mean when they talk about morals. Incidentally, different moral realists have different accounts of the nature of such beliefs, and I assume the general populace that believes in morals would have different understandings of what that means, if asked.
If you think it’s so obvious that the concept of external reality is coherent, then can you give an account of it? I don’t need a super detailed technical account, but at least some description of what it would mean for something to exist or not, without referencing predictions, just in the abstract.
Are you ever going to argue for that claim? It seems unreasinable that your opponents do all the work.
And are you going to define “coherent”?
I gave arguments for that in the post. Instead of responding to those, you insisted there’s some hypothetical value system that makes the concept of truth useful, and didn’t really argue for its coherency at all.
The burden of proof is on people asserting the positive, not entirely on those trying to prove the negative. Pointing out that nobody has given an account of meaning, and 50+ comments later still nobody has done so, is very suggestive that nobody actually has one.
A binary attribute is coherent if it’s possible for the attribute to hold as well as possible for the attribute not to hold. External reality is incoherent, because it never holds and it never doesn’t hold—there’s nothing that it would mean for it to hold or not hold.
Theres a value system according to which truth is valuable and that is the one where truth itself is a terminal value.
If you want to argue that usefulness is the only true value, go ahead
Shared meaning is what allows communication to take place. Therefore, communication taking place is evidence of shared meaning.
Do you think that verificationism is filling a vacuum ? That’s it’s the only theory of meaning anyone ever came up with? There are multiple fields that deal with the subject of meaning. There are multiple theories of meaning, not zero.
Again: the existence of external reality can be supported by abductive reasoning: it’s the best explanation for why science works at all.
You’re entitled to put whatever you want in your value system, but if it references incoherent things then you just can’t achieve it. Personally I would much prefer that 1+1=blrgh, but sadly not only does it equal 2, blrgh is meaningless. If I kept insisting that blrgh was meaningful and played a central role in my value system, what would you say? Also, I think truth can be defined without referencing external reality, just observations.
Evidence, but not definitive evidence, as I pointed out elsewhere. It’s possible for people to be mistaken about what has meaning, and I argue they are so mistaken when talking about external reality.
I’m talking specifically about the meaning of external reality. I looked up critiques of verificationism and didn’t see any accounts of such meaning. If there’s a particular account of the meaning of external reality that you’d like me to look at, let me know.
I responded to the explain argument elsewhere. Regardless, it’s unclear how an incoherent claim can serve as an explanation of anything.
Well, if it’s rational to only value things that are really achievable, you need a concept of reality .
Its where your sense data come from. You need the idea of an external world to define “sense organs”.
Incoherent != Impossible. It’s not possible for me to get a unicorn, but it’s coherent for me to want one. It’s incoherent for me to want to both have a unicorn and not have a unicorn at the same time. It’s incoherent for me to want the “true map” to have an XML tag saying “this really exists”, because the concept of a “true map” is meaningless, and the map with the tag and the map without yield the same predictions and so neither is more correct.
Disagree. If you’re defining it as the source of sense data, that’s roughly equivalent to the definition in The Simple Truth essay I linked to and responded to in OP.
My sense data can be predicted well by my map. Why do I need to define a concept of external reality to make sense of that?
You need it to explain why science works at all.
To summarise:
There’s a definition of “external reality”: it’s where your sense data come from.
There is a purpose served by the posit of an external reality: explaining how science works .
There is a meaning of “true model”, based on correspondence.
So every one of your detailed objections has been answered.
>You need it to explain why science works at all.
Why? How does realism explain why science works? What is the exact argument, premises, and conclusion?
>There’s a definition of “external reality”: it’s where your sense data come from.
This definition is incoherent, and I addressed it directly in OP. It’s also circular—it assumes your sense data comes from somewhere, which is precisely what I’m claiming is incoherent. Giving a definition that assumes the matter in debate is begging the question.
>There is a purpose served by the posit of an external reality: explaining how science works .
I still don’t know how your proposed explanation works. I think any such explanation can be replaced with an equally valid explanation that doesn’t assume incoherent concepts such as realism. And even if you managed to show that, all you’d be showing is that the concept is useful, which neither implies coherency nor truth.
>There is a meaning of “true model”, based on correspondence.
This again relies on the assumption of realism, which is circular.
Are you going to hold yourself to same standard?
Exactly and precisely how?
That’s what “sense data” means.
Circular “definitions* are ubiquitous. Circular arguments are the problem .
I’ve already given a definition of incoherent. A belief is coherent if it constrains expectations. A definition is incoherent if it relies on incoherent beliefs.
You’re making three claims: one, that realism is coherent; two, that realism explains how science works; three, that one can’t explain science as easily without realism.
My claim is that realism is incoherent, and that whatever argument you have for your second claim, I can find one just as good without using realism.
I only asked you to provide details on the second claim. If you do, I will hold myself to the same standards when arguing against the third. I can’t give a precise argument to explain science without realism now, because I don’t know what you want in an explanation until I’ve seen your argument.
So reality is defined in terms of sense data, and sense data is defined in terms of reality. No, that doesn’t work.
The definition you are now giving:
Isnt’t the same as the one you gave before:
But that would certainly explain why seem to have been using “incoherent” and “meaningless” interchangeably.
Binary attributes aren’t the same as beliefs.