I recently started reading up on the standard approaches to epistemology. Much of the primary discussion seems to be focused on the question “what constitutes knowledge?”. The basic definition used to be that to count as knowledge it needs to be a belief, it needs to be justified, and it needs to be true. But there’s the Gettier Problem which points out that there are cases that satisfy the above criteria but which we wouldn’t normally consider “knowledge”. Numerous alternative “theories of knowledge” have been proposed, new counter-examples have been pointed out, philosophers have split into competing camps (each under its own “-ism” title), and hundreds if not thousands of papers have been published on this topic.
But I’m totally confused. It sounds like they’re just arguing about basically arbitrary definitions. So agree on a definition and get on with it. Or define different types of knowledge if that suits you better. And if that doesn’t perfectly capture everything we might mean by the word “knowledge”, what difference does it make? If they’d taboo the word “knowledge” would there be anything left to discuss?
I assume I’m just missing something. But if in fact they could just taboo the term and get on with more important discussions, then could someone please explain to me why so many highly intelligent, extremely thoughtful philosophers have spent so much time on a (seemingly) ridiculous discussion?
I agree with this. However, there are philosophers who criticize this practice. For instance, Peter Unger recently published a vehement criticism of mainstream analytic philosophy, Empty Ideas.
One influential view is that we should not try to “analyze” pre-theoretical concepts, but rather construct fruitful, exact and simple “explications”. If you have that view, definitions do not become interesting for their own sake. Rather, terms and concepts are a tool in the pursuit of knowledge, which can be more or less effective. See Carnap’s dicussion in Logical Foundations of Probability, pp. 3-20 (esp. p. 7).
That said, it is true that many philosophers continue to write papers on the Gettier problem in a very classical essentialist fashion, along the lines you are describing. The same goes for many other philosophical discussions (e.g. on truth, reasons, etc).
I think that there is a selection effect at work here: those who think this is silly move on to other things while those who think that it isn’t keep on doing it. This creates the illusion that more people think this is a good and interesting form of philosophy than is actually the case.
Of course now and again some outsiders get so fed up with this that they write a book on it to attack it. Another similar example of this (in addition to Unger) is Ladyman and Ross’s attack on mainstream analytic metaphysics (which treats questions like “is the statue and the lump of clay that is made of distinct or identical objects?). I suspect that many others feel, however, that although this kind of philosophy is a bit of a nuisance, there are other more pressing problems more worth focusing on. For instance, I suspect Nick Bostrom doesn’t like this kind of philosophy, but as far as I know he hasn’t spent much time criticizing it, thinking there are other problems which are more important to spend time on.
Also, it seems surprisingly hard to weed out. The kind of criticism that Carnap gave is at least a century old, but the Gettier problem and other similar problems are still treated seriously.
An interesting argument for why people who are critical of this kind of philosophy should do something about it is, though, that it presents a great opportunity cost:
Why should something as “quixotic”, “mostly harmless”, and null as academic philosophy rouse any strong feelings whatsoever?
Because of the opportunity cost. Harmless-and-null philosophy is crowding out something better, and has been doing so since 1950 or so. Philosophy did not have to be what it is today; it was made what it is by purposeful, destructive action.
No, though I understand my comment could be read in that way. I have thought and read a lot about these questions (and written some things) and sometimes get a bit frustrated with them. I have started to become more pessimistic about the possibilities of convincing mainstream philosophers who like to work on these questions (“scholasticism with a dull knife”, as a brilliant colleague of mine scribbled on his noteblock during a talk on the Gettier problem).
Perhaps we should instead focus on showing what alternative things philosophers could do. Also we should make alliances with other subjects. People outside the discipline are much more likely to want to fund work on business ethics or medical ethics than yet another go at some concept or metaphysical question.
Without borrowing wholesale Kuhn’s picture of science, I think some ideas Kuhn introduced are important to keep in mind when considering the trajectory of philosophy. Research programs are adopted, consciously or not, by a certain part of the philosophical community: certain tenets are taken for granted, certain notions are regarded as the proper ones to use as tools, and certain puzzles are regarded as the ones to focus attention on. The research program isn’t abandoned simply on the ground that seemingly compelling arguments against its fundamental assumptions are presented. Rather, it is abandoned when research conducted within its confines is no longer seen as fruitful, and when a new alternative, with some promise of success, is available.
If we can’t disprove the Gettier stuff, perhaps we can hope that people will get bored of it (if we provide them with a less boring alternative).
If we can’t disprove the Gettier stuff, perhaps we can hope that people will get bored of it (if we provide them with a less boring alternative).
When you can’t disprove something the straightforward way is to accept it. In this case you can switch to a more construtivist notion of knowledge. To quote Heinz von Förster: “Truth is the invention of a liar.”
Perhaps we should instead focus on showing what alternative things philosophers could do. Also we should make alliances with other subjects. People outside the discipline are much more likely to want to fund work on business ethics or medical ethics than yet another go at some concept or metaphysical question.
The problem isn’t that you can’t do anything useful with ontology but that a lot of analytic philosophers are confused about the subject and produce papers that provide no value.
Barry Smith does deal with the question of knowledge and get’s funded because he actually does something useful. Applied ontology is useful for bioinformatics and other fields likely also would profit from it.
It possible that in one or two decades bioinformatic inspired mapping of mental states is good enough that the psychology folks with their DSM simply loses it’s authority.
why so many highly intelligent, extremely thoughtful philosophers have spent so much time on a (seemingly) ridiculous discussion?
That has been my exact reaction to most of the “big” questions, with a few exceptions.
As for the “justified true belief” “definition”, each word in it is already so poorly defined, you might as well give up on it. My personal working definition of knowledge is the capability to accurately predict the outcome(s) of a certain (set of) action(s). Which probably matches an existing camp or two, not that I care.
If they’d taboo the word “knowledge” would there be anything left to discuss?
Yes.
If we want to have an AI that knows things we have to be specific about what knowledge is.
If we have unrealistic naive concepts we will never get the knowledge into the AI.
If you want an university to teach knowledge, then it makes sense to have an idea of what the university is supposed to teach.
If you want to decide whether someone has depression, than it makes sense to ask what you mean with the sentence: “Alice has depression.”
Currently it might mean: “A trained psychologist has found that Alice fulfills the criteria of the DSM-V for depression.”
It would be possible to get saner way to deal with the issue if we would have a better grasp on the underlying ontology.
Based on certain ideas of knowledge people reject certain approaches as pseudoscience because they don’t fulfill the criteria of what’s believed to be necessary for knowledge generation.
Yes, philosophers tend to be interested in the issue of conceptual analysis. Different philosophers will have a different understanding of what conceptual analysis is but one story goes something like the following. First, we start out with a rough, intuitive sense of the concepts that we use and this gives us a series of criteria for each concept (perhaps with free will one criteria would be that it relates to moral responsibility in some way and another would be it relates to the ability to do otherwise in some way). Then we try to find a more precise account of the concept that does the best (though not necessarily perfect) job of satisfying these criteria.
I personally find the level of focus on conceptual analysis in philosophy frustrating so I’m not sure that I can do justice to a defence of it. I know many very intelligent people who think it is indispensible to our reasoning though so it may well be deserving of further reflection. If you’re interested in such reflection I suggest that you read Frank Jacksons, “From Metaphysics to Ethics: A Defence of Conceptual Analysis”. It’s a short book and gives a good sense of how contemporary philosophers think about conceptual analysis (in terms of what conceptual analysis is, btw, Jackson says the following: “The short answer is that conceptual analysis is the very business of addressing when and whether a story told in one vocabulary is made true by one told in some allegedly more fundamental vocabulary.”)
Off the top of my head, why might someone think conceptual analysis is important? First, conceptual analysis is all about getting clear on our terms. If you’re discussing free will, it seems like a really bad idea to just debate without making clear what you mean by free will. So it seems useful to get clear on our terms.
Myself, I’m tempted to say we should get clear on our terms by stipulation (though note that even this involves a small amount of conceptual analysis or I would be just as likely to stipulate that “free will” means “eating my hat” as I would be to stipulate that it means “my decisions flow from my deliberative process”: and many philosophers only use conceptual analysis in this easily-defendable manner). So I would stipulate what various meanings of free will are, say which we do and do not have and leave it to each individual to figure out how that makes them feel (which type of free will they care about).
However, a lot of people don’t find this very helpful. They care about FREE WILL or MORALITY or BEING THE SAME PERSON TOMORROW (a.k.a. personal identity). And they need to know what the best realiser of the criteria that they have for this concept is to know what they care about. So if you tell a lot of people that they have free-will-1 and not free-will-2, they don’t care about this: they care whether they have FREE WILL and so they need to find out which of free-will-1 and free-will-2 is a better realiser of their concept before they can work out whether they have the sort of free will that they care about. Insofar as I don’t think telling people what they should desire (no, you should desire free-will-1, regardless of the nature of your concept of FREE WILL), I don’t really have any objection to the claim that such a person needs to carry out a more robust project of conceptual analysis (though I feel no need to join them on the road they’re travelling).
All that said, standard epistemology (as opposed to formal epistemology) is one of the worst areas of philosophy to study if you’re uninterested in these conceptual debates as such debates are pretty much the entire of the field (where in many other areas of philosophy they play a smaller, though often still substantial, role).
I recently started reading up on the standard approaches to epistemology. Much of the primary discussion seems to be focused on the question “what constitutes knowledge?”. The basic definition used to be that to count as knowledge it needs to be a belief, it needs to be justified, and it needs to be true. But there’s the Gettier Problem which points out that there are cases that satisfy the above criteria but which we wouldn’t normally consider “knowledge”. Numerous alternative “theories of knowledge” have been proposed, new counter-examples have been pointed out, philosophers have split into competing camps (each under its own “-ism” title), and hundreds if not thousands of papers have been published on this topic.
But I’m totally confused. It sounds like they’re just arguing about basically arbitrary definitions. So agree on a definition and get on with it. Or define different types of knowledge if that suits you better. And if that doesn’t perfectly capture everything we might mean by the word “knowledge”, what difference does it make? If they’d taboo the word “knowledge” would there be anything left to discuss?
I assume I’m just missing something. But if in fact they could just taboo the term and get on with more important discussions, then could someone please explain to me why so many highly intelligent, extremely thoughtful philosophers have spent so much time on a (seemingly) ridiculous discussion?
I agree with this. However, there are philosophers who criticize this practice. For instance, Peter Unger recently published a vehement criticism of mainstream analytic philosophy, Empty Ideas.
One influential view is that we should not try to “analyze” pre-theoretical concepts, but rather construct fruitful, exact and simple “explications”. If you have that view, definitions do not become interesting for their own sake. Rather, terms and concepts are a tool in the pursuit of knowledge, which can be more or less effective. See Carnap’s dicussion in Logical Foundations of Probability, pp. 3-20 (esp. p. 7).
That said, it is true that many philosophers continue to write papers on the Gettier problem in a very classical essentialist fashion, along the lines you are describing. The same goes for many other philosophical discussions (e.g. on truth, reasons, etc).
I think that there is a selection effect at work here: those who think this is silly move on to other things while those who think that it isn’t keep on doing it. This creates the illusion that more people think this is a good and interesting form of philosophy than is actually the case.
Of course now and again some outsiders get so fed up with this that they write a book on it to attack it. Another similar example of this (in addition to Unger) is Ladyman and Ross’s attack on mainstream analytic metaphysics (which treats questions like “is the statue and the lump of clay that is made of distinct or identical objects?). I suspect that many others feel, however, that although this kind of philosophy is a bit of a nuisance, there are other more pressing problems more worth focusing on. For instance, I suspect Nick Bostrom doesn’t like this kind of philosophy, but as far as I know he hasn’t spent much time criticizing it, thinking there are other problems which are more important to spend time on.
Also, it seems surprisingly hard to weed out. The kind of criticism that Carnap gave is at least a century old, but the Gettier problem and other similar problems are still treated seriously.
An interesting argument for why people who are critical of this kind of philosophy should do something about it is, though, that it presents a great opportunity cost:
Do you think the question whether or not core foundations of analytical philosophy are correct is unimportant?
No, though I understand my comment could be read in that way. I have thought and read a lot about these questions (and written some things) and sometimes get a bit frustrated with them. I have started to become more pessimistic about the possibilities of convincing mainstream philosophers who like to work on these questions (“scholasticism with a dull knife”, as a brilliant colleague of mine scribbled on his noteblock during a talk on the Gettier problem).
Perhaps we should instead focus on showing what alternative things philosophers could do. Also we should make alliances with other subjects. People outside the discipline are much more likely to want to fund work on business ethics or medical ethics than yet another go at some concept or metaphysical question.
I think this view of Matti Eklund’s has a lot to be said for it:
If we can’t disprove the Gettier stuff, perhaps we can hope that people will get bored of it (if we provide them with a less boring alternative).
When you can’t disprove something the straightforward way is to accept it. In this case you can switch to a more construtivist notion of knowledge. To quote Heinz von Förster: “Truth is the invention of a liar.”
The problem isn’t that you can’t do anything useful with ontology but that a lot of analytic philosophers are confused about the subject and produce papers that provide no value.
Barry Smith does deal with the question of knowledge and get’s funded because he actually does something useful. Applied ontology is useful for bioinformatics and other fields likely also would profit from it.
It possible that in one or two decades bioinformatic inspired mapping of mental states is good enough that the psychology folks with their DSM simply loses it’s authority.
That has been my exact reaction to most of the “big” questions, with a few exceptions.
As for the “justified true belief” “definition”, each word in it is already so poorly defined, you might as well give up on it. My personal working definition of knowledge is the capability to accurately predict the outcome(s) of a certain (set of) action(s). Which probably matches an existing camp or two, not that I care.
Yes.
If we want to have an AI that knows things we have to be specific about what knowledge is. If we have unrealistic naive concepts we will never get the knowledge into the AI.
If you want an university to teach knowledge, then it makes sense to have an idea of what the university is supposed to teach.
If you want to decide whether someone has depression, than it makes sense to ask what you mean with the sentence: “Alice has depression.” Currently it might mean: “A trained psychologist has found that Alice fulfills the criteria of the DSM-V for depression.” It would be possible to get saner way to deal with the issue if we would have a better grasp on the underlying ontology.
Based on certain ideas of knowledge people reject certain approaches as pseudoscience because they don’t fulfill the criteria of what’s believed to be necessary for knowledge generation.
Yes, philosophers tend to be interested in the issue of conceptual analysis. Different philosophers will have a different understanding of what conceptual analysis is but one story goes something like the following. First, we start out with a rough, intuitive sense of the concepts that we use and this gives us a series of criteria for each concept (perhaps with free will one criteria would be that it relates to moral responsibility in some way and another would be it relates to the ability to do otherwise in some way). Then we try to find a more precise account of the concept that does the best (though not necessarily perfect) job of satisfying these criteria.
I personally find the level of focus on conceptual analysis in philosophy frustrating so I’m not sure that I can do justice to a defence of it. I know many very intelligent people who think it is indispensible to our reasoning though so it may well be deserving of further reflection. If you’re interested in such reflection I suggest that you read Frank Jacksons, “From Metaphysics to Ethics: A Defence of Conceptual Analysis”. It’s a short book and gives a good sense of how contemporary philosophers think about conceptual analysis (in terms of what conceptual analysis is, btw, Jackson says the following: “The short answer is that conceptual analysis is the very business of addressing when and whether a story told in one vocabulary is made true by one told in some allegedly more fundamental vocabulary.”)
Off the top of my head, why might someone think conceptual analysis is important? First, conceptual analysis is all about getting clear on our terms. If you’re discussing free will, it seems like a really bad idea to just debate without making clear what you mean by free will. So it seems useful to get clear on our terms.
Myself, I’m tempted to say we should get clear on our terms by stipulation (though note that even this involves a small amount of conceptual analysis or I would be just as likely to stipulate that “free will” means “eating my hat” as I would be to stipulate that it means “my decisions flow from my deliberative process”: and many philosophers only use conceptual analysis in this easily-defendable manner). So I would stipulate what various meanings of free will are, say which we do and do not have and leave it to each individual to figure out how that makes them feel (which type of free will they care about).
However, a lot of people don’t find this very helpful. They care about FREE WILL or MORALITY or BEING THE SAME PERSON TOMORROW (a.k.a. personal identity). And they need to know what the best realiser of the criteria that they have for this concept is to know what they care about. So if you tell a lot of people that they have free-will-1 and not free-will-2, they don’t care about this: they care whether they have FREE WILL and so they need to find out which of free-will-1 and free-will-2 is a better realiser of their concept before they can work out whether they have the sort of free will that they care about. Insofar as I don’t think telling people what they should desire (no, you should desire free-will-1, regardless of the nature of your concept of FREE WILL), I don’t really have any objection to the claim that such a person needs to carry out a more robust project of conceptual analysis (though I feel no need to join them on the road they’re travelling).
All that said, standard epistemology (as opposed to formal epistemology) is one of the worst areas of philosophy to study if you’re uninterested in these conceptual debates as such debates are pretty much the entire of the field (where in many other areas of philosophy they play a smaller, though often still substantial, role).