What Do Philosophers Believe?
May 2, 2013 11:31 AM Subscribe
What Do Philosophers Believe? David Bourget and David Chalmers, co-directors of Philpapers.com, have written an article based on the PhilPapers Survey of professional philosophers. It covers the popularity of various views, correlations with age, gender, and geography, a factor analysis that tries to isolate important underlying factors; and discussion of the results of the Metasurvey, bringing out just how surprising some of the survey results are. The article is forthcoming in Philosophical Studies.
There is famously no consensus on the answers to most major philosophical questions. Still, some of the questions on the survey came closer to drawing a consensus than others. In particular, the following views all had normalized positive answer rates of approximately 70% or more: a priori knowledge, the analytic-synthetic distinction, non-skeptical realism, compatibilism, atheism, non-Humeanism about laws, cognitivism about moral judgment, classicism about logic, externalism about mental content, scientific realism, and trolley switching.
The Metasurvey indicates that a number of the preceding positions were not expected to reach this level of agreement: a priori knowledge, the analytic-synthetic distinction, non-Humeanism about laws, cognitivism about moral judgment, scientific realism, and trolley switching were all predicted to achieve rates at least 15% lower. For most of these questions, respondents to the Metasurvey underestimated agreement on the leading positions. Two notable exceptions are subjectivism about aesthetic value (estimate: 67.7%, actual: 45.7%) and empiricism (estimate: 66.5%, actual: 55.7%)
The correlations and principal component analysis reported in the preceding sections suggest that philosophical views tend to come in packages. Our analysis reveals five major choice points in logical space: naturalism vs anti-naturalism, objectivism vs subjectivism, rationalism vs empiricism, realism vs anti-realism (of the kind associated with epistemic theories of truth), internalism vs externalism.
[...]
The Metasurvey suggests that philosophers often have highly inaccurate sociological beliefs. The Survey itself may contribute to the project of correcting these beliefs. Given the important roles that sociological beliefs sometimes play in philosophy, there may well be room for more surveys of the philosophical views of professional philosophers.
There is famously no consensus on the answers to most major philosophical questions. Still, some of the questions on the survey came closer to drawing a consensus than others. In particular, the following views all had normalized positive answer rates of approximately 70% or more: a priori knowledge, the analytic-synthetic distinction, non-skeptical realism, compatibilism, atheism, non-Humeanism about laws, cognitivism about moral judgment, classicism about logic, externalism about mental content, scientific realism, and trolley switching.
The Metasurvey indicates that a number of the preceding positions were not expected to reach this level of agreement: a priori knowledge, the analytic-synthetic distinction, non-Humeanism about laws, cognitivism about moral judgment, scientific realism, and trolley switching were all predicted to achieve rates at least 15% lower. For most of these questions, respondents to the Metasurvey underestimated agreement on the leading positions. Two notable exceptions are subjectivism about aesthetic value (estimate: 67.7%, actual: 45.7%) and empiricism (estimate: 66.5%, actual: 55.7%)
The correlations and principal component analysis reported in the preceding sections suggest that philosophical views tend to come in packages. Our analysis reveals five major choice points in logical space: naturalism vs anti-naturalism, objectivism vs subjectivism, rationalism vs empiricism, realism vs anti-realism (of the kind associated with epistemic theories of truth), internalism vs externalism.
[...]
The Metasurvey suggests that philosophers often have highly inaccurate sociological beliefs. The Survey itself may contribute to the project of correcting these beliefs. Given the important roles that sociological beliefs sometimes play in philosophy, there may well be room for more surveys of the philosophical views of professional philosophers.
now you need a meta-meta survey about what schemas of beliefs categorizing philosophy that philosophers believe...
posted by ennui.bz at 11:52 AM on May 2, 2013 [1 favorite]
posted by ennui.bz at 11:52 AM on May 2, 2013 [1 favorite]
Analyticity. The a priori. Two-boxing. Deontology. Correspondence theory of truth. Conceivable zombies. Ugh.
But otherwise, everyone did pretty well!
posted by painquale at 12:05 PM on May 2, 2013 [1 favorite]
But otherwise, everyone did pretty well!
posted by painquale at 12:05 PM on May 2, 2013 [1 favorite]
It should be acknowledged that this target group has a strong (although not exclusive) bias toward analytic or Anglocentric philosophy. As a consequence, the results of the survey are a much better guide to what analytic/Anglocentric philosophers (or at least philosophers in strong analytic/Anglocentric departments believe) believe than to what philosophers from other traditions believe.
Yeah that pretty much sums it up.
posted by shivohum at 12:24 PM on May 2, 2013 [5 favorites]
Yeah that pretty much sums it up.
posted by shivohum at 12:24 PM on May 2, 2013 [5 favorites]
How does one disprove Hume's argument...?
Humeanism with respect to laws of nature is the view that the laws of nature are nothing more than regularities among particular matters of fact. This is quite a bit stronger than your example makes out. One way to see why someone might disagree with this view is to distinguish between "governing" and "non-governing" conceptions of the laws of nature. Ordinarily we talk about laws of nature as governing (and thereby explaining) what happens in particular observable cases. So regularities in particular natural phenomena are explained by the fact that the phenomena in question are governed by certain natural laws. On the Humean view, by contrast, the laws are nothing more than the regularities in the particular phenomena. What explains why the phenomena are subject to those regularities, and not others? Nothing -- it's a brute fact about the world that they do.
So, to take up your example, both sides could agree that the law of gravity might fail tomorrow. But on the Humean view, if that happened, there could in principle be no explanation for it, whereas the non-Humean might think that something had better explain the change.
posted by voltairemodern at 12:26 PM on May 2, 2013 [1 favorite]
Humeanism with respect to laws of nature is the view that the laws of nature are nothing more than regularities among particular matters of fact. This is quite a bit stronger than your example makes out. One way to see why someone might disagree with this view is to distinguish between "governing" and "non-governing" conceptions of the laws of nature. Ordinarily we talk about laws of nature as governing (and thereby explaining) what happens in particular observable cases. So regularities in particular natural phenomena are explained by the fact that the phenomena in question are governed by certain natural laws. On the Humean view, by contrast, the laws are nothing more than the regularities in the particular phenomena. What explains why the phenomena are subject to those regularities, and not others? Nothing -- it's a brute fact about the world that they do.
So, to take up your example, both sides could agree that the law of gravity might fail tomorrow. But on the Humean view, if that happened, there could in principle be no explanation for it, whereas the non-Humean might think that something had better explain the change.
posted by voltairemodern at 12:26 PM on May 2, 2013 [1 favorite]
previously on metafilter (they certainly took their time analysing the results of this)
posted by leibniz at 12:28 PM on May 2, 2013 [1 favorite]
posted by leibniz at 12:28 PM on May 2, 2013 [1 favorite]
Analyticity. The a priori. Two-boxing. Deontology. Correspondence theory of truth. Conceivable zombies. Ugh.
To be fair, two-boxing was a close call. And I think they rejected the right premise of the zombie argument (the inference from conceivability to possibility).
posted by voltairemodern at 12:31 PM on May 2, 2013
To be fair, two-boxing was a close call. And I think they rejected the right premise of the zombie argument (the inference from conceivability to possibility).
posted by voltairemodern at 12:31 PM on May 2, 2013
Interestingly, specialists were more likely than non-specialists to reject the dichotomies posed.
How totally unsurprising does a result have to be before it doesn't count as "interesting"?
posted by RogerB at 12:45 PM on May 2, 2013 [1 favorite]
How totally unsurprising does a result have to be before it doesn't count as "interesting"?
posted by RogerB at 12:45 PM on May 2, 2013 [1 favorite]
@painquale: What's the problem?
I was just picking out the winning positions that I found most ugh-worthy. (Not that I agree with all the other winners... they're just not as ugh-worthy.)
posted by painquale at 12:58 PM on May 2, 2013
I was just picking out the winning positions that I found most ugh-worthy. (Not that I agree with all the other winners... they're just not as ugh-worthy.)
posted by painquale at 12:58 PM on May 2, 2013
painquale, I'm curious as to why you don't believe in analyticity or apriori knowledge.
The surprise for me is that a plurality thinks aesthetic value is objective. I'm glad we're not settling these questions for all time based on first-past-the-post rules.
posted by justsomebodythatyouusedtoknow at 1:00 PM on May 2, 2013
The surprise for me is that a plurality thinks aesthetic value is objective. I'm glad we're not settling these questions for all time based on first-past-the-post rules.
posted by justsomebodythatyouusedtoknow at 1:00 PM on May 2, 2013
@painquale: Why do you disagree with these positions? This may come off as rude and I honestly don't intend it to be this way nor do I want to single you out as such, but I come to Metafilter for more interesting discussions and comments and seeing a lot of "[X] is dumb" or "ugh, more [x]?" makes it hard to have any kind of meaningful exchange. I'd be interested in learning why you have some objection to these things.
posted by koavf at 1:04 PM on May 2, 2013 [3 favorites]
posted by koavf at 1:04 PM on May 2, 2013 [3 favorites]
To some extent this survey does not lend itself to talking about the issues. We're presented with conclusions without arguments. This study, while interesting as sociology, is methodologically the opposite of philosophy.
posted by justsomebodythatyouusedtoknow at 1:07 PM on May 2, 2013
posted by justsomebodythatyouusedtoknow at 1:07 PM on May 2, 2013
This study, while interesting as sociology, is methodologically the opposite of philosophy.
I think that is the point.
posted by MisantropicPainforest at 1:18 PM on May 2, 2013 [3 favorites]
I think that is the point.
posted by MisantropicPainforest at 1:18 PM on May 2, 2013 [3 favorites]
The real shocker for me here is that 60% think free will is compatible with determinism. Weird.
This seems like a great outline for a book for non-philosophers. I'd love to see a brief overview explaining the basic argument behind each of the 30 questions (yeah, I know, wikipedia, but I'm lazy).
And since, as others point out, this is methodologically more like sociology than philosophy, I'd love to see someone tackle a more sociological (borderline ad hominem) perspective on these questions, "Here's some non-philosophical reasons that might explain why a majority of today's philosophers say they'd switch the trolley tracks or take both Newcomb boxes."
posted by straight at 2:02 PM on May 2, 2013 [1 favorite]
This seems like a great outline for a book for non-philosophers. I'd love to see a brief overview explaining the basic argument behind each of the 30 questions (yeah, I know, wikipedia, but I'm lazy).
And since, as others point out, this is methodologically more like sociology than philosophy, I'd love to see someone tackle a more sociological (borderline ad hominem) perspective on these questions, "Here's some non-philosophical reasons that might explain why a majority of today's philosophers say they'd switch the trolley tracks or take both Newcomb boxes."
posted by straight at 2:02 PM on May 2, 2013 [1 favorite]
The real shocker for me here is that 60% think free will is compatible with determinism. Weird.
I've talked with so many people who don't seem to understand free will that I'm beginning to wonder if there are multiple definitions out there. They start saying things about how we should act if we don't have free will, as if we'd have a choice. And these are smart people so I find it very confusing.
posted by sineater at 2:07 PM on May 2, 2013 [2 favorites]
I've talked with so many people who don't seem to understand free will that I'm beginning to wonder if there are multiple definitions out there. They start saying things about how we should act if we don't have free will, as if we'd have a choice. And these are smart people so I find it very confusing.
posted by sineater at 2:07 PM on May 2, 2013 [2 favorites]
NYTimes piece about Daniel Dennett: Philosophy That Stirs the Waters
posted by homunculus at 2:09 PM on May 2, 2013
posted by homunculus at 2:09 PM on May 2, 2013
I'm beginning to wonder if there are multiple definitions out there
Well that's definitely true. "Free Will" seems to combine several ideas - the concept of moral responsibility, the concept of having the ability to do otherwise, the concept of being the genuine source of one's actions.
posted by straight at 2:13 PM on May 2, 2013 [3 favorites]
Well that's definitely true. "Free Will" seems to combine several ideas - the concept of moral responsibility, the concept of having the ability to do otherwise, the concept of being the genuine source of one's actions.
posted by straight at 2:13 PM on May 2, 2013 [3 favorites]
"Free Will" seems to combine several ideas
interesting. I'm way out of my depth here. But both of those links indicate that determinism and free will are incompatible.
I think what confuses my friends is that it obviously seems so deeply that we do have free will that it's difficult to imagine what it means to not have it. I'm not sure I grasp it myself other than very simply. More and more experiments are showing that our brains are making decisions before the conscious part of our brain comes into play in the way we normally think of making choices. These experiments don't yet prove that we don't have free will (and maybe never will) but I always find it amusing to have friends react by discussing ways in which our behavior should change if it turns out we don't have free will. Talk about not missing the forest and the trees :-)
posted by sineater at 2:24 PM on May 2, 2013 [2 favorites]
interesting. I'm way out of my depth here. But both of those links indicate that determinism and free will are incompatible.
I think what confuses my friends is that it obviously seems so deeply that we do have free will that it's difficult to imagine what it means to not have it. I'm not sure I grasp it myself other than very simply. More and more experiments are showing that our brains are making decisions before the conscious part of our brain comes into play in the way we normally think of making choices. These experiments don't yet prove that we don't have free will (and maybe never will) but I always find it amusing to have friends react by discussing ways in which our behavior should change if it turns out we don't have free will. Talk about not missing the forest and the trees :-)
posted by sineater at 2:24 PM on May 2, 2013 [2 favorites]
Why do you disagree with these positions? This may come off as rude and I honestly don't intend it to be this way nor do I want to single you out as such, but I come to Metafilter for more interesting discussions and comments and seeing a lot of "[X] is dumb" or "ugh, more [x]?" makes it hard to have any kind of meaningful exchange. I'd be interested in learning why you have some objection to these things.
Well, I was just bleating into the void, but if you want me to defend my ughs, you'll have to permit me to give some very cavalier and pithy defenses. I'm hardly going to give a compelling defense of consequentialism over deontology in a Metafilter comment. Please forgive some sloppiness.
Analyticity: Quine and others have tried to offer principled arguments against analytic truths, but I find Fodor's empirical argument best: in hundreds of years, pretty much no one has been able to come up with something that is a clear analytic truth in natural language. Examples that are offered, like "bachelors are unmarried males" or "conception causes pregnancy", have counterexamples. If there are analytic statements, they're extremely rare. (I admit that "vixens are female foxes" is pretty good.) Also, analytic statements are supposed to be different than logical truths, and I haven't seen a great explanation of the difference (without begging questions about what counts as a logical symbol).
The Apriori: I have problems with the category itself. An apriori proposition is one that could be known independently of sense experience. Known to whom? Pretty much any true proposition is something that a possible creature could be born innately believing. If you think that innate beliefs can be justified, or reliably formed, or whatever it takes for a belief to be knowledge, then it looks pretty easy to make it so that any proposition could be known innately by a possible cognizer; it would follow from the standard definition that every proposition is a priori. So, it's a bad definition. But is there a better one? Saying that propositions are a priori if we could know them innately seems to get the extension wrong entirely, because we apparently have some innate knowledge that no philosopher would want to say is a priori (e.g. about causation, about word segmentation, etc.). I don't think that there's a useful regimentation of the concept. The a priori is supposed to be different from the innate or the analytic, but I don't think there's space for such a category.
Two-boxing: Basically, two-boxing makes sense if you subscribe to causal decision theory rather than evidential decision theory. I don't see why causation is something that we should build into our epistemology, I think our intuitions about it are all over the map, and CDT seems to me to get a whole bunch of toy cases wrong compared to EDT. (I talked about CDT and EDT a little in this thread about voting.)
Correspondence Theory of Truth: The theory claims that a sentence is true if it corresponds to a fact. You have to be committed to facts for this to make sense. So what are they? For various reasons, the correspondence theorist cannot think that facts are true sentences or states of affairs. You very get into horrible questions about the individuation conditions of facts, their ontological status, etc. Is the fact that some rectangle is twice as wide as it is tall the same fact as the fact that it's half as tall as it is wide? These are awful questions. I think facts just are true sentences (or propositions). So they cannot be used to explain how it is that a sentence is true.
Deontology: Super pithy here. A consequentialist theory says that if you can do something that will make the world a better place overall, then you should do it instead of something that will make the world a less good place. That is a principle I am happy to commit myself to and I think it's kinda gross to deny it. If a deontological theory can't be "consequentialized", then it denies that principle. Ugh.
Conceivable Zombies: If we're talking about ideal conceivability here, then I do think that ideal conceivability entails possibility. But I don't think we have much insight into what really is ideally conceivable. When people claim that they are conceiving of a zombie, is there any reason to take their hunch seriously and believe they are conceiving of something coherent? I don't think so. (Dennett has a nice parallel: imagine someone who believes in zombikes. A zombike is physically exactly like a bike, but when you turn the pedals, the wheels don't move. It's pretty clear that there's something incoherent there, but if you had a mechanism that was more complex than a bike, you'd be harder pressed to spot the incoherence.) I think we do have independent reasons to think that consciousness at least supervenes on physics, and because I don't put much credence in the zombie intuition, I don't see why we should think zombies are ideally conceivable.
The surprise for me is that a plurality thinks aesthetic value is objective. I'm glad we're not settling these questions for all time based on first-past-the-post rules.
I am very torn about this topic. Arguments that aesthetic value is subjective (or agent-relative or something of the sort) always seem to threaten realism about moral norms, practical norms, or epistemic norms. I think philosophers are driven to objectivism about aesthetic value by considering other forms of normativity.
This study, while interesting as sociology, is methodologically the opposite of philosophy.
That would assume that philosophy has a single consistent method. It doesn't. Despite the scorn of naysayers, experimental philosophy is philosophy.
posted by painquale at 2:31 PM on May 2, 2013 [13 favorites]
Well, I was just bleating into the void, but if you want me to defend my ughs, you'll have to permit me to give some very cavalier and pithy defenses. I'm hardly going to give a compelling defense of consequentialism over deontology in a Metafilter comment. Please forgive some sloppiness.
Analyticity: Quine and others have tried to offer principled arguments against analytic truths, but I find Fodor's empirical argument best: in hundreds of years, pretty much no one has been able to come up with something that is a clear analytic truth in natural language. Examples that are offered, like "bachelors are unmarried males" or "conception causes pregnancy", have counterexamples. If there are analytic statements, they're extremely rare. (I admit that "vixens are female foxes" is pretty good.) Also, analytic statements are supposed to be different than logical truths, and I haven't seen a great explanation of the difference (without begging questions about what counts as a logical symbol).
The Apriori: I have problems with the category itself. An apriori proposition is one that could be known independently of sense experience. Known to whom? Pretty much any true proposition is something that a possible creature could be born innately believing. If you think that innate beliefs can be justified, or reliably formed, or whatever it takes for a belief to be knowledge, then it looks pretty easy to make it so that any proposition could be known innately by a possible cognizer; it would follow from the standard definition that every proposition is a priori. So, it's a bad definition. But is there a better one? Saying that propositions are a priori if we could know them innately seems to get the extension wrong entirely, because we apparently have some innate knowledge that no philosopher would want to say is a priori (e.g. about causation, about word segmentation, etc.). I don't think that there's a useful regimentation of the concept. The a priori is supposed to be different from the innate or the analytic, but I don't think there's space for such a category.
Two-boxing: Basically, two-boxing makes sense if you subscribe to causal decision theory rather than evidential decision theory. I don't see why causation is something that we should build into our epistemology, I think our intuitions about it are all over the map, and CDT seems to me to get a whole bunch of toy cases wrong compared to EDT. (I talked about CDT and EDT a little in this thread about voting.)
Correspondence Theory of Truth: The theory claims that a sentence is true if it corresponds to a fact. You have to be committed to facts for this to make sense. So what are they? For various reasons, the correspondence theorist cannot think that facts are true sentences or states of affairs. You very get into horrible questions about the individuation conditions of facts, their ontological status, etc. Is the fact that some rectangle is twice as wide as it is tall the same fact as the fact that it's half as tall as it is wide? These are awful questions. I think facts just are true sentences (or propositions). So they cannot be used to explain how it is that a sentence is true.
Deontology: Super pithy here. A consequentialist theory says that if you can do something that will make the world a better place overall, then you should do it instead of something that will make the world a less good place. That is a principle I am happy to commit myself to and I think it's kinda gross to deny it. If a deontological theory can't be "consequentialized", then it denies that principle. Ugh.
Conceivable Zombies: If we're talking about ideal conceivability here, then I do think that ideal conceivability entails possibility. But I don't think we have much insight into what really is ideally conceivable. When people claim that they are conceiving of a zombie, is there any reason to take their hunch seriously and believe they are conceiving of something coherent? I don't think so. (Dennett has a nice parallel: imagine someone who believes in zombikes. A zombike is physically exactly like a bike, but when you turn the pedals, the wheels don't move. It's pretty clear that there's something incoherent there, but if you had a mechanism that was more complex than a bike, you'd be harder pressed to spot the incoherence.) I think we do have independent reasons to think that consciousness at least supervenes on physics, and because I don't put much credence in the zombie intuition, I don't see why we should think zombies are ideally conceivable.
The surprise for me is that a plurality thinks aesthetic value is objective. I'm glad we're not settling these questions for all time based on first-past-the-post rules.
I am very torn about this topic. Arguments that aesthetic value is subjective (or agent-relative or something of the sort) always seem to threaten realism about moral norms, practical norms, or epistemic norms. I think philosophers are driven to objectivism about aesthetic value by considering other forms of normativity.
This study, while interesting as sociology, is methodologically the opposite of philosophy.
That would assume that philosophy has a single consistent method. It doesn't. Despite the scorn of naysayers, experimental philosophy is philosophy.
posted by painquale at 2:31 PM on May 2, 2013 [13 favorites]
I've talked with so many people who don't seem to understand free will that I'm beginning to wonder if there are multiple definitions out there.
Yup. Here are two of most important.
Freedom-1: Indeterministic. There are multiple possible futures. Suppose you choose option A. If we rewound history you might choose option B. Your choices determine what will come to pass. You could choose either way.
Freedom-2: Compatibilist. Your internal machinery (so to speak) determines what you will want to do. You will want to do what your genes/upbringing/personalhistory/culture/situation/randomneuronfirings/etc incline you to do. We say that you are free if nothing in your situation is forcing you to do something you don't want to do. Freedom-2 is compatible with determinism. You can't do otherwise but you're free in the sense that you wouldn't want to do otherwise.
On preview, what straight said.
how we should act if we don't have free will
Read this story/article by Ted Chiang. Some ethical theories suggest that we should try to have/spread false beliefs if false beliefs would make us happier. Chiang suggests that the suicide rate would go up if we had concrete proof that indeterminism was false.
posted by justsomebodythatyouusedtoknow at 2:48 PM on May 2, 2013 [4 favorites]
Yup. Here are two of most important.
Freedom-1: Indeterministic. There are multiple possible futures. Suppose you choose option A. If we rewound history you might choose option B. Your choices determine what will come to pass. You could choose either way.
Freedom-2: Compatibilist. Your internal machinery (so to speak) determines what you will want to do. You will want to do what your genes/upbringing/personalhistory/culture/situation/randomneuronfirings/etc incline you to do. We say that you are free if nothing in your situation is forcing you to do something you don't want to do. Freedom-2 is compatible with determinism. You can't do otherwise but you're free in the sense that you wouldn't want to do otherwise.
On preview, what straight said.
how we should act if we don't have free will
Read this story/article by Ted Chiang. Some ethical theories suggest that we should try to have/spread false beliefs if false beliefs would make us happier. Chiang suggests that the suicide rate would go up if we had concrete proof that indeterminism was false.
posted by justsomebodythatyouusedtoknow at 2:48 PM on May 2, 2013 [4 favorites]
That would assume that philosophy has a single consistent method. It doesn't.
I agree that philosophy doesn't have a consistent method. However, I think that all branches of philosophy involve arguments in some way. No arguments at all? Not philosophy.
I agree that experimental philosophy is philosophy. Here's one argument that many experimental philosophers advance.
- Many ethicists say that armchair intuitions lead us to universal moral truths.
- Empirical studies show that some intuitions vary from person to person and from culture to culture.
- Therefore my having an intuition about morality is not in itself grounds for making claims about universal moral truths.
posted by justsomebodythatyouusedtoknow at 3:01 PM on May 2, 2013
I agree that philosophy doesn't have a consistent method. However, I think that all branches of philosophy involve arguments in some way. No arguments at all? Not philosophy.
I agree that experimental philosophy is philosophy. Here's one argument that many experimental philosophers advance.
- Many ethicists say that armchair intuitions lead us to universal moral truths.
- Empirical studies show that some intuitions vary from person to person and from culture to culture.
- Therefore my having an intuition about morality is not in itself grounds for making claims about universal moral truths.
posted by justsomebodythatyouusedtoknow at 3:01 PM on May 2, 2013
Freedom-1: Indeterministic. There are multiple possible futures. Suppose you choose option A. If we rewound history you might choose option B. Your choices determine what will come to pass. You could choose either way.
It seems to me this notion of choice is incoherent. If a choice is not the action of internal machinery, then what is it and where does it come from?
posted by Golden Eternity at 3:02 PM on May 2, 2013
It seems to me this notion of choice is incoherent. If a choice is not the action of internal machinery, then what is it and where does it come from?
posted by Golden Eternity at 3:02 PM on May 2, 2013
It seems to me this notion of choice is incoherent. If a choice is not the action of internal machinery, then what is it and where does it come from?
A soul? That's a bit unfair, but seriously, Freedom-1 is hard to defend because the question you just asked is hard to answer. That's why a majority of philosophers are compatibilists.
posted by justsomebodythatyouusedtoknow at 3:10 PM on May 2, 2013
A soul? That's a bit unfair, but seriously, Freedom-1 is hard to defend because the question you just asked is hard to answer. That's why a majority of philosophers are compatibilists.
posted by justsomebodythatyouusedtoknow at 3:10 PM on May 2, 2013
(ooh thanks, I'm a big Ted Chiang fan. can't read it now, but soon)
I'm not sure I get Freedom-2 being compatible with determinism since I find what I "want" is usually very complex and can include conflicting desires. If we can't do otherwise but feel like we're making a choice or don't question our choices because we always do what we "want" then that's not free will. but it gets murky for me here, can we think we have free will even if we don't? and if so, does it matter?
posted by sineater at 3:11 PM on May 2, 2013
I'm not sure I get Freedom-2 being compatible with determinism since I find what I "want" is usually very complex and can include conflicting desires. If we can't do otherwise but feel like we're making a choice or don't question our choices because we always do what we "want" then that's not free will. but it gets murky for me here, can we think we have free will even if we don't? and if so, does it matter?
posted by sineater at 3:11 PM on May 2, 2013
sineater, an example of the second conception of freedom might clarify it. Think of two drug addicts, one who is happy to be an addict, and one who struggles constantly but ineffectively against his addiction. It seems that there is a sense in which we can say that the first is freely pursuing the activities involved in his addiction, while the second isn't. But since both are determined to act, it can't be that mere causal determination accounts for the difference. The relevant difference, put briefly, is that the first addict is doing what he wants, while the second isn't.
posted by voltairemodern at 3:51 PM on May 2, 2013 [1 favorite]
posted by voltairemodern at 3:51 PM on May 2, 2013 [1 favorite]
sineater, try turning it around: why would you choose to do something you don't want to do?
If our desires and our decision making process are entirely explained by states of affairs of the meat inside our skulls then we would expect that what we will do in future should be in principle as predictable as any process governed by the laws of physics.
That doesn't mean we're railroaded. Compatibilists aren't saying that based on your current brain state we could predict what you will choose to have for lunch on May 2, 2014. Some of what we do might best be explained in terms of random chance (e.g. neurons getting hit by cosmic rays and firing at random). However, randomness does not prove that we have freedom-1. Randomness just means that what we will choose is determined by some combination of our current desires, randomness in the way our brains work and the effects that the outside world has on our brains.
Compatibilists have no problem acknowledging that you have a decision-making process. Sure, we have conflicting desires pulling every which way. We do question our choices, in the sense that we deliberate about which desire is most important. It's also true that some of our desires are meta-desires. For example wanting to enjoy jogging more in future is a reason to go jogging now. None of that proves that we have freedom-1. If after having thought carefully about our options we don't want to do something we won't do it.
That's a bit quick and dirty but try it on for size. If the above is true, you have to be either a compatibilist (we are free whenever circumstances don't prevent us from doing what we want) or give upon the notion of free will. If.
posted by justsomebodythatyouusedtoknow at 4:08 PM on May 2, 2013
If our desires and our decision making process are entirely explained by states of affairs of the meat inside our skulls then we would expect that what we will do in future should be in principle as predictable as any process governed by the laws of physics.
That doesn't mean we're railroaded. Compatibilists aren't saying that based on your current brain state we could predict what you will choose to have for lunch on May 2, 2014. Some of what we do might best be explained in terms of random chance (e.g. neurons getting hit by cosmic rays and firing at random). However, randomness does not prove that we have freedom-1. Randomness just means that what we will choose is determined by some combination of our current desires, randomness in the way our brains work and the effects that the outside world has on our brains.
Compatibilists have no problem acknowledging that you have a decision-making process. Sure, we have conflicting desires pulling every which way. We do question our choices, in the sense that we deliberate about which desire is most important. It's also true that some of our desires are meta-desires. For example wanting to enjoy jogging more in future is a reason to go jogging now. None of that proves that we have freedom-1. If after having thought carefully about our options we don't want to do something we won't do it.
That's a bit quick and dirty but try it on for size. If the above is true, you have to be either a compatibilist (we are free whenever circumstances don't prevent us from doing what we want) or give upon the notion of free will. If.
posted by justsomebodythatyouusedtoknow at 4:08 PM on May 2, 2013
thanks for examples, but I remain confused :-) wikipedia seems to indicate the compatibilists have redefined free will so we're not really talking about the same things. for me the notion of free will has to do with whether or not we are able to actually make choices or if what feels like choosing is just an illusion. in other words, do i have any control over my decisions? can I, taking into account the current structure of my brain created by my personal history, genetics and any external influences and my current circumstances, make a decision that goes against all that? that doesn't have to do with what I want, but what is possible.
posted by sineater at 4:48 PM on May 2, 2013
posted by sineater at 4:48 PM on May 2, 2013
Arguments that aesthetic value is subjective (or agent-relative or something of the sort) always seem to threaten realism about moral norms, practical norms, or epistemic norms.
This and consequentialism seem to me like examples of philosophers losing their nerve. It seems to me that a lot of philosophical perspectives entail subjectivity of aesthetics as well as moral norms, or that determinism entails the non-existence of free will, but that a lot of philosophers just can't bring themselves to say things like "free will doesn't exist" or "morality is just an enumeration and prioritization of our preferences."
posted by straight at 4:53 PM on May 2, 2013
This and consequentialism seem to me like examples of philosophers losing their nerve. It seems to me that a lot of philosophical perspectives entail subjectivity of aesthetics as well as moral norms, or that determinism entails the non-existence of free will, but that a lot of philosophers just can't bring themselves to say things like "free will doesn't exist" or "morality is just an enumeration and prioritization of our preferences."
posted by straight at 4:53 PM on May 2, 2013
The relevant difference, put briefly, is that the first addict is doing what he wants, while the second isn't.
Except that determinism seems to entail that the wanting itself is determined. We feel that we can (sometimes) choose what to want (so-called "second-order desires") and maybe that we can chose what to want to want ("third order desires") but eventually what we want is simply a given--a product of our genes and environment and past--not something we freely choose.
I'm not a Sam Harris fan at all, but I think his statement, "I cannot take credit for the fact that I do not have the soul of a psychopath" is pretty compelling.
posted by straight at 4:59 PM on May 2, 2013
Except that determinism seems to entail that the wanting itself is determined. We feel that we can (sometimes) choose what to want (so-called "second-order desires") and maybe that we can chose what to want to want ("third order desires") but eventually what we want is simply a given--a product of our genes and environment and past--not something we freely choose.
I'm not a Sam Harris fan at all, but I think his statement, "I cannot take credit for the fact that I do not have the soul of a psychopath" is pretty compelling.
posted by straight at 4:59 PM on May 2, 2013
People who've thought seriously about it actually think that there can be realism about moral norms? I'd think that was obviously impossible, especially in a non-theistic universe, but actually even in a theistic one... you don't have to agree with the deity.
And how do you even define what "realism" is without reference to an "epistemic norm"? If epistemic norms are prior to realism, then how could you ever be realistic about them? Or am I completely misunderstanding what this means?
On the other hand, aren't "practical norms" something that you choose according to something very much like aesthetic values... so what's "practical" depends on your choice of what's "aesthetic" (in that it's desirable)? And, if so, what element of realism is lost that you haven't already lost by accepting that aesthetic value is subjective? And why should you care? Again, is there some technical term I'm missing here, or isn't a "practical norm" just a "way of bringing the physical state of the world closer to what you value"?
posted by Hizonner at 5:28 PM on May 2, 2013
And how do you even define what "realism" is without reference to an "epistemic norm"? If epistemic norms are prior to realism, then how could you ever be realistic about them? Or am I completely misunderstanding what this means?
On the other hand, aren't "practical norms" something that you choose according to something very much like aesthetic values... so what's "practical" depends on your choice of what's "aesthetic" (in that it's desirable)? And, if so, what element of realism is lost that you haven't already lost by accepting that aesthetic value is subjective? And why should you care? Again, is there some technical term I'm missing here, or isn't a "practical norm" just a "way of bringing the physical state of the world closer to what you value"?
posted by Hizonner at 5:28 PM on May 2, 2013
People who've thought seriously about it actually think that there can be realism about moral norms?
Moral realism just means thinking that moral questions have right answers. Lots of thoughtful people are realists.
even in a theistic one...you don't have to agree with the deity
Let's try a case study. Job is a good man. To settle a bet with Satan God kills Job's children and gives Job AIDS. Job thinks he did not deserve tobe treated this way by God and demands that God explain why God did that to him. God says STFU, I made the universe and who are you to question what I do? Then God heals Job because reasons.
God and Job had a difference of opinion about what if anything God had a duty to do. Are God's views and Job's views equally valid? If you say no, you might be a realist.
I say Job was right. If you disagree, you are wrong. That's not just my opinion; I can argue for my view in terms we can both understand. You could try to convince me otherwise with an argument -- go ahead and try -- but on the basis of the arguments I've grappled with so far I think I'm justified in saying Job was right. That's realism.
posted by justsomebodythatyouusedtoknow at 6:17 PM on May 2, 2013
Moral realism just means thinking that moral questions have right answers. Lots of thoughtful people are realists.
even in a theistic one...you don't have to agree with the deity
Let's try a case study. Job is a good man. To settle a bet with Satan God kills Job's children and gives Job AIDS. Job thinks he did not deserve tobe treated this way by God and demands that God explain why God did that to him. God says STFU, I made the universe and who are you to question what I do? Then God heals Job because reasons.
God and Job had a difference of opinion about what if anything God had a duty to do. Are God's views and Job's views equally valid? If you say no, you might be a realist.
I say Job was right. If you disagree, you are wrong. That's not just my opinion; I can argue for my view in terms we can both understand. You could try to convince me otherwise with an argument -- go ahead and try -- but on the basis of the arguments I've grappled with so far I think I'm justified in saying Job was right. That's realism.
posted by justsomebodythatyouusedtoknow at 6:17 PM on May 2, 2013
I, too, say Job is right, but I don't kid myself that there's anything behind that but my own personal judgement. If I tried to convince you of my view with an argument, I would make sure to couch that argument in terms of your own values, without reference to whether or not I shared them.
To support the idea of "moral realism", it wouldn't be enough for you to be able to argue me around to agreeing with you on an issue. It wouldn't even be enough if you argued me around to agreeing with you on every issue. You would still have to show me that there was some moral fact involved, not just a matter of personally assigned value.
It seems obvious to me that moral propositions are neither analytic, nor grounded in external reality... there is no fact that Job was right. There is only the fact that I don't approve of God's conduct (and I doubt most others would either, if it weren't for the "God" part).
As to "equally valid", tell me what you mean by "valid" and I'll tell you whether the two positions are equally valid.
If you mean equally valid in an epistemic sense, then, yes, they are equally valid (more like equally invalid). There is no external standard by which one can be chosen over the other.
If you mean equally valid in the sense of representing the right thing to do, then I say they're not equally valid... but the phrase "the right thing" doesn't even make sense except from inside my own point of view. You can't do moral judgement without a value system. There is nothing in the external world that can give you a value system. Therefore your value system must be internal (even if your value system was in fact created by identifiable external forces). So there can't be any kind of "objective" universal morality. My view may differ from your view, or from God's view. From my point of view, I am still right... but that says nothing at all about your point of view.
Since there's no external way for moral judgement to be universally agreed upon, it can't be "realistic". An impartial arbiter that started out with no value system would not be able to make a choice by reference to any actual facts, and sure as hell couldn't make one analytically.
I really can't see how anybody can come to any other conclusion, other than by wishful thinking. No matter how much anybody wishes for a universal, external moral standard, reality isn't obligated to actually provide one.
posted by Hizonner at 6:55 PM on May 2, 2013 [2 favorites]
To support the idea of "moral realism", it wouldn't be enough for you to be able to argue me around to agreeing with you on an issue. It wouldn't even be enough if you argued me around to agreeing with you on every issue. You would still have to show me that there was some moral fact involved, not just a matter of personally assigned value.
It seems obvious to me that moral propositions are neither analytic, nor grounded in external reality... there is no fact that Job was right. There is only the fact that I don't approve of God's conduct (and I doubt most others would either, if it weren't for the "God" part).
As to "equally valid", tell me what you mean by "valid" and I'll tell you whether the two positions are equally valid.
If you mean equally valid in an epistemic sense, then, yes, they are equally valid (more like equally invalid). There is no external standard by which one can be chosen over the other.
If you mean equally valid in the sense of representing the right thing to do, then I say they're not equally valid... but the phrase "the right thing" doesn't even make sense except from inside my own point of view. You can't do moral judgement without a value system. There is nothing in the external world that can give you a value system. Therefore your value system must be internal (even if your value system was in fact created by identifiable external forces). So there can't be any kind of "objective" universal morality. My view may differ from your view, or from God's view. From my point of view, I am still right... but that says nothing at all about your point of view.
Since there's no external way for moral judgement to be universally agreed upon, it can't be "realistic". An impartial arbiter that started out with no value system would not be able to make a choice by reference to any actual facts, and sure as hell couldn't make one analytically.
I really can't see how anybody can come to any other conclusion, other than by wishful thinking. No matter how much anybody wishes for a universal, external moral standard, reality isn't obligated to actually provide one.
posted by Hizonner at 6:55 PM on May 2, 2013 [2 favorites]
Chiang suggests that the suicide rate would go up if we had concrete proof that indeterminism was false.
All part of the plan, I suppose.
posted by AElfwine Evenstar at 7:01 PM on May 2, 2013 [1 favorite]
All part of the plan, I suppose.
posted by AElfwine Evenstar at 7:01 PM on May 2, 2013 [1 favorite]
Except that determinism seems to entail that the wanting itself is determined.
Right. But the compatibilist replies: "So what? The question of what causes me to have the desires I do just isn't relevant to whether I'm free to pursue those desires." That's what most people find infuriating about the position, but there's no logical incompatibility in the view. And it relies on a sense of freedom that seems pretty straightforward and easy to understand.
posted by voltairemodern at 7:55 PM on May 2, 2013 [1 favorite]
Right. But the compatibilist replies: "So what? The question of what causes me to have the desires I do just isn't relevant to whether I'm free to pursue those desires." That's what most people find infuriating about the position, but there's no logical incompatibility in the view. And it relies on a sense of freedom that seems pretty straightforward and easy to understand.
posted by voltairemodern at 7:55 PM on May 2, 2013 [1 favorite]
Does anyone else have this sense that a huge portion of western philosophy is just boldly, hopelessly ignorant of any sort of eastern/non-reductionist thought, and lost in its own echo chamber of abstractions?
posted by crayz at 8:18 PM on May 2, 2013
posted by crayz at 8:18 PM on May 2, 2013
People say that about western science and medicine too.
posted by painquale at 8:27 PM on May 2, 2013
posted by painquale at 8:27 PM on May 2, 2013
Define "eastern thought." I'm kinda into warring states Chinese ethics and political philosophy and I don't see any incompatibility between that and western thought. The Mencius/Xunzi debate within Confucianism is something anyone interested in neo-Aristotelianism might want to look at. Mozi is a straight up consequentialist.
posted by justsomebodythatyouusedtoknow at 8:36 PM on May 2, 2013 [1 favorite]
posted by justsomebodythatyouusedtoknow at 8:36 PM on May 2, 2013 [1 favorite]
There are many things in the sociology of philosophy that are strange, but one of the strangest is that people have strong intuitions about Newcomb's problem.
posted by LobsterMitten at 9:22 PM on May 2, 2013 [1 favorite]
posted by LobsterMitten at 9:22 PM on May 2, 2013 [1 favorite]
painquale,
Set aside natural language (which I think is probably a red herring). Why don't you think that logical truths are analytic?
Relatedly, why not think that something being an a priori truth is less about how one might come to know it and more about whether it is vulnerable to empirical refutation?
posted by Jonathan Livengood at 9:53 PM on May 2, 2013 [1 favorite]
Set aside natural language (which I think is probably a red herring). Why don't you think that logical truths are analytic?
Relatedly, why not think that something being an a priori truth is less about how one might come to know it and more about whether it is vulnerable to empirical refutation?
posted by Jonathan Livengood at 9:53 PM on May 2, 2013 [1 favorite]
@painquale: "A consequentialist theory says that if you can do something that will make the world a better place overall, then you should do it instead of something that will make the world a less good place. That is a principle I am happy to commit myself to and I think it's kinda gross to deny it."
Well, committing oneself (e.g. myself) to human rights is a pretty good reason to reject that principle, don't you think?
posted by koavf at 10:05 PM on May 2, 2013
Well, committing oneself (e.g. myself) to human rights is a pretty good reason to reject that principle, don't you think?
posted by koavf at 10:05 PM on May 2, 2013
Does your commitment to human rights make the world a better place overall? I'd certainly think so. Having consequentialist ethics doesn't mean you don't consider the consequences of broad ethical principals.
Or to turn it around - I am committed to human rights because I believe that that makes the world a better place overall. If someone could convince me otherwise, that a commitment to human rights makes the world a worse place because of xyz, that would lead me to abandon that commitment. I do not believe that any such argument is possible, mind. Are you committed to human rights qua human rights, or are you committed to that because it is the best possible thing?
posted by vibratory manner of working at 10:18 PM on May 2, 2013
Or to turn it around - I am committed to human rights because I believe that that makes the world a better place overall. If someone could convince me otherwise, that a commitment to human rights makes the world a worse place because of xyz, that would lead me to abandon that commitment. I do not believe that any such argument is possible, mind. Are you committed to human rights qua human rights, or are you committed to that because it is the best possible thing?
posted by vibratory manner of working at 10:18 PM on May 2, 2013
painquale, I'm curious about your (it appears) issue with the correspondence theory of truth.
I recently learned its name over at the Stanford Encyclopedia of Philosophy (discovered via AskMeFi, it's really good!) essay on Truth here and it's the only version of Truth that makes much sense to me. There's a discussion in the essay about the identity theory of truth, which appears to be what you're advocating? From that essay, it seems to be problematic.
I'm a complete layman when it comes to this, so I'm happy to be enlightened.
posted by But tomorrow is another day... at 2:07 AM on May 3, 2013
I recently learned its name over at the Stanford Encyclopedia of Philosophy (discovered via AskMeFi, it's really good!) essay on Truth here and it's the only version of Truth that makes much sense to me. There's a discussion in the essay about the identity theory of truth, which appears to be what you're advocating? From that essay, it seems to be problematic.
I'm a complete layman when it comes to this, so I'm happy to be enlightened.
posted by But tomorrow is another day... at 2:07 AM on May 3, 2013
Almost all of this is over my head, but a quote from the FPP has made me curious and I hope one of you more knowledgeable folks can answer me.
"The Metasurvey suggests that philosophers often have highly inaccurate sociological beliefs."
What are the inaccurate sociological beliefs referred to here?
posted by harriet vane at 3:12 AM on May 3, 2013
"The Metasurvey suggests that philosophers often have highly inaccurate sociological beliefs."
What are the inaccurate sociological beliefs referred to here?
posted by harriet vane at 3:12 AM on May 3, 2013
The inaccurate beliefs are beliefs about what most philosophers believe. For example, philosophers consistently under-estimated the number of "other" answers for each question. Philosophers surveyed also missed by fairly wide margins on several specific questions. For example, they under-estimated how many philosophers support contextualism about knowledge, non-Humean laws of nature, moral realism, scientific realism, and trolley switching.
posted by Jonathan Livengood at 7:15 AM on May 3, 2013 [1 favorite]
posted by Jonathan Livengood at 7:15 AM on May 3, 2013 [1 favorite]
A thought I had about Newcomb's problem while reading this: Do I know the rules of the game before the Predictor makes the prediction? Because if yes, then the only conceivably correct answer is Box B. Since, you know, I want $1,000,000.
posted by cthuljew at 8:42 AM on May 3, 2013
posted by cthuljew at 8:42 AM on May 3, 2013
One of the problems with consequentialist ethics is that we can't know the consequences of our actions. We might successfully predict some of the short-term consequences, but we'll be wrong some of the time, and more so the longer out we look at the consequences.
Is it right to help a kid who is choking, saving his life? What if he grows up to be a serial rapist?
So clearly people who claim to be consequentialists are in practice using some other criteria to determine their ethics. In the example above, vibratory manner of working's can't know if a personal commitment to human rights will actually make the world a better place (or whether it might perversely somehow make the world worse), and so must actually be using some other criteria for deciding to be committed to human rights.
posted by straight at 8:46 AM on May 3, 2013
Is it right to help a kid who is choking, saving his life? What if he grows up to be a serial rapist?
So clearly people who claim to be consequentialists are in practice using some other criteria to determine their ethics. In the example above, vibratory manner of working's can't know if a personal commitment to human rights will actually make the world a better place (or whether it might perversely somehow make the world worse), and so must actually be using some other criteria for deciding to be committed to human rights.
posted by straight at 8:46 AM on May 3, 2013
A thought I had about Newcomb's problem while reading this: Do I know the rules of the game before the Predictor makes the prediction? Because if yes, then the only conceivably correct answer is Box B. Since, you know, I want $1,000,000.
Yes, you know the rules beforehand.
I wonder what you think about the following thought experiment. Suppose that (1) people who sleep late are much more likely to die before the age of 50, (2) the reason that people who sleep late are likely to die young is that they have a defective gene that makes them want to sleep late and also makes them likely to die young, and (3) sleeping late is pleasurable, but not preferable to living past 50.
Should you sleep in late?
posted by Jonathan Livengood at 9:43 AM on May 3, 2013
Yes, you know the rules beforehand.
I wonder what you think about the following thought experiment. Suppose that (1) people who sleep late are much more likely to die before the age of 50, (2) the reason that people who sleep late are likely to die young is that they have a defective gene that makes them want to sleep late and also makes them likely to die young, and (3) sleeping late is pleasurable, but not preferable to living past 50.
Should you sleep in late?
posted by Jonathan Livengood at 9:43 AM on May 3, 2013
Jonathan Livengood: Relatedly, why not think that something being an a priori truth is less about how one might come to know it and more about whether it is vulnerable to empirical refutation?
Well, it depends what you mean by "vulnerable to empirical refutation." If you mean that we couldn't overturn it experimentally, then the short of it is that I think this would make every proposition a priori. Any proposition could be overturned if we decided to revise our logic, and a battery of results could conceivably cause us to do so. I don't think the quantum logicians were right in thinking that quantum mechanics told us we needed to jettison the law of distribution, but their arguments had the right form. On the other hand, if when you say that a proposition isn't vulnerable to empirical refutation, you mean that we *won't* expose it to refutation (not just that we couldn't), then I think you end up including all sorts of more mundane empirical propositions that philosophers would not want to call a priori.
I accept that there's a sliding scale between propositions that we protect from revision and those that we are less enamored of, but I don't think this maps very nicely to the a priori/a posteriori division. There's no such thing as a logical truth independent of a specified logic, and it's an empirical issue which logic applies or is to be used in a certain domain.
Set aside natural language (which I think is probably a red herring). Why don't you think that logical truths are analytic?
Which logic? In tense logic, various statements about time will be logically true. In Hintikka's epistemic logic, the KK principle holds. I can gerrymander an artificial language right now in which 'all bachelors are unmarried males' is a logical truth. Are these analytically true? You can say that they're analytic within that formal system, but I think that misses the point.
I don't think that the focus on natural language is a red herring. Whenever people offer examples of analytic statements, or use analyticity in philosophical argument, they're employing natural language. It's always been presented as a property of language as we speak it or thought as we think it. You could probably define a formal notion of analyticity that applies to formal languages and I'd be OK with it, but it wouldn't be the thing that philosophers have been talking about. (In Two Dogmas, Quine allows that in an artificial language you can stipulate that two terms are substitutionally synonymous. He isn't concerned with that notion of synonymy.)
You can also say that a sentence is analytic if its translation into our currently best preferred logic is a logical truth. I'm fine with this notion. But I note that you only get logical truths from this definition, and analyticity is normally thought to be something different from logical truth. And I think every skeptic about analyticity will be OK with this notion.
But tomorrow is another day...: painquale, I'm curious about your (it appears) issue with the correspondence theory of truth. I recently learned its name over at the Stanford Encyclopedia of Philosophy (discovered via AskMeFi, it's really good!) essay on Truth here and it's the only version of Truth that makes much sense to me. There's a discussion in the essay about the identity theory of truth, which appears to be what you're advocating? From that essay, it seems to be problematic.
It looks like Moore and Russell abandoned their identity theory because it posited propositions, and they came to disbelieve in propositions. Philosophers today aren't quite so squeamish about allowing propositions into their ontology (I'm not). Current theories of propositions don't have problems with negative propositions or the unity of the proposition, as Russell did.
But I didn't mean to be proposing an identity theory. I just wanted to show that it's weird to have facts as a sui generis category in your ontology. When I said that I think that facts are true sentences or propositions, I was just giving an off-hand account of the sort of things that I think facts are. I didn't mean to suggest that a theory of truth will rely on facts (which Russell and Moore's identity theory does try to do).
I tend to just be a deflationist about truth. The sentence "Snow is white" is true if and only if snow is white. I don't need to appeal to facts to make that claim.
straight: One of the problems with consequentialist ethics is that we can't know the consequences of our actions. We might successfully predict some of the short-term consequences, but we'll be wrong some of the time, and more so the longer out we look at the consequences.
You make it sound like we have no idea whether our actions will be constructive or destructive. If that were the case, then we'd be well and truly screwed. But we do the best we can, and try to do what we have reason to think will maximize good in the world. We're not totally clueless. We might get it wrong, but such is the danger of living. If you really thought we had no insight into the consequences of our actions, pretty much all ethical theories would be dead in the water.
I bet I can turn your favorite ethical theory into a consequentialist theory. Suppose you think that infringing on another person's right to privacy is a harm. That is not intrinsically deontological. A consequentialist might hold that a future in which someone's privacy is infringed on is a worse outcome than one in which their privacy is upheld. So, you shouldn't take actions that would result in their infringed privacy, if you can help it. That action would be worse than one which doesn't have that consequence.
You can start doing this with whatever rights you care about. It might turn out you were consequentialist all along! Consequentialism does not always just have to be about maximizing pleasure. You can maximize things like human dignity, etc.
posted by painquale at 11:14 AM on May 3, 2013
Well, it depends what you mean by "vulnerable to empirical refutation." If you mean that we couldn't overturn it experimentally, then the short of it is that I think this would make every proposition a priori. Any proposition could be overturned if we decided to revise our logic, and a battery of results could conceivably cause us to do so. I don't think the quantum logicians were right in thinking that quantum mechanics told us we needed to jettison the law of distribution, but their arguments had the right form. On the other hand, if when you say that a proposition isn't vulnerable to empirical refutation, you mean that we *won't* expose it to refutation (not just that we couldn't), then I think you end up including all sorts of more mundane empirical propositions that philosophers would not want to call a priori.
I accept that there's a sliding scale between propositions that we protect from revision and those that we are less enamored of, but I don't think this maps very nicely to the a priori/a posteriori division. There's no such thing as a logical truth independent of a specified logic, and it's an empirical issue which logic applies or is to be used in a certain domain.
Set aside natural language (which I think is probably a red herring). Why don't you think that logical truths are analytic?
Which logic? In tense logic, various statements about time will be logically true. In Hintikka's epistemic logic, the KK principle holds. I can gerrymander an artificial language right now in which 'all bachelors are unmarried males' is a logical truth. Are these analytically true? You can say that they're analytic within that formal system, but I think that misses the point.
I don't think that the focus on natural language is a red herring. Whenever people offer examples of analytic statements, or use analyticity in philosophical argument, they're employing natural language. It's always been presented as a property of language as we speak it or thought as we think it. You could probably define a formal notion of analyticity that applies to formal languages and I'd be OK with it, but it wouldn't be the thing that philosophers have been talking about. (In Two Dogmas, Quine allows that in an artificial language you can stipulate that two terms are substitutionally synonymous. He isn't concerned with that notion of synonymy.)
You can also say that a sentence is analytic if its translation into our currently best preferred logic is a logical truth. I'm fine with this notion. But I note that you only get logical truths from this definition, and analyticity is normally thought to be something different from logical truth. And I think every skeptic about analyticity will be OK with this notion.
But tomorrow is another day...: painquale, I'm curious about your (it appears) issue with the correspondence theory of truth. I recently learned its name over at the Stanford Encyclopedia of Philosophy (discovered via AskMeFi, it's really good!) essay on Truth here and it's the only version of Truth that makes much sense to me. There's a discussion in the essay about the identity theory of truth, which appears to be what you're advocating? From that essay, it seems to be problematic.
It looks like Moore and Russell abandoned their identity theory because it posited propositions, and they came to disbelieve in propositions. Philosophers today aren't quite so squeamish about allowing propositions into their ontology (I'm not). Current theories of propositions don't have problems with negative propositions or the unity of the proposition, as Russell did.
But I didn't mean to be proposing an identity theory. I just wanted to show that it's weird to have facts as a sui generis category in your ontology. When I said that I think that facts are true sentences or propositions, I was just giving an off-hand account of the sort of things that I think facts are. I didn't mean to suggest that a theory of truth will rely on facts (which Russell and Moore's identity theory does try to do).
I tend to just be a deflationist about truth. The sentence "Snow is white" is true if and only if snow is white. I don't need to appeal to facts to make that claim.
straight: One of the problems with consequentialist ethics is that we can't know the consequences of our actions. We might successfully predict some of the short-term consequences, but we'll be wrong some of the time, and more so the longer out we look at the consequences.
You make it sound like we have no idea whether our actions will be constructive or destructive. If that were the case, then we'd be well and truly screwed. But we do the best we can, and try to do what we have reason to think will maximize good in the world. We're not totally clueless. We might get it wrong, but such is the danger of living. If you really thought we had no insight into the consequences of our actions, pretty much all ethical theories would be dead in the water.
I bet I can turn your favorite ethical theory into a consequentialist theory. Suppose you think that infringing on another person's right to privacy is a harm. That is not intrinsically deontological. A consequentialist might hold that a future in which someone's privacy is infringed on is a worse outcome than one in which their privacy is upheld. So, you shouldn't take actions that would result in their infringed privacy, if you can help it. That action would be worse than one which doesn't have that consequence.
You can start doing this with whatever rights you care about. It might turn out you were consequentialist all along! Consequentialism does not always just have to be about maximizing pleasure. You can maximize things like human dignity, etc.
posted by painquale at 11:14 AM on May 3, 2013
Shoot. My second sentence: "If you mean that we couldn't overturn it experimentally, then the short of it is that I think this would make every proposition a priori." Change that to: "it would make no proposition a priori."
posted by painquale at 11:47 AM on May 3, 2013
posted by painquale at 11:47 AM on May 3, 2013
painquale, I think if you push that too hard you just end up being trivially, tautologically correct. You might as well just say "Deonotlogists believe a world where people choose to act based on duty is better (by definition) than a world where people choose to act based on consequences, therefore deontologists who act based on duty are actually consequentialists."
posted by straight at 1:28 PM on May 3, 2013
posted by straight at 1:28 PM on May 3, 2013
But the compatibilist replies: "So what? The question of what causes me to have the desires I do just isn't relevant to whether I'm free to pursue those desires."
If that were widely accepted, there would be no such thing as the insanity defense.
We commonly accept that some people are not responsible for their actions. But I don't see how a compatibilist can differentiate between desires caused by "insanity" and desires caused by other factors in a person's biology, history, and environment.
posted by straight at 2:07 PM on May 3, 2013
If that were widely accepted, there would be no such thing as the insanity defense.
We commonly accept that some people are not responsible for their actions. But I don't see how a compatibilist can differentiate between desires caused by "insanity" and desires caused by other factors in a person's biology, history, and environment.
posted by straight at 2:07 PM on May 3, 2013
Well, there are some deontological theories that you can't consequentialize that way. A theory that allows for moral tragedies -- "Sophie's Choice" situations in which all of your options are things that you shouldn't do -- can't be consequentialized, for instance. And plenty of other deontological theories that can be consequentialized look really mutilated when you do so.
But you do get other benefits from thinking about things this way. Firstly, it blocks the sorts of arguments that you're trying to make against all consequentialist theories (unless you're also willing to accept that they are arguments against deontological theories that can be consequentialized). Secondly, it allows us to apply resources from decision theory in thinking about what to do. Thirdly, it focusses our attention on comparing various decisions to one another rather than making categorical judgments about what's permissible and impermissible, which is more of a help in most applied ethics cases. Fourthly, it provides a basic framework for further debates about what the values that we're trying to maximize should be.
Really, my point for bringing this up was to make consequentialism look less crazy to you. Although I think that even hedonic utilitarians can argue for human rights, it's even easier for those consequentialists who build various rights into the values that they are trying to maximize. And there are plenty of consequentialists who do this ("objective list utilitarians"). I think this should give pause to the more easy attacks against consequentialism.
posted by painquale at 2:08 PM on May 3, 2013
But you do get other benefits from thinking about things this way. Firstly, it blocks the sorts of arguments that you're trying to make against all consequentialist theories (unless you're also willing to accept that they are arguments against deontological theories that can be consequentialized). Secondly, it allows us to apply resources from decision theory in thinking about what to do. Thirdly, it focusses our attention on comparing various decisions to one another rather than making categorical judgments about what's permissible and impermissible, which is more of a help in most applied ethics cases. Fourthly, it provides a basic framework for further debates about what the values that we're trying to maximize should be.
Really, my point for bringing this up was to make consequentialism look less crazy to you. Although I think that even hedonic utilitarians can argue for human rights, it's even easier for those consequentialists who build various rights into the values that they are trying to maximize. And there are plenty of consequentialists who do this ("objective list utilitarians"). I think this should give pause to the more easy attacks against consequentialism.
posted by painquale at 2:08 PM on May 3, 2013
My main issue with consequentialism in practice is when people seem to have unwarranted confidence in what the consequences are. Particularly when people try to argue that a given situation is like a trolley problem when in fact it seems there are actually more than two options and the consequences of each option are less certain than is claimed. (In fact, I think that's true of the trolley problem itself as classically formulated.)
posted by straight at 2:29 PM on May 3, 2013
posted by straight at 2:29 PM on May 3, 2013
My main issue with consequentialism in practice is when people seem to have unwarranted confidence in what the consequences are.
Some consequentialists draw a distinction between objective and subjective consequentialism. Objective consequentialists evaluate your actions based on the effects they actually have. Subjective consequences evaluate actions on the basis of what you reasonably ought to have expected the consequences of your actions to be.
Subjective consequentialism: throwing a brick into a crowd of people is morally wrong even if by dumb luck you happen to knock out a terrorist.
posted by justsomebodythatyouusedtoknow at 3:15 PM on May 3, 2013
Some consequentialists draw a distinction between objective and subjective consequentialism. Objective consequentialists evaluate your actions based on the effects they actually have. Subjective consequences evaluate actions on the basis of what you reasonably ought to have expected the consequences of your actions to be.
Subjective consequentialism: throwing a brick into a crowd of people is morally wrong even if by dumb luck you happen to knock out a terrorist.
posted by justsomebodythatyouusedtoknow at 3:15 PM on May 3, 2013
painquale: Thanks for the clarification.
I'm afraid I don't understand the distinction you've made between "Snow is white" and snow is white. It seems tautological, and I'm not clear on how you're not appealing to facts here in assessing the truth of the statement "Snow is white".
I suspect enlightenment will only come from my doing more reading on the topic so that I have the tools to think and talk about this more precisely and succinctly.
posted by But tomorrow is another day... at 3:31 PM on May 3, 2013
I'm afraid I don't understand the distinction you've made between "Snow is white" and snow is white. It seems tautological, and I'm not clear on how you're not appealing to facts here in assessing the truth of the statement "Snow is white".
I suspect enlightenment will only come from my doing more reading on the topic so that I have the tools to think and talk about this more precisely and succinctly.
posted by But tomorrow is another day... at 3:31 PM on May 3, 2013
It's an empirical issue which logic applies or is to be used in a certain domain.
Even if one agrees with this, it doesn't touch a whole raft of claims about what sentences are related to what other sentences by this or that consequence relation.
Which logic? In tense logic, various statements about time will be logically true. In Hintikka's epistemic logic, the KK principle holds. I can gerrymander an artificial language right now in which 'all bachelors are unmarried males' is a logical truth. Are these analytically true?
Yes.
You can say that they're analytic within that formal system, but I think that misses the point.
That's certainly the line Quine takes in Two Dogmas, but I think that line is ultimately kind of silly. It's like saying, "I know what it's like for water to boil on my stove and on your stove and any other stove you care to name, but I have no idea what it's like for water to boil." I just want to give the incredulous stare here.
I don't think that the focus on natural language is a red herring. Whenever people offer examples of analytic statements, or use analyticity in philosophical argument, they're employing natural language.
Okay, let me take English a bit more seriously for a second. I'll stick my neck out and say that all of the following sentences are analytic in English.
* Green grass is green.
* If red and green were the only colors, then everything colored would be red and/or green.
* If Bucky is taller than Silvia and Silvia is taller than Joe, then Bucky is taller than Joe.
* Nothing is heavier than itself.
What moves do you want to make here and why?
You could probably define a formal notion of analyticity that applies to formal languages and I'd be OK with it, but it wouldn't be the thing that philosophers have been talking about.
Well, that seems a bit contentious. ;)
You can also say that a sentence is analytic if its translation into our currently best preferred logic is a logical truth. I'm fine with this notion. But I note that you only get logical truths from this definition, and analyticity is normally thought to be something different from logical truth. And I think every skeptic about analyticity will be OK with this notion.
Again, this is something I don't get. When I say that there is an analytic/synthetic distinction, I mean roughly that there are some sentences that are true in virtue of their forms and/or in virtue of the meanings of their constituent terms. That notion includes logical truths if it includes anything. It might include more besides. But if you grant the first, then you grant the analytic/synthetic distinction in some form. You just aren't so sure there are what we might call "extra-logical" analytic sentences.
Fun! Real philosophy on metafilter!
posted by Jonathan Livengood at 6:52 PM on May 3, 2013 [1 favorite]
Even if one agrees with this, it doesn't touch a whole raft of claims about what sentences are related to what other sentences by this or that consequence relation.
Which logic? In tense logic, various statements about time will be logically true. In Hintikka's epistemic logic, the KK principle holds. I can gerrymander an artificial language right now in which 'all bachelors are unmarried males' is a logical truth. Are these analytically true?
Yes.
You can say that they're analytic within that formal system, but I think that misses the point.
That's certainly the line Quine takes in Two Dogmas, but I think that line is ultimately kind of silly. It's like saying, "I know what it's like for water to boil on my stove and on your stove and any other stove you care to name, but I have no idea what it's like for water to boil." I just want to give the incredulous stare here.
I don't think that the focus on natural language is a red herring. Whenever people offer examples of analytic statements, or use analyticity in philosophical argument, they're employing natural language.
Okay, let me take English a bit more seriously for a second. I'll stick my neck out and say that all of the following sentences are analytic in English.
* Green grass is green.
* If red and green were the only colors, then everything colored would be red and/or green.
* If Bucky is taller than Silvia and Silvia is taller than Joe, then Bucky is taller than Joe.
* Nothing is heavier than itself.
What moves do you want to make here and why?
You could probably define a formal notion of analyticity that applies to formal languages and I'd be OK with it, but it wouldn't be the thing that philosophers have been talking about.
Well, that seems a bit contentious. ;)
You can also say that a sentence is analytic if its translation into our currently best preferred logic is a logical truth. I'm fine with this notion. But I note that you only get logical truths from this definition, and analyticity is normally thought to be something different from logical truth. And I think every skeptic about analyticity will be OK with this notion.
Again, this is something I don't get. When I say that there is an analytic/synthetic distinction, I mean roughly that there are some sentences that are true in virtue of their forms and/or in virtue of the meanings of their constituent terms. That notion includes logical truths if it includes anything. It might include more besides. But if you grant the first, then you grant the analytic/synthetic distinction in some form. You just aren't so sure there are what we might call "extra-logical" analytic sentences.
Fun! Real philosophy on metafilter!
posted by Jonathan Livengood at 6:52 PM on May 3, 2013 [1 favorite]
Again, this is something I don't get. When I say that there is an analytic/synthetic distinction, I mean roughly that there are some sentences that are true in virtue of their forms and/or in virtue of the meanings of their constituent terms. That notion includes logical truths if it includes anything.
This is tricky because I think that there are some distinctions that are close to traditional formulations of the analytic/synthetic distinction, so a rough characterization might well have me agreeing with you. The term 'analytic' might be so indeterminate that some precisifications would be ones that everyone would agree with. So, to some extent, maybe there aren't many disputes of substance here. Maybe the distinctions I think are genuine are ones you think deserve to be called variants of the analytic/synthetic distinction.
But I do think that we'll find disputes of substance here. I reject the characterizations of analyticity that you mention. I don't think it's quite right to say that analytic statements are true in virtue of their constituent meanings: all sentences are. Analytic statements are true solely in virtue of the meanings of their terms. Let's take this as our definition.
Here is my claim: there are no analytic statements of this sort in non-artificial languages. (I'm not going to make a claim about artificial languages.) Confirmation or disconfirmation of any statement in a non-artificial language can come from anywhere. As Quine puts it in Philosophy of Logic, even logical connectives are true not solely in virtue of their meaning, but "in virtue of anything and everything." No sentence is true come what may.
Before addressing the sentences that you present, I want to clarify my position on logical truths. You write:
if you grant [there are logical truths], then you grant the analytic/synthetic distinction in some form. You just aren't so sure there are what we might call "extra-logical" analytic sentences.
The Fodorian argument I made at first would suggest that I think this. But I was just giving a quick and pithy argument. I do want to go further and claim that even logical truths are not analytic.
There's a really fractured and really interesting history about whether logical truths are analytic. It's mostly due to Quine being entirely unable to make up his mind. In Two Dogmas, Quine defined analyticity as a disjunction, and attacked only the 'Fregean sense' of analyticity (which depended on a notion of synonymy), leaving the 'logical truth' half of the definition untouched. On the other hand, he also claims in that paper that logic is revisable and that logical truths aren't true in virtue of the meaning of the connectives. From then on, he was pretty inconsistent about the topic.
The real problem was that Quine was a conservative about first-order logic. He recognized that his anti-analyticity line in Two Dogmas would give deviant logicians ammo, so he ended up arguing that they "changed the subject" when they used their deviant connectives (Philosophy of Logic), and that "prelogical people" (that is, people who didn't use FOL) were a myth of bad translation (Word and Object). He thought he could get away with this by arguing that the sentential connectives could be identified and translated non-holistically, solely in virtue of peoples' assent and dissent to sentences, which he thought were unproblematic behaviorist cues. (For various reasons, this does not work.)
The upshot is that although Quine actually did try to find a wedge between the logical connectives and all other words, he wasn't able to do it. I think that if you're tempted to reject that non-logical vocabulary can contribute to the analyticity of a statement (as I am), then you should probably go whole hog and make that claim about logical vocabulary as well.
Okay, let me take English a bit more seriously for a second. I'll stick my neck out and say that all of the following sentences are analytic in English. ... What moves do you want to make here and why?
OK, this is tough... I think that all the sentences are true, that they are all very unlikely to be disconfirmed by empirical evidence, that they all suggest fairly ready translations, and that 'green grass is green' would be a logical truth in the logic it is best translated into.
However, I dispute that they have determinate meanings that could render them impervious to possible empirical counterargument. Every sentence has various precisifications of its indeterminacy -- various translation manuals. (A translation manual will specify not just the extension of the terms, but also the logic it is translated into, the consequence relations, etc.) There are many coexisting manuals, but we prefer some over others. I claim that there is always some possible finding -- some possible empirical evidence, or maybe some theoretical work that systematizes our evidence -- that will cause some precisifications to make the sentence come out false, and moreover, cause us to prefer those precisifications.
Let me demonstrate my strategy by attacking two other sentences that you might have written if we had lived long ago: "there are no round squares" and "there is at most one straight line between any two points." These sentences have seemed conceptually true to many, and they have said so in writing. We now know that both are hostage to Euclidean geometry. There are round squares in taxicab geometry, and there is more than one straight line between two points in elliptical geometry.
It turns out that before we had considered it, 'straight line' had various indeterminate intensions. It could have been synonymous with 'geodesic', in which case the sentence "there is at most one straight line between any two points" was false, or it could have been synonymous with something like 'straight line in Euclidean space', in which case it'd be true. (I once had a debate with a philosopher who tried to argue that there are no straight lines in a curved spacetime.) Which of those did it really mean? That seems like a bad question to me. Why suppose that it needed to determinately mean one of them?
Empirical and theoretical considerations caused mathematicians to prefer translation manuals that favor the geodesic interpretation. However, imagine that our universe really had been Euclidean. They might well have gone the other route. It was an empirical matter. And it's conceivable that we'll decide that we were wrong about the structure of the universe and precisify the word differently once again. I claim that is there wasn't a fact of the matter what the word meant before mathematicians cleaned their language up, and there still isn't. And there isn't ever a single determinately preferred translation manual.
I think that what happened with "there are no round squares" could happen with all of the sentences you mention. Various empirical and theoretical considerations could cause us to prefer translation manuals that would cause the sentence to come out false. For instance, various considerations could cause us to reject FOL and become intuitionists across the board, making it so that a bunch of supposed logical truths ended up being empirically disconfirmed. That's the general idea. The problem is that disconfirming your sentences would require really radical revisions to our conceptual scheme, and they're pretty hard to make plausible. I'm going to have to just hint at how they could be disconfirmed. (And I'm going to punt on the one about nothing being heavier than itself.)
* If red and green were the only colors, then everything colored would be red and/or green.
This one's pretty easy. Philosophical considerations about the nature of colors could render this false. Two philosophers are arguing about the metaphysics of color. Philosopher Alpha includes colors in his ontology. When an object is colored, it bears an instantiation relation to one of these colors. Philosopher Beta is an adverbialist. She does not admit colors into her ontology. However, she allows that there are different ways that things can be colored. An object can be colored bluishly, reddishly, etc. Philosopher Delta comes in, and trying to be conciliatory, splits the difference. Red and green are the only colors! Nevertheless, he's an adverbialist with respect to blue and yellow and purple, so he still thinks things are colored blue and yellow and purple. A dumb position, but not self-contradictory, and he would deny the statement that you say is analytic.
* If Bucky is taller than Silvia and Silvia is taller than Joe, then Bucky is taller than Joe.
I'm uncertain about this, but this very much looks to me like it depends on certain assumptions about geometrical space. It looks to me like "If Bucky is parallel to Silvia and Silvia is parallel to Joe, then Bucky is parallel to Joe", which is equivalent ot the parallel postulate. I can't think of one off the top of my head, but I am really willing to bet that there are crazy geometries or affine spaces in which a natural translation of 'taller than' is not transitive. Part of the problem is that 'taller than' is difficult to precisify. There are certainly space in which AB is less than AC and AC is less than AD but AD is less than AB, but I'm not sure that captures the logic of 'taller than'. I also started playing around with the logic of 'taller than' when you consider relativistic time dilation, but I gave up. Let's just say that this looks like a straightfowardly geometrical relation, and I bet you could construct a geometry in which it does not hold.
* Green grass is green.
This one is pretty brutal. Here are a couple of possibilities. 1) Imagine that we kill off all green grass so that no green grass exists. Then, suppose that we find it scientifically advantageous to go back to Aristotelian logic (I have no idea why). Now you have a universal statement about a non-existent, which according to Aristotle was false.
2) I note that you didn't use the syntactically analagous sentence "red meat is red." Because not all red meat is red. One translation of the sentence is that "green grass" functions for us like a structured name... a name for our kind of grass. Suppose we find some other planet with blue grass. Now everyone starts talking about green grass and blue grass. At some point, all our grass goes yellow. Now green grass is not green. It's yellow. I suspect you might think that you think I've swapped out your sentence for one with a different meaning... one in which 'green grass' functions as a name. But my claim is that it's the very same English sentence, and that reading 'green grass' as a name is just one of many viable translations. It's just one that we currently don't find useful. (The fact that the sentence has indeterminate readings explains how linguistic evolution and drift is possible. 'Red meat' was a description at one point, and our translation preferences gradually shifted over time to reading it as a name.)
Fun! Real philosophy on metafilter!
Yup! Oh, one last thing:
That's certainly the line Quine takes in Two Dogmas, but I think that line is ultimately kind of silly. It's like saying, "I know what it's like for water to boil on my stove and on your stove and any other stove you care to name, but I have no idea what it's like for water to boil." I just want to give the incredulous stare here.
Compare: "You say you know how much this bowling ball weighs on Earth, and you know how much it weighs on the moon, but you really have no idea how much the bowling ball just weighs? I want to give the incredulous stare." I didn't expect you to go the way you did. I thought you'd say that it's fine for there to be a whole bunch of different notions of analytic-in-L; we don't need a notion of analyticity simpliciter.
posted by painquale at 5:19 PM on May 4, 2013 [3 favorites]
This is tricky because I think that there are some distinctions that are close to traditional formulations of the analytic/synthetic distinction, so a rough characterization might well have me agreeing with you. The term 'analytic' might be so indeterminate that some precisifications would be ones that everyone would agree with. So, to some extent, maybe there aren't many disputes of substance here. Maybe the distinctions I think are genuine are ones you think deserve to be called variants of the analytic/synthetic distinction.
But I do think that we'll find disputes of substance here. I reject the characterizations of analyticity that you mention. I don't think it's quite right to say that analytic statements are true in virtue of their constituent meanings: all sentences are. Analytic statements are true solely in virtue of the meanings of their terms. Let's take this as our definition.
Here is my claim: there are no analytic statements of this sort in non-artificial languages. (I'm not going to make a claim about artificial languages.) Confirmation or disconfirmation of any statement in a non-artificial language can come from anywhere. As Quine puts it in Philosophy of Logic, even logical connectives are true not solely in virtue of their meaning, but "in virtue of anything and everything." No sentence is true come what may.
Before addressing the sentences that you present, I want to clarify my position on logical truths. You write:
if you grant [there are logical truths], then you grant the analytic/synthetic distinction in some form. You just aren't so sure there are what we might call "extra-logical" analytic sentences.
The Fodorian argument I made at first would suggest that I think this. But I was just giving a quick and pithy argument. I do want to go further and claim that even logical truths are not analytic.
There's a really fractured and really interesting history about whether logical truths are analytic. It's mostly due to Quine being entirely unable to make up his mind. In Two Dogmas, Quine defined analyticity as a disjunction, and attacked only the 'Fregean sense' of analyticity (which depended on a notion of synonymy), leaving the 'logical truth' half of the definition untouched. On the other hand, he also claims in that paper that logic is revisable and that logical truths aren't true in virtue of the meaning of the connectives. From then on, he was pretty inconsistent about the topic.
The real problem was that Quine was a conservative about first-order logic. He recognized that his anti-analyticity line in Two Dogmas would give deviant logicians ammo, so he ended up arguing that they "changed the subject" when they used their deviant connectives (Philosophy of Logic), and that "prelogical people" (that is, people who didn't use FOL) were a myth of bad translation (Word and Object). He thought he could get away with this by arguing that the sentential connectives could be identified and translated non-holistically, solely in virtue of peoples' assent and dissent to sentences, which he thought were unproblematic behaviorist cues. (For various reasons, this does not work.)
The upshot is that although Quine actually did try to find a wedge between the logical connectives and all other words, he wasn't able to do it. I think that if you're tempted to reject that non-logical vocabulary can contribute to the analyticity of a statement (as I am), then you should probably go whole hog and make that claim about logical vocabulary as well.
Okay, let me take English a bit more seriously for a second. I'll stick my neck out and say that all of the following sentences are analytic in English. ... What moves do you want to make here and why?
OK, this is tough... I think that all the sentences are true, that they are all very unlikely to be disconfirmed by empirical evidence, that they all suggest fairly ready translations, and that 'green grass is green' would be a logical truth in the logic it is best translated into.
However, I dispute that they have determinate meanings that could render them impervious to possible empirical counterargument. Every sentence has various precisifications of its indeterminacy -- various translation manuals. (A translation manual will specify not just the extension of the terms, but also the logic it is translated into, the consequence relations, etc.) There are many coexisting manuals, but we prefer some over others. I claim that there is always some possible finding -- some possible empirical evidence, or maybe some theoretical work that systematizes our evidence -- that will cause some precisifications to make the sentence come out false, and moreover, cause us to prefer those precisifications.
Let me demonstrate my strategy by attacking two other sentences that you might have written if we had lived long ago: "there are no round squares" and "there is at most one straight line between any two points." These sentences have seemed conceptually true to many, and they have said so in writing. We now know that both are hostage to Euclidean geometry. There are round squares in taxicab geometry, and there is more than one straight line between two points in elliptical geometry.
It turns out that before we had considered it, 'straight line' had various indeterminate intensions. It could have been synonymous with 'geodesic', in which case the sentence "there is at most one straight line between any two points" was false, or it could have been synonymous with something like 'straight line in Euclidean space', in which case it'd be true. (I once had a debate with a philosopher who tried to argue that there are no straight lines in a curved spacetime.) Which of those did it really mean? That seems like a bad question to me. Why suppose that it needed to determinately mean one of them?
Empirical and theoretical considerations caused mathematicians to prefer translation manuals that favor the geodesic interpretation. However, imagine that our universe really had been Euclidean. They might well have gone the other route. It was an empirical matter. And it's conceivable that we'll decide that we were wrong about the structure of the universe and precisify the word differently once again. I claim that is there wasn't a fact of the matter what the word meant before mathematicians cleaned their language up, and there still isn't. And there isn't ever a single determinately preferred translation manual.
I think that what happened with "there are no round squares" could happen with all of the sentences you mention. Various empirical and theoretical considerations could cause us to prefer translation manuals that would cause the sentence to come out false. For instance, various considerations could cause us to reject FOL and become intuitionists across the board, making it so that a bunch of supposed logical truths ended up being empirically disconfirmed. That's the general idea. The problem is that disconfirming your sentences would require really radical revisions to our conceptual scheme, and they're pretty hard to make plausible. I'm going to have to just hint at how they could be disconfirmed. (And I'm going to punt on the one about nothing being heavier than itself.)
* If red and green were the only colors, then everything colored would be red and/or green.
This one's pretty easy. Philosophical considerations about the nature of colors could render this false. Two philosophers are arguing about the metaphysics of color. Philosopher Alpha includes colors in his ontology. When an object is colored, it bears an instantiation relation to one of these colors. Philosopher Beta is an adverbialist. She does not admit colors into her ontology. However, she allows that there are different ways that things can be colored. An object can be colored bluishly, reddishly, etc. Philosopher Delta comes in, and trying to be conciliatory, splits the difference. Red and green are the only colors! Nevertheless, he's an adverbialist with respect to blue and yellow and purple, so he still thinks things are colored blue and yellow and purple. A dumb position, but not self-contradictory, and he would deny the statement that you say is analytic.
* If Bucky is taller than Silvia and Silvia is taller than Joe, then Bucky is taller than Joe.
I'm uncertain about this, but this very much looks to me like it depends on certain assumptions about geometrical space. It looks to me like "If Bucky is parallel to Silvia and Silvia is parallel to Joe, then Bucky is parallel to Joe", which is equivalent ot the parallel postulate. I can't think of one off the top of my head, but I am really willing to bet that there are crazy geometries or affine spaces in which a natural translation of 'taller than' is not transitive. Part of the problem is that 'taller than' is difficult to precisify. There are certainly space in which AB is less than AC and AC is less than AD but AD is less than AB, but I'm not sure that captures the logic of 'taller than'. I also started playing around with the logic of 'taller than' when you consider relativistic time dilation, but I gave up. Let's just say that this looks like a straightfowardly geometrical relation, and I bet you could construct a geometry in which it does not hold.
* Green grass is green.
This one is pretty brutal. Here are a couple of possibilities. 1) Imagine that we kill off all green grass so that no green grass exists. Then, suppose that we find it scientifically advantageous to go back to Aristotelian logic (I have no idea why). Now you have a universal statement about a non-existent, which according to Aristotle was false.
2) I note that you didn't use the syntactically analagous sentence "red meat is red." Because not all red meat is red. One translation of the sentence is that "green grass" functions for us like a structured name... a name for our kind of grass. Suppose we find some other planet with blue grass. Now everyone starts talking about green grass and blue grass. At some point, all our grass goes yellow. Now green grass is not green. It's yellow. I suspect you might think that you think I've swapped out your sentence for one with a different meaning... one in which 'green grass' functions as a name. But my claim is that it's the very same English sentence, and that reading 'green grass' as a name is just one of many viable translations. It's just one that we currently don't find useful. (The fact that the sentence has indeterminate readings explains how linguistic evolution and drift is possible. 'Red meat' was a description at one point, and our translation preferences gradually shifted over time to reading it as a name.)
Fun! Real philosophy on metafilter!
Yup! Oh, one last thing:
That's certainly the line Quine takes in Two Dogmas, but I think that line is ultimately kind of silly. It's like saying, "I know what it's like for water to boil on my stove and on your stove and any other stove you care to name, but I have no idea what it's like for water to boil." I just want to give the incredulous stare here.
Compare: "You say you know how much this bowling ball weighs on Earth, and you know how much it weighs on the moon, but you really have no idea how much the bowling ball just weighs? I want to give the incredulous stare." I didn't expect you to go the way you did. I thought you'd say that it's fine for there to be a whole bunch of different notions of analytic-in-L; we don't need a notion of analyticity simpliciter.
posted by painquale at 5:19 PM on May 4, 2013 [3 favorites]
Scumbag analytic philosopher looks very vaguely like me. I'm considering making him my Halloween costume.
posted by painquale at 3:03 PM on May 8, 2013 [1 favorite]
posted by painquale at 3:03 PM on May 8, 2013 [1 favorite]
Sorry to take so long in getting back to this. End-of-semester grading has been killing me this term.
Which of those did it really mean? That seems like a bad question to me. Why suppose that it needed to determinately mean one of them?
I assume that you think the same thing about theoretical terms, like mass. But this move strikes me as a mistake. When Newton used the word "mass" he had a determinate meaning, I think. This despite the fact that later developments in physics make "mass" ambiguous between "proper mass" and "relativistic mass." It is one thing to think that words could change their meanings in various ways, which is how I would describe your very clever reply to "Green grass is green." (Or maybe we ought to say that the same phonemes do different semantic work at different times.) But just because words can change their meanings in response to various empirical and practical demands does not mean that they have no determinate meaning at a time. And Quinean arguments for indeterminacy are, I think, entirely unpersuasive. Yes, you can rig up proxy functions to swap around labels, but that is all that is happening: label-swapping. Games like that no more change the meaning of sentences than toy "alternative" theories (e.g. theory T is just like GR except when no one is looking, in which case, everything disappears) are really rival theories.
Anyway, the replies you offer are definitely clever. Nonetheless, I am unmoved. They strike me as cheats -- in the best possible sense. ;) Something like if I had offered, "If Bucky is taller than Suzy, then Suzy is shorter than Bucky." And you had replied, "Well, Suzy could have more money than Bucky." Okay, there is a reading of "shorter than" that might work that way. But is it really a fair challenge to the claim that the sentence is analytic? I am skeptical.
These aren't arguments, of course. But I think it is suspicious that the replies to suggested analytic sentences are almost always amazingly contrived. I am always left feeling like I've just witnessed a very impressive magic show. And the magicians don't get me to believe that they have really made anything disappear, even if I don't really understand the trick.
Compare: "You say you know how much this bowling ball weighs on Earth, and you know how much it weighs on the moon, but you really have no idea how much the bowling ball just weighs? I want to give the incredulous stare." I didn't expect you to go the way you did. I thought you'd say that it's fine for there to be a whole bunch of different notions of analytic-in-L; we don't need a notion of analyticity simpliciter.
Yes, maybe I should have taken that line. I like your reply here, incidentally, though I'm still not sure if the relevant metaphor is knowing the weight or knowing how to determine the weight. The first pushes your way, the second pushes mine.
posted by Jonathan Livengood at 7:41 AM on May 9, 2013
Which of those did it really mean? That seems like a bad question to me. Why suppose that it needed to determinately mean one of them?
I assume that you think the same thing about theoretical terms, like mass. But this move strikes me as a mistake. When Newton used the word "mass" he had a determinate meaning, I think. This despite the fact that later developments in physics make "mass" ambiguous between "proper mass" and "relativistic mass." It is one thing to think that words could change their meanings in various ways, which is how I would describe your very clever reply to "Green grass is green." (Or maybe we ought to say that the same phonemes do different semantic work at different times.) But just because words can change their meanings in response to various empirical and practical demands does not mean that they have no determinate meaning at a time. And Quinean arguments for indeterminacy are, I think, entirely unpersuasive. Yes, you can rig up proxy functions to swap around labels, but that is all that is happening: label-swapping. Games like that no more change the meaning of sentences than toy "alternative" theories (e.g. theory T is just like GR except when no one is looking, in which case, everything disappears) are really rival theories.
Anyway, the replies you offer are definitely clever. Nonetheless, I am unmoved. They strike me as cheats -- in the best possible sense. ;) Something like if I had offered, "If Bucky is taller than Suzy, then Suzy is shorter than Bucky." And you had replied, "Well, Suzy could have more money than Bucky." Okay, there is a reading of "shorter than" that might work that way. But is it really a fair challenge to the claim that the sentence is analytic? I am skeptical.
These aren't arguments, of course. But I think it is suspicious that the replies to suggested analytic sentences are almost always amazingly contrived. I am always left feeling like I've just witnessed a very impressive magic show. And the magicians don't get me to believe that they have really made anything disappear, even if I don't really understand the trick.
Compare: "You say you know how much this bowling ball weighs on Earth, and you know how much it weighs on the moon, but you really have no idea how much the bowling ball just weighs? I want to give the incredulous stare." I didn't expect you to go the way you did. I thought you'd say that it's fine for there to be a whole bunch of different notions of analytic-in-L; we don't need a notion of analyticity simpliciter.
Yes, maybe I should have taken that line. I like your reply here, incidentally, though I'm still not sure if the relevant metaphor is knowing the weight or knowing how to determine the weight. The first pushes your way, the second pushes mine.
posted by Jonathan Livengood at 7:41 AM on May 9, 2013
Sorry for not responding in so long, JL... I keep meaning to find time to write another reply, but I've been under the gun with a lot of deadlines, and the posts in this thread aren't something that I want to just fire off. I'll try to return to this thread in the future, but I'll just say for now that your response is completely reasonable and it's where I expected resistance!
posted by painquale at 4:54 AM on May 24, 2013
posted by painquale at 4:54 AM on May 24, 2013
Do novels tell us anything about the ethical life that analytic philosophy cannot?
via Bookforum, with more philosophy links!
posted by the man of twists and turns at 1:16 PM on May 25, 2013
via Bookforum, with more philosophy links!
posted by the man of twists and turns at 1:16 PM on May 25, 2013
« Older Miracle twins | Press A to make these patronizing screens go away. Newer »
This thread has been archived and is closed to new comments
posted by yoink at 11:41 AM on May 2, 2013 [1 favorite]