Monday, February 05, 2007

What Do Analytic Philosophers Analyze? (by guest blogger Jonathan Ichikawa)

Hello, Splintered Mind readers! I'm pleased to be guest-blogging, and
thanks to Eric for the invitation. I'm looking forward to some good
discussions over the next couple of weeks.

I thought I'd start off with a question in philosophical methodology.
There's a certain kind of disagreement that sometimes comes up about
how to go about doing philosophy -- it connects to lots of deep
issues, and from the way people talk about it, it sure seems like it's
a big ideological disagreement, but I have a pretty hard time wrapping
my head around just what the disagreement is supposed to be.

The debate I have in mind tends to occur between traditional
philosophers who think that lots of interesting philosophy is a
priori
, and philosophers who think that we have to go out and do
empirical investigation to learn anything interesting. Roughly -- and
only roughly -- speaking, some philosophers like thought experiments,
and others like psychological experiments.

Sometimes, I feel debate between these camps goes something like this.
Maybe I'm caricaturing, but I don't think so:

Archie: I'm using a priori investigation to learn about the
nature of knowledge.

Eddie: When you say 'knowledge', are you talking about something out
there in the world? Or are you just studying the concept of
knowledge?

Archie: Oh, I definitely am interested in knowledge out there in
the world
. I'm using a priori investigation to learn which
things are instances of knowledge -- that thing out there in
the world.

Eddie: That's hopeless. How can a priori investigation teach
you something about something out there in the world? At best, a
priori
investigation could give you insight into which things fall
under the concept knowledge. That's a merely psychological
matter.

Archie: No, it's not just a psychological question -- I agree that
would be uninteresting. I'm learning about the nature of knowledge,
that thing in the world!

You may anticipate my confusion from the way I set up this dialogue.
I can't see what this disagreement is supposed to be about. Here are
the two questions that we're supposed to be contrasting:

(1) Which things are knowledge?
(2) Which things fall under the concept KNOWLEDGE?

But these two questions are identical. Or, at least they're
equivalent. For after all, it is a tautology that all and only things
that are knowledge are things that fall under the concept KNOWLEDGE.

So what is it that this disagreement is supposed to be about? It
can't be that Armchair Archie is illicitly attempting to adjudicate
from the armchair what actual cases of knowledge there are in the
world, because this would be an obvious and ridiculous mistake. It
can't be that Experimental Eddie's question presupposes that
'knowledge' is a natural kind term, because it doesn't; "which things
are trains?" has just the same structure as "which things are
knowledge?", and nobody thinks trains are natural kinds.
(Furthermore, "Which things fall under the concept WATER?" can't be
answered a priori.)

So what gives? Is there a genuine disagreement in the ballpark?

33 comments:

Eric Schwitzgebel said...

Nice post, Jonathan!

I'm with Eddie. When you're reporting on your intuitions, you're revealing something about your (or our) concepts. Either that's a matter of your (or our) psychology, or you're saying something about a concept that it's possible to have. But whether that concept maps usefully onto anything in the real world is an empirical matter -- or, more precisely, a pragmatic matter (since it involves our values, what we want to capture).

This isn't to disrespect a priori philosophical reflection, which is useful both in revealing our concepts and in generating interesting new concepts likely to be of use (because developed not wholly in ignorance of empirical facts); but I think a priori philosophers are often mistaken about the epistemic status of their judgments.

Brad said...

Hi Jonathan,

Nice question. I think one difference in the area relates to how you initially described the two camps: "traditional philosophers who think that lots of interesting philosophy is a priori, and philosophers who think that we have to go out and do empirical investigation to learn anything interesting."

One worry I have about empirically driven research is that it turns attention towards questions about the concepts we happen to use and away from the concepts we should use. I think many drawn to the traditional approach think philosophers should be centrally concerned with normative questions (those in the second group).

I think of the issue in terms of Plato's Euthyphro. My worry about methodological empiricism arises when that view leads philosophers to think of philosophy as the empirical investigation of the actual concepts various communities use. Now if this is what philosophers should do, then Socrates should have gone around polling people about whether they agree with Euthyphro that hauling his dad into court was the pious thing to do or not.

Traditionalists think that philosophers should, instead, follow Socrates lead; he tries to (1) get Euthyphro to think about the concepts were are giving sense to his way of life and (2) to then submit those concepts, and his way of life, to rational critique.

Finally, I emphasize that this is just a worry. It is possible, of course, to both investigate the concepts that people use and to do normative philosophy too. After all, we might see the former as an extension of 1 above, and as, consequently, an important part of the traditional approach. But in my (admittedly small) exposure to work flying under the flag of "experimental philosophy," I have to admit I have not seen a lot of normative philosophical reflection and I have worried that this is not merely an accident. I bet this has something to do with sympathy for naturalized epistemology as a model for philosophic reflection...

In sum, I think we need to distinguish normative inquiry from the descriptive inquiry, and to then ask about their relations and relative importance to the philosophic enterprise (if any).

A first go at identifying some characteristic questions of different lines of inquiry:

(N) What are rightness or goodness? What things are right or good? What makes something right or good?

(D1) How do people use the concepts of right or good? What things do people say are right or good when you ask them? What do people think makes things right or good?

(D2) How are conceptions of right and good embodied in people's behavior? How do emotions and desires betray people's evaluative stances? How to their "explicit beliefs" relate to their "embodied beliefs"? How can normative reflection impact each of these?

It is hard to see how anyone would think that empirical inquiry is the only route to interesting answers to the questions in the first (normative) group - in fact, it is hard to see how empirical inquiry could answer them at all.

Perhaps this is behind the traditionalist attitude you mention: "traditional philosophers who think that lots of interesting philosophy is a priori, and philosophers who think that we have to go out and do empirical investigation to learn anything interesting."

Jonathan Ichikawa said...

Hi Eric, I agree with you and Eddie to the following extent: I agree that, say, the Gettier intuition about a thought experiment reveals something about the our concept. Does that mean it's a matter of psychology? I guess I don't have a strong view about which things should count as "matters of psychology", but if being "about psychology" is meant to contrast with "about knowledge", I certainly disagree. I'm disagreeing with both Archie and Eddie that there's an important question they're disagreeing about; when you say you agree with Eddie, I take it you take yourself also to be disagreeing with Archie?

My point is that if you can answer a priori what it takes to fall under the concept KNOWLEDGE, then you can a forteriori answer a priori what it takes to be knowledge. And then you've learned something worthwhile about knowledge.

There are two points Eddie could jump off the train here. He could deny that we can know what falls under KNOWLEDGE a priori, or he can deny that it's worthwhile to learn facts about knowledge. I haven't argued against either of these moves, of course. But that's because the character in my dialogue didn't make them. The argument I find baffling is the one that says: "You can't learn facts about knowledge from the armchair; you could only learn facts about the concept KNOWLEDGE."

Lest I be taken to be beating up on Eddie, I think Archie's response is very weird, too. Instead of insisting that he's not engaging in a psychological question, why doesn't he say that the 'psychological question' IS the same question as the knowledge question?

Jonathan Ichikawa said...

Brad, I have often heard people say things like what you say here:

One worry I have about empirically driven research is that it turns attention towards questions about the concepts we happen to use and away from the concepts we should use.

I have never really understood the thinking behind talk about what concepts we should use, beyond a basic consistency restraint ("don't use TONK"). What would ground these normative facts? Pragmatic reasons ("having a word for GRUE would be a waste of mental energy")? Moral reasons ("no BOSCH")? I guess I can see that there could be room to debate these questions, but it just seems wrong that Plato or traditional philosophers or really anybody else that seems philosophically prominent is making these arguments.

michael metzler said...

Eric’s way of setting it up seems helpful. Not sure this matches up precisely, but: Our reporting of our intuitions about what we take philosophers as doing reveals something about our concept of ‘concept.’ Whether or not our philosophical use of ‘concept’ usefully maps onto anything in the real world—including the workings of our unconscious mind—is largely a pragmatic matter.

Brad C said...

Jonathan,

Perhaps my way of characterizing this was misleading, because it encourages the picture of us somehow stepping outside of our normative conceptions and then evaluating them. As your response indicates, that encourages us to ask, "which normative conceptions should we use to judge the ones out of which we are stepping?"

We can avoid this problem, I think, because inquiry aimed at determining which normative conceptions to adopt is best pursued by asking the sorts of questions I put under normative inquiry at the end of the last post. Figuring out which conception of 'good' to adopt would, for example, involve trying to understand the concepts 'good' and 'bad' and trying to reason about what things are good and bad , what makes these things good and bad, etc.

We can pursue similar normative inquiry, which is, I take it, characteristic of the tradition, in order to choose amongst other normative conception as well.

The general point: since the concepts themselves are normative, deciding how best to use them is not something apart from thinking in terms of different conceptions and arguing some are better than others.

The terms of criticism here are debatable and vary from domain to domain, but I see no reason to think the reasons for criticizing a conception should be restricted to the types of considerations that would tell against 'tonk'.

Brad C said...

Also: Your claim that questions about the instrumental or prudential value of concepts have not been pursued by "Plato or traditional philosophers or really anybody else that seems philosophically prominent" is, unless I misunderstand you, simply false.

In practical philosophy there is a long tradition of people claiming that the moral conceptions that animate "our" (at least western) practical thinking and our emotional lives are open to criticism on the ground that they are bad for us. Obviously Freud, Nietzsche, and Marx come to mind here, but the idea that conventional moral conceptions are an illusion we, or at least the strong amongst us, could do *better* without goes back to Plato (think of Callicles) and includes many lesser know figures (e.g., Mandeville's fable of the Bees).

I also would suggest that Williams' "Truth and Truthfulness", Craig's "Knowledge and the State of Nature", Richard Joyce's "Myth of Morality" might fall roughly into this category. We might also include indirect utilitarian accounts of ordinary moral thought (e.g. Sidgwick's government house until)

Pete Mandik said...

Nice Post, Jonathan.

I agree with your point about how it doesn't look like there's much of a difference between what EE and AA are up to. I would add that what they are both doing isn't terribly empirical. Further, I think that EE isn't a particularly good illustration of what empirical investigation amounts to.

To spell this out further, consider the difference between these two questions:

1. What's the mass of Pluto?
2. Is Pluto a planet?

Question 1 strikes me as straightforwardly empirical. If you want to know Pluto's mass, you have to go and check. Question 2, while not the sort of thing that can be settled solely in the armchair (I don't believe anything can be settled like that) is the sort of thing that invloves more of a "looking inward" that is associated with the a priori. People look at their theories and make appropriate adjustments and draw appropriate inferences to settle the question. We can't just whip out the planet detector and point it at Pluto and see if it lights up the right way.

While I don't think your EE and AA have a real disagreement, here is what I think amounts to a real disagreement about methodolgy:

To what degree can we figure out what to say about questions like #2 just by "looking inward" at our intuitions, at our current theoretical entailments, and to what degree is this open to revision in light of further discovery? Real life enemies of armchair methodology like the Churchlands point to examples like the theory of electromagnetic radiation to show how our concept of light was maleable and changed in light (pun intended) of empricial discoveries.

Eric Schwitzgebel said...

I could say a lot, but let me just stick with this:

I'm not sure how useful it is to learn what knowledge is, if by that you just mean learn what kind of thing falls (a priori) under our concept of knowledge -- except insofar as we think our concept of knowledge reflects some kind of empirical (or innate?) knowledge about the world (which surely it does). But then the epistemic power comes mostly from the knowledge implicit in our concept of knowledge. Compare a priori analysis of our concept of "psychic powers" or "a Scorpio".

(Sheesh, what an ugly paragraph!)

Tanasije Gjorgoski said...

Hi Jonatan,

I think the issue is not if a priori can tell us something about the world (as Eric puts it), but how much it can help us understand it.

I think the form in which a priori thinking helps us understand it is in form of couterfactuals. That is, *if* something falls under that concept, it will necessarily fall under that other concept. And things do fall under our concepts, in fact that is the only way we can think about the world.

Separate question is how complex things it can help us understand.
Simplest forms would be applied logic and math. For example a priori reasoning can tell us that 10 things can't be equally divided to 3 people. So in concrete situation if we have 10 things, a priori reasoning tell us something about them. (i.e. that they can't be divided equally to 3 people).

More complex explanations are used e.g by sciences where the explanations consist in telling us how *given this or that physical law* a phenomenon that can be characterized by the concepts used in the law, will show such and such characteristics.

Jonathan Ichikawa said...

Ok, lots to respond to!

Brad: Ok, I take the point that insofar as some concepts have important normative roles to play, normative considerations may bear on how we use them. I guess I'm having trouble seeing the connection to the methodological disagreement I'm highlighting, though. Do you think that the armchair folks are assuming, while the experimentals folks are ruling out, that all concepts are like this? Help me out with connecting the point to this methodological disagreement.

Jonathan Ichikawa said...

Pete: thanks, that's a helpful diagnosis. Here are a couple of points. First, I'm not sure I see the disanalogy between your (1) and (2):

1. What's the mass of Pluto?
2. Is Pluto a planet?

You say that (1) is "straightforwardly empirical" and (2) "involves more of a 'looking inward' that is associated with the a priori". (I disagree with this characterization of apriority, but set that aside.)

I guess the idea here is that when evaluating (2), we need to pay special attention to what "planet" means; what would count as a planet? But if that's the point, I don't see how (2) is supposed to contrast with (1) -- after all, the proper evaluation of (1) depends on what terms like "Pluto" and "mass" mean.

Can you say more about what the difference you have in mind is?

Separate (I think?) point: 'planet' and 'light' are scientific, theoretical terms. I agree that it's probably not the best idea for most of us to be doing a lot of a priori investigation of planets or light, because most of us don't sufficiently understand how they fit into physical and astronomical theories. I'm told by smart scientists that Pluto is not a planet; I can't recall if I have heard why not -- if I have, I've forgotten. But I believe them. I have only a very incomplete idea of what it takes to be a planet, so I wouldn't presume to judge planethood in borderline thought experiments. Part of my concept there includes deference to relevant authorities. Likewise with light.

Admitting all this, though, isn't to give up on apriority. This is all consistent with the thought that for some other concepts, I do have a sufficiently complete grasp. I think KNOWLEDGE is like this. If a big group of psychologists -- or epistemologists, for that matter -- told me that such-and-such Gettier case is an instance of knowledge after all, I would not defer to their authority; I would either attribute an error to them or, perhaps more plausibly, attribute a different concept to them.

It's worth pointing out, too, that it's at least in principle possible to recognize cases where these conceptual roles are susceptible to change, in advance, from the armchair. Any time an empirical discovery would cause us to alter our theories about a concept, a smart enough person could imagine that discovery ahead of time, and simulate the theory change.

Jonathan Ichikawa said...

Eric, I guess I'm not sure if I disagree with anything in your last comment. But it all sounds pretty consistent with a pretty traditional methodology, apriority and all. One might say:

"We seem to have this pretty useful notion of knowledge, that plays a bunch of important roles. (After all, we're smart creatures, and probably wouldn't have a word as prominent as 'knowledge' if it didn't pick out anything interesting.) Let's figure out what knowledge is. Let's do it by a priori investigation into the content of our concept."

We might be convinced that there is good reason to care about something for a posteriori reasons, then investigate it a priori.

I'm not sure what to make of this, line from your comment, Eric, in light of what I've just said:
But then the epistemic power comes mostly from the knowledge implicit in our concept of knowledge.

Can you help to explain this?

Jonathan Ichikawa said...

Tanasije, I'm sorry, but I'm having trouble understanding what you're up to. How does the knowledge/understanding distinction help?

Also, this seems false:
That is, *if* something falls under that concept, it will necessarily fall under that other concept.

I fall under the concept GRADUATE STUDENT, but do so only contingently. (Maybe that I choose a trivial example shows that I don't understand what you're up to.)

Eric Schwitzgebel said...

Hi, Jonathan. I agree that what I say here (and think in general) is quite consistent with valuing a priori philosophy. Reading David Lewis is such good fun, how could it be pointless?!

The main area where I'd part ways with certain views of what's going on in a priori philosophy is in the epistemic status of the claims so generated. A priori philosophy is only as certain, and as valuable, as the folk conceptions it illuminates (if it's descriptive of our concepts); or (if it's constructing new concepts or altering existing ones) it's not so much the discovery of truths as the making of practical recommendations. In the first disjunct, I take myself to be with Austin, in the second, with Carnap. In the first, a priori philosophy just brings out what is already known to us by empirical means. In the second, the value of the term depends on how well it serves us in light of the empirical facts and facts about what we care about.

If Archie's okay with all that, I'm okay with Archie. But my guess is that he'll balk. At least, if he's George Bealer, he'll balk!

Jonathan Ichikawa said...

Thanks, Eric -- that is helpful.

Suppose Archie takes your first, Austinean choice. A priori investigation illuminates concepts. Then, you say:

A priori philosophy is only as certain, and as valuable, as the folk conceptions it illuminates...

And:

a priori philosophy just brings out what is already known to us by empirical means.

Of these three claims, only the second is obvious to me. I don't really understand what the first amounts to. What is it for a concept to be more or less certain?

Regarding the third: grant that a priori investigation reveals something we tacitly knew or were sensitive to all along. Why think that this means it's "something already known to us by empirical means"?

By what empirical means do we know that knowledge isn't justified true belief?

In the background, of course, is this question about how the various views being floated correspond to views that actual philosophers actually hold. I'm happy to let that stay mostly in the background still, lest various contingent commitments of particular philosophers get in the way of this methodological question. Also, it's just a lot more work responsibly to attribute views on subtle distinctions to real people than to fictional characters. But I will say that I have a hard time seeing what Bealer is up to; I do have the impression that he would balk. But maybe somebody like Frank Jackson would be happy with this kind of line?

Tanasije Gjorgoski said...

Jonathan,

Hi, sorry for not being more clear.
Let me rephrase the part that you point to as false, so it will be more clear. I meant that a priori gives us possibility to form counterfactuals - *If* something falls under concept A, it will necessarily fall under other concept B.

And as far as we talk about some concrete in the world as falling under A, the a priori reasoning can tell us that it will fall under B.

Let's say that I want to focus more on the understanding vs. knowledge, because as far the a priori reasoning can give us necessary relations between universals, it can be applied only so far we do have a phenomenon in the world which falls under some universal, and we want to understand it. (A priori doesn't tell us that there is a phenomenon).

Hope this makes my previous comment more clear.

Eric Schwitzgebel said...

Thanks for the reply, Jonathan. I'd say the empirical knowledge behind "knowledge is not (just) justified true belief" is in the implicit assumption that "knowledge" in that sense is the interesting and important concept we should be working with rather than say "schmoledge" (which is justified true belief).

I did not mean to imply that a concept is certain or uncertain. Rather, a concept is more or less useful. Whether such-and-such is the right articulation of our concept, or whether the concept maps usefully onto something we care about in the world -- those are the things that are more or less certain.

The claim that it is "a priori certain" that knowledge is a form of belief (say) is toxically ambiguous between various background metaphilosophical and epistemic projects and assumptions. It has an air of certainty about it for one reason (as an analysis of a concept presently being employed by the philosopher) and an air of substance about it for another (because it sounds like a claim about the world beyond our concepts). To conflate these two is to be like "Phil" in Alison Gopnik's and my 1998 dialogue in "Whose Concepts Are They Anyway?". See http://schwitzsplinters.blogspot.com/2006/10/
intuitions-in-sandbox.html

Pete Mandik said...

Jonathan,

Ok, good questions. Let me see if the following helps.

Re: (1) What's the mass of Pluto? and (2) Is Pluto a planet?

I intend the distinction to be one of degree. And I agree with you that answering both involves a sensitivity to the meanings of the terms. Where they differ, though, is not something that can simply be read off of the form of the sentences. I had something in mind that involves the history of the terms involved. In particular, "planet" has been in the news relatively recently as a term of contention among scientists. There was a lack of consesnus about which objects should be counted among the planets. The terms "mass" and "Pluto" however, have sufficiently fixed meanings that there is little else to do but take your means for determining masses and apply it to the object in question. In that sense, then, is the answering of the question about mass more straightforward. It is straightforward what our theory of mass tells us about how to figure out what Pluto's mass is. Our theory of planethood, on the other hand, required further tweaking before we could answer question 2.

I see the kind of tweaking involved as the sort of things philosophers do all the time. Our concepts of, e.g. freewill, obligation, knowledge, are insufficently explicit and clear to just straightforwardly tell whether something counts as an obligation or an instance of free will.

I think your remark about 'light' being disanalogous from 'knowledge' is interesting, but I don't think there's much I agree with you on there.

I think 'light' had a life prior to any explicitly scientfic investigation on light was conducted and the scientific use of 'light' is continuous with the non-scientific use. A prescientific philosopher may have attempted an a priori argument that the idea of invisible light was incoherent, and current consensus is that such a philosopher would have been wrong. I suppose that philosopher might want to insulate himself from scientific criticism by saying that his use of 'light' picks out somthing different from the scientists' use, but I'm not sure how the argument would go from there.

I think perhaps you and I disagree on whether there is a distinction between terms that are, on the one hand scientific and theoretical and terms that, on the other hand, aren't. I would say that all terms are theoretical, though often the theory they are embedded in has not been made explicit. Further, the process of making them explicit is not insulated from input from scientists or other experts.

I find it curious that you think you have such a complete grasp of the concept of knowledge that you wouldn't defer to a gaggle of experts who disagreed with you over some Gettier case. I think that the concepts so central to philosophy (like knoweldge, obligation, free will) are precisely those concepts that no one has a particularly complete grasp of. Why else is philosophy so hard?

Jonathan Ichikawa said...

Eric: You say:

I'd say the empirical knowledge behind "knowledge is not (just) justified true belief" is in the implicit assumption that "knowledge" in that sense is the interesting and important concept we should be working with rather than say "schmoledge" (which is justified true belief).

I think we must be talking about different things. I'm not sure if we actually disagree about anything yet. I think that it's a priori that knowledge isn't JTB because our warrant for that knowledge doesn't rely on any empirical knowledge. You say there is empirical knowledge 'behind' it in the assumption that knowledge is important and worth talking about.

I agree that people bother to try to analyze knowledge because they are making this assumption. I'll assume for (at least) the purpose of argument that this assumption can only be known empirically. But none of that prevents K not JTB from being a priori. The knowledge that KNOWLEDGE is worth thinking about and understanding isn't warranting for, say, the Gettier intuition. (There is, after all, no contradiction or incoherence in the claim that knowledge is not justified true belief, and it's also not worth analyzing or caring about. I used to think this, until I read Williamson.)

Thanks for the link; I'll take a look at that paper when I get a chance, which may or may not be soon.h

Jonathan Ichikawa said...

Pete: thanks for your thoughtful comment. There's lots there I need to spend more time thinking about. Here is one thing I feel prepared to say off the bat. In response to your last paragraph:

I find it curious that you think you have such a complete grasp of the concept of knowledge that you wouldn't defer to a gaggle of experts who disagreed with you over some Gettier case.

Well, any time a bunch of people seem to disagree with me, I can either give up my belief, attribute to them a confusion, or figure we're talking about different concepts. I guess I just think that last option sounds really attractive in the imaginary sort of case you suggest. It seems like, given how fine-grained our concepts are, disagreement in cases like this would be just the way to recognize that different people are using different concepts.

I think that the concepts so central to philosophy (like knoweldge, obligation, free will) are precisely those concepts that no one has a particularly complete grasp of. Why else is philosophy so hard?

Well, it's hard for lots of reasons. Conceptual analysis is hard because it's hard to articulate the features we're recognizing when we make knowledge-attributions.

I'm not claiming, by the way, to be in possession of a complete undersatnding of KNOWLEDGE. Just enough to be sure of some Gettier case that it doesn't fall under it.

Joachim Horvath said...

Hi Jonathan,

I was puzzled, too, about how certain you seem to feel concerning your Gettier intuitions. You responded to that worry in the following way:

Well, any time a bunch of people seem to disagree with me, I can either give up my belief, attribute to them a confusion, or figure we're talking about different concepts. I guess I just think that last option sounds really attractive in the imaginary sort of case you suggest. It seems like, given how fine-grained our concepts are, disagreement in cases like this would be just the way to recognize that different people are using different concepts.

Now, it clearly won't do to attribute confusion to a bunch of major figures in epistemology, if they really did disagree with you. That leaves you with just two options: (1) Deference or (2) difference in concepts.

(2) seems to be a strange move, though, because according to it every controversial debate in philosophy where differing intuitions are at play (e.g. does Norman know, or TrueTemp, or Fake Barn Fred?) would not be a case of substantial disagreement at all, but merely indicating some conceptual difference or verbal disagreement.

Thus, (1) seems to be the only plausible alternative, given that you're facing some massive multiply corroborated testimonial evidence that contradicts your own Gettier intuitions. So, you could only coherently insist on your intuitions here, if you actually did regard them as infallible (or, alternatively, your intuition that in such cases you are entitled to assume a difference in concepts). If that really were AAs position, then EE just is someone who (reasonably) denies infallible a priori intuitions!

Eric Schwitzgebel said...

Interesting discussion! Maybe Jonathan and I disagree less than I thought.

Here's a possibility, Joachim: If the dispute is not just an empirical one about what our concepts antecedently are (and thus, I still think, amenable to polling), then something like your #2 is going on; and maybe that's the more interesting way to take it, anyway.

But then how do you decide between competing concepts? By appeal to the pragmatic advantages of adopting one rather than another -- i.e., its usefulness for other projects we're engaged in. Although some philosophers seem to adopt this perspective, most don't, instead announcing that one approach or another captures the real metaphysical (or whatever) facts. And that's a key metaphilosophical issue on which I'll depart from most of the "Archies" in this discipline.

Jonathan Ichikawa said...

Yes, the view I'm suggesting is basically one like the one Eric is articulating.

On this point, though:

But then how do you decide between competing concepts? By appeal to the pragmatic advantages of adopting one rather than another -- i.e., its usefulness for other projects we're engaged in.

I guess I think this is right, but I'd put it this way instead: once you realize you have different concepts at work, be careful not to confuse the two -- maybe use a different word for each -- and you'll realize you agree on more than you thought you did, and can move forward from there. (I wouldn't worry too much about whether one concept is 'pragmatically better' than the other or not; that will all sort itself out when we use our concepts in fruitful theorizing.)

Although some philosophers seem to adopt this perspective, most don't, instead announcing that one approach or another captures the real metaphysical (or whatever) facts.

This, I just don't get. Maybe your sociological claim is right; I'm inexperienced enough not to know. But I just can't see how that line would go. "I'm using one concept and calling it 'knowledge', and you're using a different one and calling it 'knowledge', and mine is the one that gets the metaphysical facts right"? What does that even mean?

I guess he could truly say "...and mine is the one that corresponds to knowledge", but of course he knows that his dialectical opponent can also truthfully utter that sentence.

Joachim Horvath said...

But still, both of you - Eric and Jonathan - say preciously little about what justifies such a view in the imagined situation. This was a case where all the experts (or maybe even all other people) say that Gettier cases are (intuitively) instances of knowledge. Then, I guess, the deviating individual has at least the following two options:
(A) My grasp of the concept of knowledge is incomplete or deficient.
(B) The concept that I associate with "knowledge" is another concept than the one the others associate.

You opt for (B), but I still would like to know how that move is justified in such a situation (given that there are reasonable alternatives available)?

Eric Schwitzgebel said...

Why couldn't (B) be the most reasonable option, in many cases? It could just be a feature of my psychology that I'm different; or I could choose to be approaching the concept as a technical one that I think there are good practical reasons to shape somewhat differently from the concept usually associated with the word in question. No?

Joachim Horvath said...

If all the experts say that whales are mammals but I think they are fish, or all the experts say that electrons have negative charge yet I think they have positive charge, then it seems most reasonable to defer to them instead of assuming a difference in concepts. What is so different about philosophical concepts that here it might "in many cases" be reasonable not to defer? Wouldn't this, on the one hand, avoid substantial disagreement way too easily and, on the other hand, just prevent learning and the correction of errors? Of course, there are situations where it is fully justified to ascribe different concepts, but I think of this move as something like a last resort and not as the first option (or one of the first options) to consider.

Eric Schwitzgebel said...

Well, you have a point! This starts to get into Burgean issues of externalism, I suppose, and issues about the extent to which we can and should have authority over our concepts. Tough issues! But I am inclined to think that technical concepts are more subject to authority and control than the concepts of ordinary life.

Jonathan Ichikawa said...

Yes, I'm in full agreement with Eric in these last few posts. (Sorry for not checking in for a little while!)

I think that some examples should make this really clear. The example Ben Jarvis and I use in a paper we're working on on these issues is drowning. If we did a bunch of thought experiments and discovered a bunch of people who said that if you submerged yourself in orange juice and it filled up your lungs and you died, you wouldn't count as DROWNING. Maybe they think you can only drown in water, or maybe only clear liquids, or something like that.

It seems that in this case, the only plausible interpretation is that they have a different concept at work.

If you agree about drowning, then you think that some concepts are like this. Which ones? Good question. But I think that knowledge is a lot more like drowning than like electron charge.

Joachim Horvath said...
This comment has been removed by the author.
Joachim Horvath said...

Disclaimer: Sorry, I just realized that I misread Eric's last post (yes, I am not a native speaker!). So, here's my corrected comment:

Jonathan, I agree that your drowning case is quite likely a case of differing concepts - maybe even a case of different speech communities. However, you've changed the original example significantly. Because this was supposed to be the case of an individual who deviates against all the others. And the question was: What is the most rational option from her point of view? And my concern was: I don't see why deference is not an equally good or maybe even better option than ascribing a different concept to all the others guys. I still don't see why.

Joachim Horvath said...

Hmm, maybe I haven't misread Eric after all and he really claims that deference is more plausible in case of "concepts of ordinary life" like "drowning" or "knowledge". But then, Jonathan doesn't seem to agree with him at all because he claims exactly the opposite, namely that deference is more plausible in case of technical concepts like "electron". Maybe that's what confuses me here...

Eric Schwitzgebel said...

Nice point, Joachim -- maybe that is a difference. I'd probably rank deferences like this:

(1.) [most deferential] concepts like 'electron' which are technical but outside your specialization

(2.) [medium] ordinary concepts like 'drowning'

(3.) [least deferential] concepts in your specialization that you control to some extent, like 'introspection' as used by me in an essay in which I discuss how reliable we are as introspectors and stipulatively characterize what I mean by the term 'introspection'; if another philosopher uses the term very differently we might well be talking past each other with different concepts than simply disagreeing about the same concept.