Assume that Anselm's argument is valid - that it follows from the definition of "God" that God exists. In that case, couldn't we simply refuse to use the word "God" that Anselm has defined for us? If one of the premises of the argument is a definition, and definitions are constitutive of languages, isn't one rebuttal just not to speak an Anselmian language? If someone defined "phlogiston" so that it necessarily existed, she wouldn't thereby have a knock-down argument against the ontology of modern biology. She would have a linguistic quirk to be corrected or ignored.
A similar objection comes to mind in certain ethical and epistemological discussions, where it is argued that some substantive theory is just analytic. Someone might argue that it follows from the meanings of the words "will" and "good" that the object of ethical judgment is always the will, or that it follows from the meanings of the words "state" and "know" that we should state only what we know to be true. In response to this, can't we always simply refuse to use the relevant words with the meanings they are taken to have? Sometimes philosophers focus so much on the meanings and entailments associated with particular words because we're interested in better understanding the conceptual scheme we actually employ. But at some point, in these sorts of discussions of folk vocabulary, isn't it available to us to make new concepts? If we're clever enough, can't we cook up a "know" or a "good" that is better suited to our purposes than the "know" and "good" we have received from our ancestors?
I've assumed here that, at the crucial point, someone can't make a transcendental argument that we have to keep our folk vocabulary just as it is. But what form could such an argument take?
Showing posts with label phil phil. Show all posts
Showing posts with label phil phil. Show all posts
Tuesday, September 16, 2008
Wednesday, August 6, 2008
Explications and Empirical Revision
In what senses are explications open to empirical revision?
Say you have a term t and an explication of that term t`. For instance, t could be “heat” and t` could be “mean kinetic energy”. Or t could be “volume” and t` could be “amplitude”. Or t could be “more probable than” and t` could be “has a greater limiting relative frequency than”.
One obvious sense in which the explication t` could be empirically revised is that we could have some theory T1 in which t` occurs essentially, and then, in light of new empirical evidence, we replace T1 with some theory T2 in which t` does not occur. For instance, we could have a Newtonian mechanical theory, in which “heat” is explicated with a certain definition of “kinetic energy”, and (loosely speaking) the evidence could suggest that we replace it with a relativistic theory, in which the definition changes. Or the evidence could suggest that some classical theory of statistical mechanics be replaced by a quantum mechanical theory, in which (so I understand) the frequentist explication of "probability" is not appropriate.
Another slightly stronger sense in which t` could be empirically revised would be this. Suppose we have some theory T, containing the term t, then we replace every occurrence of t in T with the explication t`, producing the theory T`. Then, since T is presumably more vague or ambiguous than T`, there is some conclusive evidence against T` which is not conclusive evidence against T. If we discover just this evidence, then we should abandon T`, but we should not necessarily abandon T. I imagine that some would want to say, in this situation, that they had learned that they had chosen the wrong explication. This would be especially tempting if there were another explication t`` of t, such that the theory T``, gotten by replacing every occurrence of t in T with t``, were true, or supported by all of the relevant evidence at hand.
I feel like these senses aren’t strong enough to capture what someone might want to express by saying that explications are open to empirical revision. An explication is empirically revised in this stronger sense not only if the theories in which it occurs are found to be false, and not only if some competing explication better preserves the truth of theories in which the explicandum occurs. But is there such a sense in which explications are open to empirical revision?
Perhaps the idea is that there is some source of empirical evidence that can tell for or against the decision to explicate a term a certain way that is not empirical evidence for or against any particular theory in which the explicatum is used. This seems unlikely to me. What form could this sort of evidence possibly take? If we choose an explication, say, to render the empirical consequences of a theory more transparent, then we might have evidence that the explication fails or succeeds in doing so. We might have evidence, for instance, that “…t`…” is a borderline case but “…t…” isn’t. But would this be empirical evidence? Can’t we tell borderline cases from the armchair, given enough background information? Note that we can supply the necessary background information ourselves, using thought experiments. We might be incapable of telling what the empirical consequences of some explicated or unexplicated theory are if we don’t know what the theory says. So we should be familiar with theories in which explicanda occur in order to make our explications work best. But is this really empirical knowledge, in the relevant sense? What is necessary is not knowledge that any given theory is true, or knowledge of the evidence for or against a theory, but knowledge of what the theory says. (Of course, what makes what one theory says more or less important than what another theory says is the relative likelihood of the truth of the theories, but philosophers qua explicators don’t generally take on the task of empirically assessing the relative likelihoods of theories – generally, I suppose, they assess the importance of a scientific theory on the basis of its prominence in modern scientific discussion.)
Perhaps the idea is that, since no one sentence in which a (non-observational) term occurs is immune to empirical revision, then no sentence we pick out as the explication of a term is immune to empirical revision. This also seems implausible to me. As I understand it, part of what it is to explicate t with t` is to treat “t = t`” (perhaps relative to a given theory) as if it were the definition of t (in that theory). Part of what that amounts to is agreeing to the eliminability of t for t` in all extensional contexts (in a given theory). When we cease to agree to this, we change the subject. In general, as I see it, the explicit adoption of an explication of a term in a theory is an attempt to escape the consequences of various sorts of holism about that theory. We explicate, in part, to make the meanings of terms less diffuse. Of course, if a popular theory containing t is rendered trivially but empirically false by replacing all occurrences of t with t`, then t` is useless as an explication of t in that theory. Nobody would agree to explicate t with t` in such a case, and everyone’s refusal to adopt the explication would be on empirical grounds.
As it stands, this satisfies me that explications are open to empirical revision only in our first two senses. Am I missing something?
Say you have a term t and an explication of that term t`. For instance, t could be “heat” and t` could be “mean kinetic energy”. Or t could be “volume” and t` could be “amplitude”. Or t could be “more probable than” and t` could be “has a greater limiting relative frequency than”.
One obvious sense in which the explication t` could be empirically revised is that we could have some theory T1 in which t` occurs essentially, and then, in light of new empirical evidence, we replace T1 with some theory T2 in which t` does not occur. For instance, we could have a Newtonian mechanical theory, in which “heat” is explicated with a certain definition of “kinetic energy”, and (loosely speaking) the evidence could suggest that we replace it with a relativistic theory, in which the definition changes. Or the evidence could suggest that some classical theory of statistical mechanics be replaced by a quantum mechanical theory, in which (so I understand) the frequentist explication of "probability" is not appropriate.
Another slightly stronger sense in which t` could be empirically revised would be this. Suppose we have some theory T, containing the term t, then we replace every occurrence of t in T with the explication t`, producing the theory T`. Then, since T is presumably more vague or ambiguous than T`, there is some conclusive evidence against T` which is not conclusive evidence against T. If we discover just this evidence, then we should abandon T`, but we should not necessarily abandon T. I imagine that some would want to say, in this situation, that they had learned that they had chosen the wrong explication. This would be especially tempting if there were another explication t`` of t, such that the theory T``, gotten by replacing every occurrence of t in T with t``, were true, or supported by all of the relevant evidence at hand.
I feel like these senses aren’t strong enough to capture what someone might want to express by saying that explications are open to empirical revision. An explication is empirically revised in this stronger sense not only if the theories in which it occurs are found to be false, and not only if some competing explication better preserves the truth of theories in which the explicandum occurs. But is there such a sense in which explications are open to empirical revision?
Perhaps the idea is that there is some source of empirical evidence that can tell for or against the decision to explicate a term a certain way that is not empirical evidence for or against any particular theory in which the explicatum is used. This seems unlikely to me. What form could this sort of evidence possibly take? If we choose an explication, say, to render the empirical consequences of a theory more transparent, then we might have evidence that the explication fails or succeeds in doing so. We might have evidence, for instance, that “…t`…” is a borderline case but “…t…” isn’t. But would this be empirical evidence? Can’t we tell borderline cases from the armchair, given enough background information? Note that we can supply the necessary background information ourselves, using thought experiments. We might be incapable of telling what the empirical consequences of some explicated or unexplicated theory are if we don’t know what the theory says. So we should be familiar with theories in which explicanda occur in order to make our explications work best. But is this really empirical knowledge, in the relevant sense? What is necessary is not knowledge that any given theory is true, or knowledge of the evidence for or against a theory, but knowledge of what the theory says. (Of course, what makes what one theory says more or less important than what another theory says is the relative likelihood of the truth of the theories, but philosophers qua explicators don’t generally take on the task of empirically assessing the relative likelihoods of theories – generally, I suppose, they assess the importance of a scientific theory on the basis of its prominence in modern scientific discussion.)
Perhaps the idea is that, since no one sentence in which a (non-observational) term occurs is immune to empirical revision, then no sentence we pick out as the explication of a term is immune to empirical revision. This also seems implausible to me. As I understand it, part of what it is to explicate t with t` is to treat “t = t`” (perhaps relative to a given theory) as if it were the definition of t (in that theory). Part of what that amounts to is agreeing to the eliminability of t for t` in all extensional contexts (in a given theory). When we cease to agree to this, we change the subject. In general, as I see it, the explicit adoption of an explication of a term in a theory is an attempt to escape the consequences of various sorts of holism about that theory. We explicate, in part, to make the meanings of terms less diffuse. Of course, if a popular theory containing t is rendered trivially but empirically false by replacing all occurrences of t with t`, then t` is useless as an explication of t in that theory. Nobody would agree to explicate t with t` in such a case, and everyone’s refusal to adopt the explication would be on empirical grounds.
As it stands, this satisfies me that explications are open to empirical revision only in our first two senses. Am I missing something?
Wednesday, July 23, 2008
Some Thoughts About Intuitions
Why are intuitions with narrower content so much more evidentially trustworthy or warrant-conferring than intuitions with broader content? Why is it that if many of our narrower intuitions conflict with some of our broader intuitions, we often tend to drop the latter and not the former?
Better to intuit that “water” doesn’t mean the same thing on Earth as on Twin Earth than to intuit semantic externalism. If we intuit the former, this is better evidence against semantic internalism than an intuition that semantic internalism is true would be evidence against it.
Better to intuit that n grains of sand is not a heap and that if n grains of sand is not a heap, then n + 1 grains of sand is not a heap than to intuit that there are no heaps. In this case, I think, our intuition that there are heaps trumps our intuitions of the premises. Perhaps the premises, combined, are “broader” or stronger than the intuition that there are heaps, since they are recursive, and so have an infinite number of instances. Perhaps that is because I already know that there are heaps on the basis, say, of my perceptual evidence and my intuitions that, in this or that case, what I am looking at is a heap.
Better to intuit that I don’t really know that that is the façade of a barn than to intuit that knowledge is not justified true belief. Sometimes I think that my intuition that knowledge is justified true belief trumps my intuition about fake barn country. Perhaps this is because I sometimes think that K = JTB has too many merits to give up so easily. Of course, most of the time, people take the intuition about fake barn country (along with intuitions about other Gettier cases) to trump K = JTB.
I think these examples show that, in general, narrower intuitions are better evidence for their claims than broader intuitions. They don’t show that a narrower intuition always trumps a conflicting broader intuition, since the broader intuition might have more going for it than the conflicting narrower intuition brings against it. They suggest that a narrower intuition trumps a conflicting broader intuition, ceteris paribus.
How do we explain all of this? Part of it is that a narrower claim just, in general, has more warrant or a higher probability, relative to some piece of evidence, than any broader claim (of which the narrower claim is “an instance”), relative to that same evidence. That is, if I see Mike park the car, that is better evidence that Mike knows how to park a car than it is evidence that everyone named “Mike” knows how to park a car, or that everyone Mike’s age knows how to park a car. So, my intuition that p is, in general, better evidence that p than it is evidence that (a stronger claim) p & q, and better evidence that p & q than it is evidence that (an even stronger claim) p & q & r, and so on. If my intuition is that all bachelors are unmarried men, that is better evidence that all bachelors are unmarried men than it is evidence that, say, all men are unmarried men, or that all bachelors are unmarried men and all male college students are unmarried men.
Part of it is that we get more disagreement with intuitions on broader claims. Since much of the time the conflicting intuitions of two individuals will carry just as much evidential weight, these sorts of disagreements don’t affect the evidential standing of the claims of interest one way or another. Intuitions about semantic internalism and semantic externalism won’t get us very far, because people’s intuitions differ and carry equal evidential weight.
But I think the biggest part of the explanation is this. The trustworthiness or warrant-conferring-ness of intuitions is often, if not always, grounded in linguistic competence. If one assents to a sentence in a certain context strictly because one knows how to use a language, then, I think, we generally treat that assent as (prima facie, defeasibly) warranted. The idea might be that if pure knowledge of a language alone prompts assent, then that sentence must command assent in that context purely in virtue of linguistic rules or conventions; and whatever rules or conventions philosophers have to obey, the linguistic rules or conventions of the language of inquiry are surely among them. Whether or not this meta-explanation is right, I think we do, in fact, generally treat such assent as warranted.
But how does one know that the assent is grounded in linguistic competence alone, and not some other feature of the assenter’s cognitive make-up? One test is whether other speakers of the assenter’s language behave similarly. This will yield some false positives, though, since speakers of the assenter’s language have more in common with her, cognitively, than the language alone. This could be what grounds the assent, not linguistic competence, and there is no very compelling reason to think that just any shared cognitive trait will issue warranted pronouncements or steer us towards the true and away from the false.
Another test has fewer false positives. Specify all of the relevant facts that do not formally or pragmatically entail the claim of interest. Then, query the subject on the claim. If she intuitively assents or fails to assent, what could explain it? If we’ve specified the facts the right way, and made sure she understands them, then her assent could not be based on failure to believe the right facts. Perhaps she has failed to reason with the facts the right way? That is one explanation of her assent. But if we have no antecedent reason to doubt her reasoning skills, this is implausible. Perhaps her tacit linguistic knowledge disposes her to respond one way, but some consciously held theoretical commitment trumps this more trustworthy disposition. We should try to rule out this possibility somehow, if we want our intuitions to help us decide the claims they are about. If we do successfully rule it out, I can only see one other alternative explanation. This is that the subject's assent (or failure to assent) is grounded in her linguistic competence.
How does this explain that intuitions with narrower content are more trustworthy than intuitions with broader content? First, because it is easier to specify all of the relevant facts for a narrow claim than for a broader claim. It is easier to say everything about fake barn country, a certain heap of sand, or Twin Earth than it is to say everything about justified true belief, heaps (or, for that matter, vagueness in general), and semantic externalism. That is one thing our second test requires we do. Another thing is to make sure that no conscious theory is over-riding what her linguistic competence disposes the subject to do. Narrower claims meet this requirement more easily, since they are less likely to be obviously consistent or inconsistent with a high-level theory than broader claims at the theoretical level. Lastly, the test requires that we make sure that our subject has not failed to reason with the premises correctly. Since it is easier to reason with fewer premises, ceteris paribus, and we need to specify fewer relevant facts for narrower claims, narrower claims are more likely to pass our second test than broader claims.
Better to intuit that “water” doesn’t mean the same thing on Earth as on Twin Earth than to intuit semantic externalism. If we intuit the former, this is better evidence against semantic internalism than an intuition that semantic internalism is true would be evidence against it.
Better to intuit that n grains of sand is not a heap and that if n grains of sand is not a heap, then n + 1 grains of sand is not a heap than to intuit that there are no heaps. In this case, I think, our intuition that there are heaps trumps our intuitions of the premises. Perhaps the premises, combined, are “broader” or stronger than the intuition that there are heaps, since they are recursive, and so have an infinite number of instances. Perhaps that is because I already know that there are heaps on the basis, say, of my perceptual evidence and my intuitions that, in this or that case, what I am looking at is a heap.
Better to intuit that I don’t really know that that is the façade of a barn than to intuit that knowledge is not justified true belief. Sometimes I think that my intuition that knowledge is justified true belief trumps my intuition about fake barn country. Perhaps this is because I sometimes think that K = JTB has too many merits to give up so easily. Of course, most of the time, people take the intuition about fake barn country (along with intuitions about other Gettier cases) to trump K = JTB.
I think these examples show that, in general, narrower intuitions are better evidence for their claims than broader intuitions. They don’t show that a narrower intuition always trumps a conflicting broader intuition, since the broader intuition might have more going for it than the conflicting narrower intuition brings against it. They suggest that a narrower intuition trumps a conflicting broader intuition, ceteris paribus.
How do we explain all of this? Part of it is that a narrower claim just, in general, has more warrant or a higher probability, relative to some piece of evidence, than any broader claim (of which the narrower claim is “an instance”), relative to that same evidence. That is, if I see Mike park the car, that is better evidence that Mike knows how to park a car than it is evidence that everyone named “Mike” knows how to park a car, or that everyone Mike’s age knows how to park a car. So, my intuition that p is, in general, better evidence that p than it is evidence that (a stronger claim) p & q, and better evidence that p & q than it is evidence that (an even stronger claim) p & q & r, and so on. If my intuition is that all bachelors are unmarried men, that is better evidence that all bachelors are unmarried men than it is evidence that, say, all men are unmarried men, or that all bachelors are unmarried men and all male college students are unmarried men.
Part of it is that we get more disagreement with intuitions on broader claims. Since much of the time the conflicting intuitions of two individuals will carry just as much evidential weight, these sorts of disagreements don’t affect the evidential standing of the claims of interest one way or another. Intuitions about semantic internalism and semantic externalism won’t get us very far, because people’s intuitions differ and carry equal evidential weight.
But I think the biggest part of the explanation is this. The trustworthiness or warrant-conferring-ness of intuitions is often, if not always, grounded in linguistic competence. If one assents to a sentence in a certain context strictly because one knows how to use a language, then, I think, we generally treat that assent as (prima facie, defeasibly) warranted. The idea might be that if pure knowledge of a language alone prompts assent, then that sentence must command assent in that context purely in virtue of linguistic rules or conventions; and whatever rules or conventions philosophers have to obey, the linguistic rules or conventions of the language of inquiry are surely among them. Whether or not this meta-explanation is right, I think we do, in fact, generally treat such assent as warranted.
But how does one know that the assent is grounded in linguistic competence alone, and not some other feature of the assenter’s cognitive make-up? One test is whether other speakers of the assenter’s language behave similarly. This will yield some false positives, though, since speakers of the assenter’s language have more in common with her, cognitively, than the language alone. This could be what grounds the assent, not linguistic competence, and there is no very compelling reason to think that just any shared cognitive trait will issue warranted pronouncements or steer us towards the true and away from the false.
Another test has fewer false positives. Specify all of the relevant facts that do not formally or pragmatically entail the claim of interest. Then, query the subject on the claim. If she intuitively assents or fails to assent, what could explain it? If we’ve specified the facts the right way, and made sure she understands them, then her assent could not be based on failure to believe the right facts. Perhaps she has failed to reason with the facts the right way? That is one explanation of her assent. But if we have no antecedent reason to doubt her reasoning skills, this is implausible. Perhaps her tacit linguistic knowledge disposes her to respond one way, but some consciously held theoretical commitment trumps this more trustworthy disposition. We should try to rule out this possibility somehow, if we want our intuitions to help us decide the claims they are about. If we do successfully rule it out, I can only see one other alternative explanation. This is that the subject's assent (or failure to assent) is grounded in her linguistic competence.
How does this explain that intuitions with narrower content are more trustworthy than intuitions with broader content? First, because it is easier to specify all of the relevant facts for a narrow claim than for a broader claim. It is easier to say everything about fake barn country, a certain heap of sand, or Twin Earth than it is to say everything about justified true belief, heaps (or, for that matter, vagueness in general), and semantic externalism. That is one thing our second test requires we do. Another thing is to make sure that no conscious theory is over-riding what her linguistic competence disposes the subject to do. Narrower claims meet this requirement more easily, since they are less likely to be obviously consistent or inconsistent with a high-level theory than broader claims at the theoretical level. Lastly, the test requires that we make sure that our subject has not failed to reason with the premises correctly. Since it is easier to reason with fewer premises, ceteris paribus, and we need to specify fewer relevant facts for narrower claims, narrower claims are more likely to pass our second test than broader claims.
Friday, July 18, 2008
For the Disappointed Theorist
One of the ways you know you are doing a science is when the data force you to change your theory.
— Daniel Kahneman, APS 2006
Mutatis mutandis, "science" and "philosophy".
— Daniel Kahneman, APS 2006
Mutatis mutandis, "science" and "philosophy".
Wednesday, July 16, 2008
Even More Comments on The Philosophy of Philosophy
I see something wrong with my previous response to Williamson. The main point of Williamson’s (and our own) discussion of the possible varieties of epistemological analyticity is to explain how we can legitimately do philosophy from the armchair. To avoid one of Williamson’s objections to certain forms of epistemological analyticity, I suggested that understanding and truth should be relativized to languages in such a way that the discarded theoretical terms of discarded theories are not in the language of an up-to-date theorist. There might be analytic truths about phlogiston, and we might have to assent to or know their truth in order for us to understand (some) sentences containing “phlogiston”, but the relevant sort of truth and understanding in play here should be relativized to a language not used in up-to-date theory.
But aren’t philosophers sometimes concerned to know from the armchair the propositional contents of non-metalinguistic sentences, rather than their metalinguistic counterparts? That is, aren’t philosophers sometimes concerned to know, say, that every vixen is a female fox, not just that “Every vixen is a female fox” is true in the language of modern zoology? Knowing that a sentence is true in some language certainly isn’t sufficient for knowing its content, otherwise we would know, for instance, that phlogiston has a real functional role in combustion, and that the sun will rise tomorrow tonk the sun is purple.* Perhaps we might say that we know the content of “Every vixen is a female fox” if we know that it is analytic in our own zoological language and that we ought to be using our own zoological language. Or, to be a little more cautious, we might say that we know the content of "Every vixen is a female fox" if we know that it is analytic in every zoological language L such that (a) we know that L is maximally useful for us and (b) we know that the translation of "Every vixen is a female fox" in L is analytic. But certainly we don’t know what sort of zoological language we ought to be using from the armchair. For instance, we don’t know from the armchair whether the terms “vixen” or “fox” mark useful distinctions.
Still, as long as we understand a chunk of modern zoological theory in its customary linguistic guise, then we are in a position to know from the armchair what language that chunk of zoological theory does use, or what (slightly formalized) language or languages it can be taken to be using, or what language or languages it can be rationally reconstructed as using. This knowledge is a priori in the sense that it is guaranteed by our linguistic competence alone. And, for each of those languages, we are in a position to know from the armchair what the analytic truths are, construed epistemologically. At any rate, if I’m right about how to relativize truth and understanding to languages, then Williamson hasn’t shown that we aren’t in a position to know from the armchair what the analytic truths are, construed epistemologically. If philosophers are sometimes concerned to know from the armchair the propositional contents of non-metalinguistic analytic sentences, then they’re in for a disappointment. But maybe they'll do just fine if they set their sights a little bit lower.
This was Michael Friedman’s idea with the “relativized a priori”, right? Or, if Friedman was right, then this was all Carnap’s idea.
_____________________
* - Williamson makes roughly this point himself.
But aren’t philosophers sometimes concerned to know from the armchair the propositional contents of non-metalinguistic sentences, rather than their metalinguistic counterparts? That is, aren’t philosophers sometimes concerned to know, say, that every vixen is a female fox, not just that “Every vixen is a female fox” is true in the language of modern zoology? Knowing that a sentence is true in some language certainly isn’t sufficient for knowing its content, otherwise we would know, for instance, that phlogiston has a real functional role in combustion, and that the sun will rise tomorrow tonk the sun is purple.* Perhaps we might say that we know the content of “Every vixen is a female fox” if we know that it is analytic in our own zoological language and that we ought to be using our own zoological language. Or, to be a little more cautious, we might say that we know the content of "Every vixen is a female fox" if we know that it is analytic in every zoological language L such that (a) we know that L is maximally useful for us and (b) we know that the translation of "Every vixen is a female fox" in L is analytic. But certainly we don’t know what sort of zoological language we ought to be using from the armchair. For instance, we don’t know from the armchair whether the terms “vixen” or “fox” mark useful distinctions.
Still, as long as we understand a chunk of modern zoological theory in its customary linguistic guise, then we are in a position to know from the armchair what language that chunk of zoological theory does use, or what (slightly formalized) language or languages it can be taken to be using, or what language or languages it can be rationally reconstructed as using. This knowledge is a priori in the sense that it is guaranteed by our linguistic competence alone. And, for each of those languages, we are in a position to know from the armchair what the analytic truths are, construed epistemologically. At any rate, if I’m right about how to relativize truth and understanding to languages, then Williamson hasn’t shown that we aren’t in a position to know from the armchair what the analytic truths are, construed epistemologically. If philosophers are sometimes concerned to know from the armchair the propositional contents of non-metalinguistic analytic sentences, then they’re in for a disappointment. But maybe they'll do just fine if they set their sights a little bit lower.
This was Michael Friedman’s idea with the “relativized a priori”, right? Or, if Friedman was right, then this was all Carnap’s idea.
_____________________
* - Williamson makes roughly this point himself.
I understand “Every vixen is a female fox” and it has some positive epistemic status for me. How does it get that status? … The lazy theorist may try to dismiss the question, saying that it is simply part of our linguistic practice that “Every vixen is a female fox” has positive epistemic status for whoever understands it. But the examples of defective practices [surrounding “phlogiston”, “tonk”, racial pejoratives, and so on] show that it is not simply up to linguistic practices to distribute positive epistemic status as they please. That the practice is to treat some given sentence as having positive epistemic status for competent speakers of the language does not imply that it really has that epistemic status for them. (The Philosophy of Philosophy, 84)
Monday, July 14, 2008
How to Approach a Linguistic Item
When I encounter a word, phrase, or grammatical construction in need of philosophical explanation or clarification, what do I do? Sometimes the item derives its interest from its relation to something I already find interesting, but sometimes the interest is somehow intrinsic to the item itself. I’m more interested in what I do when the latter happens. What are the questions I have to bear in mind on my first encounter, as a philosopher, with a linguistic item of self-luminescent interest?
First, I think of examples. I look for seemingly typical or non-distinctive instances of the item. I try to situate it in a handful of different sentential and pragmatic contexts.
The next question I ask is: What is useful about this item? I think there are two ways of approaching this from the armchair. The first is to ask how things change after a sentence containing the item of interest is uttered in the intuitively typical or non-distinctive pragmatic contexts. What do I imagine would happen after the utterance? How might an interlocutor respond? What is true now of speaker and listener that was not true before?
The second approach is to ask what we would lose if we were not to allow this (or any heteronymous substitute) into our speech. On this approach, what I imagine is more a whole linguistic and theoretical world than a set of particular speech situations. For instance, would we fail to mark an important distinction? Would we lose some pragmatic tool, some ability (very broadly speaking) to change the social status of people or things? Would we not be able to express an interesting theory?
The first approach to the question of usefulness tells us what actual linguistic facts there are for an analysis or explication of the item to capture. The second approach tells us why it is worth capturing it. Both of these approaches are profitably initiated and conducted from the armchair, but both are also susceptible to experimental and observational test.
Is this it? What am I missing?
First, I think of examples. I look for seemingly typical or non-distinctive instances of the item. I try to situate it in a handful of different sentential and pragmatic contexts.
The next question I ask is: What is useful about this item? I think there are two ways of approaching this from the armchair. The first is to ask how things change after a sentence containing the item of interest is uttered in the intuitively typical or non-distinctive pragmatic contexts. What do I imagine would happen after the utterance? How might an interlocutor respond? What is true now of speaker and listener that was not true before?
The second approach is to ask what we would lose if we were not to allow this (or any heteronymous substitute) into our speech. On this approach, what I imagine is more a whole linguistic and theoretical world than a set of particular speech situations. For instance, would we fail to mark an important distinction? Would we lose some pragmatic tool, some ability (very broadly speaking) to change the social status of people or things? Would we not be able to express an interesting theory?
The first approach to the question of usefulness tells us what actual linguistic facts there are for an analysis or explication of the item to capture. The second approach tells us why it is worth capturing it. Both of these approaches are profitably initiated and conducted from the armchair, but both are also susceptible to experimental and observational test.
Is this it? What am I missing?
The Second Maxim
I feel as if, when philosophers attempt to rationally reconstruct or formalize or explicate some fragment of discourse, we should construe things so that as few (kinds of) sentences are truth-apt as possible. If the first maxim of scientific philosophizing is always to replace undefined primitives with logical constructs, the second maxim is to construe as few utterances as truth-apt assertions as possible. If we are concerned to reconstruct the most parsimonious theory of the world from some corpus of verbal behavior (and natural knowledge), then as little of the behavior should constitute endorsement of part of a theory as possible. Suppose the expressivist meta-ethical theory provides us with a range of terms and constructions (that do not yield truth-apt, assertive sentences) with which we might replace the rationally defensible bits of our current ethical discourse. The non-reductive realist meta-ethical theory provides us with a range of terms and constructions (that do yield truth-apt, assertive sentences) with which we might do the same. If both theories account equally well for the verbal behavior, don’t we have to endorse the expressivist theory? Isn’t the second maxim the reason why?
Sunday, July 13, 2008
More Comments on The Philosophy of Philosophy
In Chapter 4 of The Philosophy of Philosophy, “Epistemological Conceptions of Analyticity”, Williamson argues against epistemologies of analytic truths based on epistemological conceptions of analytic truths, which, in turn, are based on “understanding-assent” links. An understanding-assent link is a sentence like:
(1) Necessarily, whoever understands “All bachelors are unmarried” will assent to it.
I think the idea is that understanding links hold between object-level sentences corresponding to metalinguistic semantic facts, on the one hand, and assent to the content of the object-level sentences, on the other. If an understanding-assent link like (1) is supposed to provide the basis on which I know that “All bachelors are unmarried” is true, then this could be on the basis of a corresponding understanding-knowledge link:
(2) Necessarily, whoever understands “All bachelors are unmarried” knows that it is true.
Knowledge is factive. So, (2) entails:
(3) Necessarily, someone understands “All bachelors are unmarried” only if it is true.
But then, Williamson observes, it is unclear how to proceed with understanding-assent links related to several sorts of terms. I’ll just deal with “phlogiston” and other special theoretical terms from discredited theories, although what I have to say applies to the other sorts of terms he talks about.
It seems that fans of understanding-assent links will have to say that there are some even for “phlogiston”. Intuitively, part of the meaning of "phlogiston" is captured by its role in phlogiston theory. This commits us to:
(4) Necessarily, whoever understands “Phlogiston has the role R” will assent to it.
But phlogiston does not have the role R, because nothing plays the role that phlogiston plays in phlogiston theory. However, it follows from (4) that whoever does not assent to “Phlogiston has the role R” doesn’t understand it. Then it follows that people who think that nothing has role R don’t understand “Phlogiston has the role R”. Intuitively, this is not the case. So fans of understanding-assent links will have to accept something that is intuitively not the case.
I’m sure there are all sorts of ways around the problem Williamson is trying to set up for epistemologies of analytic truths based on understanding-assent links, but I’d like to propose just one.
First, excluding bizarre cases involving private codes, we understand sentences only relative to languages (or, if you like, idiolects). Second, sentences have their truth-values relative to languages (or idiolects). Assume, contrary to Williamson, that assent is generally metalinguistic – to assent to “All bachelors are unmarried” is, generally, to assent to the metalinguistic fact that “All bachelors are unmarried” is true, not to assent the corresponding object-level fact that all bachelors are unmarried. So, fully unpacked, (4) means something like:
(4`) Necessarily, whoever understands “Phlogiston has the role R” in L will assent to the metalinguistic fact that “Phlogiston has the role R” is true in L.
In a theoretically important sense, phlogiston theory is stated in a different language from our current chemical theory. (I think this is the sense in which languages classically do not contain their own truth predicate, and probably the sense in which sentences of a language have determinate logical forms.) When doing chemical theory today, we do not speak the L mentioned in (4`). We are not speaking L even when we say that phlogiston theory is false, or that nothing has the role R. If a fully-spelled out description of R invokes other special theoretical terms of phlogiston theory, perhaps a good interpretation of “Nothing has the role R” is “We should not speak a language in which a sentence of the form ‘x has the role R’ is true, for some name substituted for ‘x’.”
The problem with words like “phlogiston” and “tonk” (and, an eliminativist might say, “believes”) is that it is a bad idea to use them at all (except, perhaps, to say something like that phlogiston doesn’t exist). It is a bad idea to use them because it is a bad idea to use languages that countenance them - languages which commit us to parts of phlogiston theory, or in which everything is true or nothing is true. Still, I think we can endorse understanding-assent links for even these sorts of words by relativizing the understanding and the assent in question to languages that we do not use.
File under: Applied Carnap.
________________
Update: This is a pretty bad objection, now that I look at it again. That's because (4`) is false. If people don't know what language L is - and most people don't, on the ontology of languages necessary to make this work - then they won't (or shouldn't) assent to statements mentioning it, including (4`). And the business about the ellipticality of "Nothing has the role R" seems pretty implausible.
(1) Necessarily, whoever understands “All bachelors are unmarried” will assent to it.
I think the idea is that understanding links hold between object-level sentences corresponding to metalinguistic semantic facts, on the one hand, and assent to the content of the object-level sentences, on the other. If an understanding-assent link like (1) is supposed to provide the basis on which I know that “All bachelors are unmarried” is true, then this could be on the basis of a corresponding understanding-knowledge link:
(2) Necessarily, whoever understands “All bachelors are unmarried” knows that it is true.
Knowledge is factive. So, (2) entails:
(3) Necessarily, someone understands “All bachelors are unmarried” only if it is true.
But then, Williamson observes, it is unclear how to proceed with understanding-assent links related to several sorts of terms. I’ll just deal with “phlogiston” and other special theoretical terms from discredited theories, although what I have to say applies to the other sorts of terms he talks about.
It seems that fans of understanding-assent links will have to say that there are some even for “phlogiston”. Intuitively, part of the meaning of "phlogiston" is captured by its role in phlogiston theory. This commits us to:
(4) Necessarily, whoever understands “Phlogiston has the role R” will assent to it.
But phlogiston does not have the role R, because nothing plays the role that phlogiston plays in phlogiston theory. However, it follows from (4) that whoever does not assent to “Phlogiston has the role R” doesn’t understand it. Then it follows that people who think that nothing has role R don’t understand “Phlogiston has the role R”. Intuitively, this is not the case. So fans of understanding-assent links will have to accept something that is intuitively not the case.
I’m sure there are all sorts of ways around the problem Williamson is trying to set up for epistemologies of analytic truths based on understanding-assent links, but I’d like to propose just one.
First, excluding bizarre cases involving private codes, we understand sentences only relative to languages (or, if you like, idiolects). Second, sentences have their truth-values relative to languages (or idiolects). Assume, contrary to Williamson, that assent is generally metalinguistic – to assent to “All bachelors are unmarried” is, generally, to assent to the metalinguistic fact that “All bachelors are unmarried” is true, not to assent the corresponding object-level fact that all bachelors are unmarried. So, fully unpacked, (4) means something like:
(4`) Necessarily, whoever understands “Phlogiston has the role R” in L will assent to the metalinguistic fact that “Phlogiston has the role R” is true in L.
In a theoretically important sense, phlogiston theory is stated in a different language from our current chemical theory. (I think this is the sense in which languages classically do not contain their own truth predicate, and probably the sense in which sentences of a language have determinate logical forms.) When doing chemical theory today, we do not speak the L mentioned in (4`). We are not speaking L even when we say that phlogiston theory is false, or that nothing has the role R. If a fully-spelled out description of R invokes other special theoretical terms of phlogiston theory, perhaps a good interpretation of “Nothing has the role R” is “We should not speak a language in which a sentence of the form ‘x has the role R’ is true, for some name substituted for ‘x’.”
The problem with words like “phlogiston” and “tonk” (and, an eliminativist might say, “believes”) is that it is a bad idea to use them at all (except, perhaps, to say something like that phlogiston doesn’t exist). It is a bad idea to use them because it is a bad idea to use languages that countenance them - languages which commit us to parts of phlogiston theory, or in which everything is true or nothing is true. Still, I think we can endorse understanding-assent links for even these sorts of words by relativizing the understanding and the assent in question to languages that we do not use.
File under: Applied Carnap.
________________
Update: This is a pretty bad objection, now that I look at it again. That's because (4`) is false. If people don't know what language L is - and most people don't, on the ontology of languages necessary to make this work - then they won't (or shouldn't) assent to statements mentioning it, including (4`). And the business about the ellipticality of "Nothing has the role R" seems pretty implausible.
Wednesday, July 9, 2008
Comments on The Philosophy of Philosophy
In Chapter 2 of The Philosophy of Philosophy, “Taking Philosophical Questions at Face Value”, Timothy Williamson argues that a certain philosophical or “proto-philosophical” question is not, explicitly or implicitly, about language. This, what he calls the original question, is: “Was Mars always either dry or not dry?” He shows how a number of ways of answering the original question, through the consideration of intuitionistic, three-valued, and fuzzy logics, still don’t make it a linguistic question, since the answers are not about (i.e. don’t refer to) linguistic items. The answers are “Mars was always either dry or not dry”, “Mars was not always either dry or not dry”, and “It is indefinite whether Mars was always either dry or not dry.” Since none of these answers is about language, the question is not a question about language.
Let’s focus on yes-no questions for now. Say that A is a straightforward answer to a yes-no question Q, stated in language L, iff A is stated in L and expresses what “yes” would express or expresses what “no” would express. Clearly, not every non-straightforward answer to a question is about language. If Susan asks “You ate lunch at Vinny’s last night?”, and Tim responds “Actually, I went to Aunt Suzie’s”, Tim does not give a straightforward answer, but neither does he give a linguistic answer. Nor is every linguistic answer non-straightforward. If Tim responded “True” – as in “What you just said is true, stated indicatively” – I think the answer is both straightforward, because equivalent to “yes”, and linguistic, because about a sentence. Still, most linguistic answers are non-straightforward. If Tim responded “Depends on what you mean by ‘lunch’”, because, say, he ate a borderline meal of salad and an omelet at 11:45, that would be a typical non-straightforward, linguistic answer.
Note also that linguistic questions – questions about language – admit of non-straightforward and perhaps also straightforward non-linguistic answers.
The point is that questions that are about language admit of non-linguistic answers, and questions that aren’t about language admit of linguistic answers.
One way for a question to be implicitly, but not explicitly, about language, relative to a kind of answer K, is for all of the members of K to be explicitly about language. Williamson has shown that the original question is not in this way implicitly about language, relative to its philosophical answers, since the philosophical answers are not explicitly about language. But might the question be implicitly about language because the philosophical answers are implicitly about language? I kinda think so, for the following reasons.
(1) and (3) are obvious. Although I’m not sure Quine would agree with me on (2), I think Williamson would. (4) is probably the most controversial, but I take it that Williamson should agree with me on that as well, judging by what he has to say about Vann McGee in his paper “Understanding and Inference.” But (6) straightforwardly follows from (1)-(4).
Now, once we get to (6), it’s not obvious that every answer in L1 to a question in some other language L2 is thereby a linguistic answer. After all, if a bilingual speaker asks me how the weather is in English, and I answer “Hace fresco”, I have not thereby given a linguistic answer. But, I want to say, that is because it was merely a manner of speaking for me to answer in Spanish. The philosopher who answers in a three-valued language, or a fuzzy language, or an intuitionistic language, or a classical language thinks she has to answer in that language, because that is the right language in which to answer the question, or the only (kind of) language in which to state her theory of vagueness, then the answer is not merely a manner of speaking. The step of translation from the logical language to natural English is a necessary step for the philosopher to give the sort of answer she wants to give. I want to say that it is in virtue of this necessity that the original question is linguistic, at least relative to these sorts of answers grounded in logical metareflection.
I think it is reasonable to say that there is a sense in which a question is implicitly about language, relative to a kind of answer K, iff every member of K is in another language because it must, for the speaker’s most cherished purposes, be in another language. So it is, apparently, with the original question and its philosophical-type answers – or at least the original question and the philosophical-type answers that Williamson has on offer. I guess that if the deconstructionist wants to say (in English) that Mars was always both dry and not dry, because binary distinctions are always unstable and every inscription of both “Mars has always been dry” and “Mars has always not been dry” is internally contradictory, then we have a philosophical English-language answer to our English-language question. But that’s not the kind of philosophy we were talking about, right? Weren’t we looking for the right philosophy of analytic philosophy?
Let’s focus on yes-no questions for now. Say that A is a straightforward answer to a yes-no question Q, stated in language L, iff A is stated in L and expresses what “yes” would express or expresses what “no” would express. Clearly, not every non-straightforward answer to a question is about language. If Susan asks “You ate lunch at Vinny’s last night?”, and Tim responds “Actually, I went to Aunt Suzie’s”, Tim does not give a straightforward answer, but neither does he give a linguistic answer. Nor is every linguistic answer non-straightforward. If Tim responded “True” – as in “What you just said is true, stated indicatively” – I think the answer is both straightforward, because equivalent to “yes”, and linguistic, because about a sentence. Still, most linguistic answers are non-straightforward. If Tim responded “Depends on what you mean by ‘lunch’”, because, say, he ate a borderline meal of salad and an omelet at 11:45, that would be a typical non-straightforward, linguistic answer.
Note also that linguistic questions – questions about language – admit of non-straightforward and perhaps also straightforward non-linguistic answers.
Tim: “… but then I had to jet to the supermarket.”Tim’s first answer might be straightforward. His second answer is non-straightforward. Neither answer is about language.
Susan: “What does ‘jet’ mean?”
Tim: “Somebody jets somewhere whenever they try to get there very quickly.”
Or
Tim: “I had to get there very quickly.”
The point is that questions that are about language admit of non-linguistic answers, and questions that aren’t about language admit of linguistic answers.
One way for a question to be implicitly, but not explicitly, about language, relative to a kind of answer K, is for all of the members of K to be explicitly about language. Williamson has shown that the original question is not in this way implicitly about language, relative to its philosophical answers, since the philosophical answers are not explicitly about language. But might the question be implicitly about language because the philosophical answers are implicitly about language? I kinda think so, for the following reasons.
1) The original question is stated in English.
2) Languages are partially constituted by their logics. Two things with different logics cannot be the same language.
3) The language of each answer has some formal logic – three-valued, fuzzy, intuitionistic, classical, etc.
4) English has no formal logic – neither three-valued, nor fuzzy, nor intuitionistic, nor classical, etc.
5) Therefore, English does not have the same logic as the language of any of the answers.
6) Therefore, the language of each of the answers is not the same as the language of the original question.
(1) and (3) are obvious. Although I’m not sure Quine would agree with me on (2), I think Williamson would. (4) is probably the most controversial, but I take it that Williamson should agree with me on that as well, judging by what he has to say about Vann McGee in his paper “Understanding and Inference.” But (6) straightforwardly follows from (1)-(4).
Now, once we get to (6), it’s not obvious that every answer in L1 to a question in some other language L2 is thereby a linguistic answer. After all, if a bilingual speaker asks me how the weather is in English, and I answer “Hace fresco”, I have not thereby given a linguistic answer. But, I want to say, that is because it was merely a manner of speaking for me to answer in Spanish. The philosopher who answers in a three-valued language, or a fuzzy language, or an intuitionistic language, or a classical language thinks she has to answer in that language, because that is the right language in which to answer the question, or the only (kind of) language in which to state her theory of vagueness, then the answer is not merely a manner of speaking. The step of translation from the logical language to natural English is a necessary step for the philosopher to give the sort of answer she wants to give. I want to say that it is in virtue of this necessity that the original question is linguistic, at least relative to these sorts of answers grounded in logical metareflection.
I think it is reasonable to say that there is a sense in which a question is implicitly about language, relative to a kind of answer K, iff every member of K is in another language because it must, for the speaker’s most cherished purposes, be in another language. So it is, apparently, with the original question and its philosophical-type answers – or at least the original question and the philosophical-type answers that Williamson has on offer. I guess that if the deconstructionist wants to say (in English) that Mars was always both dry and not dry, because binary distinctions are always unstable and every inscription of both “Mars has always been dry” and “Mars has always not been dry” is internally contradictory, then we have a philosophical English-language answer to our English-language question. But that’s not the kind of philosophy we were talking about, right? Weren’t we looking for the right philosophy of analytic philosophy?
Subscribe to:
Posts (Atom)
