About Rationally Speaking


Rationally Speaking is a blog maintained by Prof. Massimo Pigliucci, a philosopher at the City University of New York. The blog reflects the Enlightenment figure Marquis de Condorcet's idea of what a public intellectual (yes, we know, that's such a bad word) ought to be: someone who devotes himself to "the tracking down of prejudices in the hiding places where priests, the schools, the government, and all long-established institutions had gathered and protected them." You're welcome. Please notice that the contents of this blog can be reprinted under the standard Creative Commons license.

Monday, December 10, 2012

Dogmas of analytic philosophy, part 1/2


1.bp.blogspot.com
by Massimo Pigliucci

If you asked me whether I consider myself an analytic or a continental philosopher, I’d immediately say analytic, but with some regret and much qualification. This particular dichotomy reflects the unfortunate way in which modern academic philosophy is divided. (Very) roughly speaking, post-Kant philosophers have either dedicated themselves to writings of broad concern, a la Nietzsche, Heidegger, Sartre, Focault and Derrida (continentals), or of a very narrow and technical sort, a la Russell, Quine, Kripe and Lewis (analyticals). The major point in favor of continental philosophy is that it tries to matter, with many of the authors in that canon addressing important social and political questions, as opposed to the all too-often legalistic, almost neo-scholastic (and I don’t mean that as a compliment), hair splitting of much analytical philosophy. Then again, most people with a PhD in philosophy can read and actually comprehend an analytic paper, while the same cannot so easily be said of much continental output — especially from the likes of Heidegger and Derrida. [1]

My new book project, which will likely be published some time in 2014 by the University of Chicago Press, deals in part with these matters, the overarching goal being that of addressing the question of whether, and if so in what sense, philosophy can be said to make progress. (Hints: yes, of course it does; no, it’s not in the same sense as progress in science; indeed, it has more to do with progress in logic/math and in literary criticism — now you figure out what the latter two have in common...)

If both analytical and continental ways of philosophizing have serious problems, what is a philosopher to do? Recover the best of both traditions and go back (or forward!) to the way people were doing philosophy before the split, i.e. up to (and including) Kant. One such suggestion comes from a recent column by Paul Thagard, a philosopher at the University of Waterloo and columnist for Psychology Today. Thagard calls this approach “natural philosophy,” and writes that it is nothing new, as it “has been practiced in various ways by such distinguished philosophers as Thales, Aristotle, Epicurus, Lucretius, Bacon, Locke, Hume, Mill, Peirce, Russell (after 1920), Dewey, Quine (after 1950), and Kuhn.” The term natural philosophy is perhaps not the best choice, historically speaking, since it usually refers to proto-science (the sort of thing that Bacon, Galilei, and Newton were doing), but I guess it will do for now.

The main body of Thagard’s essay is a step-by-step examination of what he terms 11 “dogmas” of analytic philosophy, a reference to Quine’s famous “Two Dogmas of Empiricism” paper. (Thagard doesn’t deal at all with continental philosophy, though I’d be surprised if he had a more positive take on it than the one I hinted at above.) It’s worth going through the list, to which I will add my brief commentary, as my position is very close to, but not exactly the same as, Thagard’s.

Dogma #1. “The best approach to philosophy is conceptual analysis using formal logic or ordinary language.”

The criticism here is addressed at an over-reliance of analytic philosophy on formal logicisms, which has produced, among other regrettable things, the rigidity of early 20th century logical positivism. Thagard suggests that philosophy should be in the business of theory construction, not conceptual analysis, but I’ll stick with the latter, as long as we understand it not in the narrow analytical sense of the term, but as an investigation into the “concepts and theories developed in relevant sciences” (and in other branches of human endeavor, I would add).

Dogma #2. “Philosophy is conservative, analyzing existing concepts.” 

The point, according to Thagard, is “not to interpret concepts, but to change them” (a veiled reference to Marx’s famous phrase about philosophy itself?), but I think a better take is that philosophers should be both into interpreting and changing — when necessary — the concepts deployed by philosophy itself and by other disciplines.

Dogma #3. “People’s intuitions are evidence for philosophical conclusions.” 

Ah, intuitions! They have been at the center of some interesting debate already here at Rationally Speaking. I’m certainly with Thagard when he says that intuitions need to be evaluated critically. But intuitions — in science just as in philosophy — are a good initial source of understanding, and the cognitive science literature on the role of expert intuitions (as opposed to a generic, and non-existent “intuitive sense”) is quite clear: if you spend years or decades thinking or doing X, your intuitions about X are much more likely to be correct than a layperson’s intuitions about X. X, of course, can be philosophy or one of its branches. Still, intuitions are the starting point, not the end of the matter.

Dogma #4. “Thought experiments are a good way of generating intuitive evidence.” 

No, they are not, as Thagard points out. But here too one need not go too far. First off, remember that even scientists use thought experiments (Galilei, Einstein). Second, I guess I take the point of thought experiments to be different than some of my colleagues do: when I read about Searle’s Chinese room, or even — god forbid — Chalmers’ zombies, I don’t think these exercises prove anything about the real world. What they do, however, is to make explicit the way we reason about certain things, our intuitions (see above) about a given issue. Then we can poke at the thought experiment and see what is wrong with it, sometimes even using empirical evidence!

Dogma #5. “People are rational.” 

Clearly, they (we) are not. We have more than enough evidence from cognitive science that people come equipped with all sorts of biases in the way they think about all sorts of things, from probabilities to the value of anecdotal evidence. This, however, is sometimes taken (often by psychologists) to mean that psychology has superseded philosophy (even Quine was guilty of something along these lines when he proposed to replace epistemology with psychology!). I’m with Thagard here, where he says that philosophy is needed to educate people, a task that is made more, not less, urgent by the awareness of widespread cognitive biases.

Dogma #6. “Inferences are based on arguments.” 

Thagard objects that instead “inferences operate as parallel neural processes that can use representations that involve visual and other modalities.” Hum, yes, but this strikes me as confusing the process with the mechanism (you know, like my frequent target, Sam Harris does when he thinks that a neural understanding of moral decision making is the same as answering moral questions). I’d say that inferences are a type of argument — and as such they better be logically constructed — but one that involves not just pure logic (the analytic fallacy), but also representations, observations, and so on.

Tune in next time, for the last five dogmas of analytic philosophy.

______

[1] While in this essay I join the common sport among philosophers of criticizing philosophy, those among my readers who are scientistically inclined better not snicker too much. Science itself could use quite a bit more self-doubt and internal probing. When I was a practicing scientist I saw first hand how the overwhelming majority of scientific research is just as useless and self-centered as the majority of any other academic enterprise — with the difference that science is a hell of a lot more expensive to conduct than philosophy or literary criticism! But that’s another story, and possibly another book project...

19 comments:

  1. I would have thought that progress in maths has a lot in common (and goes hand-in-hand) with progress in science.

    Progress in literary criticism? To my mind Aristotle's Poetics still stands up very well, albeit that it lacks the sophistication and richness of much twentieth-century work.

    But really, judgements on these things are so subjective. I happen to like John Updike's essays, for example, but left-leaning readers would be unlikely to warm to his Lutheran sensibility and small-town American conservatism.

    Criticism, like art, is, in the end, profoundly personal – and often ideologically sensitive.

    And why don't you tell us what you think maths/logic and literary criticism have in common?

    ReplyDelete
  2. " When I was a practicing scientist I saw first hand how the overwhelming majority of scientific research is just as useless and self-centered as the majority of any other academic enterprise ..."

    Exempt philosophy?

    ReplyDelete
  3. Mark,

    > I would have thought that progress in maths has a lot in common (and goes hand-in-hand) with progress in science. <

    No, the two are actually quite independent, especially math. Of course math is very useful to (some) science, but rarely the other way around (which is to be expected, since they are very different types of enterprises, and since math problems can’t be settled by observation or experiment).

    > Progress in literary criticism? To my mind Aristotle's Poetics still stands up very well, albeit that it lacks the sophistication and richness of much twentieth-century work <

    Yes, but one needs an account of that additional “sophistication and richness,” and I think that’s where one can find a sense in which even literary criticism makes progress.

    > Criticism, like art, is, in the end, profoundly personal – and often ideologically sensitive. <

    True, but I’m talking about criticism as scholarship, not personal aesthetics or political ideology. These are difficult to disentangle, but I think there is a sense in which it can be done (see your own statement about additional sophistication and richness since Aristotle).

    > why don't you tell us what you think maths/logic and literary criticism have in common? <

    Working on it, but my point isn’t that maths/logic and literary criticism are similar, but rather that philosophy falls somewhere between those activities, especially depending on what sort of philosophy we are talking about (analytic is closer to math/logic, continental to criticism).

    Phoenix,

    > Exempt philosophy? <

    No. Did you read only the footnote and skipped the main part of the post?

    ReplyDelete
    Replies
    1. I read the entire post - compliments to you - but when an item is footnoted in such a way as to suggest that the academic enterprise which underpins it may be useless then I wonder at possible exemptions.

      Delete
    2. >...math problems can’t be settled by observation or experiment<

      Is that *always* true, or only *mostly* true?

      What about the four color problem?

      "The four-color theorem states that any map in a plane can be colored using four-colors in such a way that regions sharing a common boundary (other than a single point) do not share the same color. This problem is sometimes also called Guthrie's problem after F. Guthrie, who first conjectured the theorem in 1852."

      A proof was offered by "Appel and Haken (1977), who constructed a computer-assisted proof that four colors were sufficient. However, because part of the proof consisted of an exhaustive analysis of many discrete cases by a computer, some mathematicians do not accept it. However, no flaws have yet been found, so the proof appears valid."

      That is, the proof was a "brute force" type of proof. Appel and Haken didn't know the result in advance and simple *observed* if the computer was successful in coloring the set of discrete cases or not.

      This, admittedly, is not the way mathematical proofs usually work, but will you admit it is a valid exception to your rule?

      Delete
    3. Tom,

      you are referring to what sometimes is called "experimental mathematics." Two points in that regard: first, it is "experimental" in a very different sense from which science is experimental, so in important respects that's a misnomer. Second, all I need for my point to stand is that *some* mathematics is done with no input at all from empirical evidence, which is most definitely the case (indeed, it is the case for most of mathematics, and logic).

      Delete
  4. It sounds like it will be an interesting work. What philosophical training I had was in analytic philosophy (or a branch of it--largely Wittgenstein, J.L. Austin, the "Oxford School") and pragmatism. I think pragmatism (but not Rortyism) offers a middle path, myself, as to most issues, with Stoicism a useful guide as to "how to live."

    ReplyDelete
    Replies
    1. Pragmatism is nice & all in principle, but in application it simply doesn't work (pun intended).

      Delete
  5. Great post! Just to play devil's advocate for a second, though:

    >if you spend years or decades thinking or doing X, your intuitions about X are much more likely to be correct than a layperson’s intuitions about X. X, of course, can be philosophy or one of its branches.

    If I were a skeptic of philosophy, I would point out that we only expect expert intuitions to track their subject matter well IF the experts are receiving some sort of objective feedback about how well their current intuitions are doing; e.g., a chess player getting positive and negative reinforcement pretty soon after they make a move.

    There are probably not many areas of philosophy (a skeptic would say) in which philosophers are receiving similar constant feedback from reality, telling them whether they are on the right track. That's not the philosophers' fault - it is just in the nature of their subject matter.

    ReplyDelete
    Replies
    1. Ian,

      > There are probably not many areas of philosophy (a skeptic would say) in which philosophers are receiving similar constant feedback from reality, telling them whether they are on the right track. <

      Well, the hypothetical skeptic would thus betray a misunderstanding of philosophy: the discipline (largely) operates in logical, not empirical, space. In that sense, would the skeptic apply a similar criticism to logic (which of course I consider a branch of philosophy) or math (which may or may not be considered a branch of logic — more on this in an upcoming post)?

      Delete
    2. >Well, the hypothetical skeptic would thus betray a misunderstanding of philosophy: the discipline (largely) operates in logical, not empirical, space.

      Nah, I appreciate the difference (after all, the chess players are in logical space, in a sense).

      Point taken in re: logicians, although I think after them you will find the counterexamples harder to find.

      The logicians have the advantage of both proving theorems, and of being able to test their principles against the metaphysical facts (such as they are). "If A then B; B, therefore A" would lead to lots of absurd conclusions, so we reject it. Likewise, mathematicians, can use proof methods to check their intuitions.

      On the other hand, an ethicist (say) has very little to check their intuitions against - arguably, only logical consistency & other intuitions.

      This is not to say that ethics is not a worthwhile enterprise (as you know, I consider it one of THE most worthwhile human endeavours), just that there are no *uncontroversial* ethical or meta-ethical facts to serve as a sanity check for ethical theorists.

      So regarding the trustworthiness of professional intuitions, if a logician tells me that some argument "sounds wrong," yes, that weighs more heavily with me than if an ethicist tells me that some argument "sounds wrong."

      Delete
    3. Maths and logic are rooted in epistemic success, and therefore have a solid starting point. People presumably started counting and doing basic arithmetic because they found that those worked. They had practical value.

      The same can't be said for most philosophy. In my view philosophers have traditionally put too much emphasis on deductive arguments from intuitively obvious premises, and not enough on inference from evidence to the best explanation. The trouble is that those intuitively obvious premises are often wrong or subtly misleading.

      Fortunately many philosophers today are taking a more naturalized, empirical approach to philosophy. Actually, as Thagard says, this isn't completely new. I think David Hume made such a big step in that direction that I like to think of him as the first modern, naturalized philosopher.

      That's not to say that, if only philosophers paid attention to the evidence, everything would be as clear cut as the hardest science. Philosophy is bound to be "soft", because the evidence doesn't speak clearly and needs a lot of interpretation, and the concepts that philosophers deal in tend to be very fuzzy ones. But philosophers need to think as empirically as possible. (And crucially that includes taking a realistic, naturalized view of language.)

      In my opinion.

      Delete
    4. Continuing my thoughts... "Philosophy" is a very broad and fuzzy rubric, under which a huge variety of subjects are grouped. These vary from the most distinctively philosophical and conceptual to the most empirical, which border on and overlap with other fields, including physics, biology, psychology, anthroplogy, cognitive science, etc. If you focus your attention on the least empirical parts of philosophy, you're less likely to see its empirical nature. Also, if your own approach to answering philosophical questions relies heavily on intuition over evidence, then you're less likely to see the empirical nature of philosophy.

      Philosophy tends to require more careful attention to language than other fields, particularly at the more conceptual end of the spectrum. And this is perhaps the thing that makes it look most distinctive. Here I'm with Wittgenstein: philosophy is a battle against the bewitchment of our intellect by our language. On some of the most distinctively philosophical questions, once we unbewitch ourselves we see that the question is meaningless, and there is no question to be answered. I take as an example the question, "Do numbers exist?" I think many people, especially mathematicians, will sense that there's something vacuous about that question. But other cases can be more subtle. The more empirical questions are those where something remains to be answered once we've unbewitched ourselves.

      Although many philosophers recognise the importance of addressing language issues, the traditional analytic approach doesn't address these issues in the right way. Again it relies too much on intuition. We need a more naturalized view of language, that pays more attention to the way language is used in practice, and sees meaning as something very fuzzy, rather than seeing ideal concepts to be uncovered. For example, we won't find the precise concept of "knowledge" if only we look hard enough, because the meaning of the word is inevitably fuzzy.

      Although Paul Thagard and I both favour a naturalized approach to philosophy, I disagree with several of his specific points. Most of all I'm concerned that he seems not to recognise the crucial place of language in philosophy.

      Delete
    5. Richard Wein,

      Re: "Maths and logic are rooted in epistemic success..." Nope. Not true at all. Perhaps the precise logic (there, of course, many from which to choose) to employ in an empirical domain depends on the extent to which a logic conduces to fruitful research programmes, but a logic itself needn't have any connection with the empirical realm whatsoever.

      Delete
    6. Eamon,

      I should have said, "Our use of maths and logic..."

      Delete
  6. Great article,

    I've been getting into frequent debates with my brother over this analytic/continental distinction, with him being more on the continental side and me being much more in line with analytic philosophy. The thing that upsets me the most about many continental philosophers is the absurdly incoherent verbosity they employ (my brother told me to read some Slavoj Zizek and Michel Foucault, and I muss confess that upon reading them I had virtually no idea what they were trying to say). You mention the close association of 'literary criticism' and continental philosophy, which I think is right on the money. If this is the case however, shouldn't these "philosophers" change their titles to "english professors"? Some in the continental school might indeed argue that analytic philosophy is too narrow or too mechanistic, but the advances made in mathematical logic, philosophy of science, and epistemology that have come from the analytic tradition are vastly superior to anything I can think of in the continental school. The real nail in the coffin for continental crackpottery comes from a book I read called "Fear of Knowledge" by Paul Boghossian, who actually had a talk up here at Cornell last semester. The whole Sokal Affair and all that nonsense that science and mathematics are all socially constructed, a view that is associated with continental braches of thought, is misplaced and pretty dangerous when it makes people suddenly believe there are no objective facts out there, and Paul does a pretty good job of dismantling them in the book. I'm a bit biased because I am a big fan of mathematics (recent Platonism article was great), the sciences, and good philosophy (which is why I decided to make it my second major after this past summer). Despite my favoritism, I don't see much objective truth coming from the continental school, but rather a plenitude of theories that can all be drawn from the same reading with no ability to ascertain the correct viewpoint. If there is no correct viewpoint, as a continental might exclaim, then we are simply back to the subjective aesthetic appeals that Mark mentioned in the first comment.

    ReplyDelete
  7. Ian,

    > The logicians have the advantage of both proving theorems, and of being able to test their principles against the metaphysical facts (such as they are). <

    Metaphysical facts??

    > On the other hand, an ethicist (say) has very little to check their intuitions against - arguably, only logical consistency & other intuitions. <

    Not necessarily. Ethics has practical consequences, but yes your point is well taken. I did not suggest that philosophical intuitions are on par with those of mathematicians, simply that in general intuitions about X are more reliable when they come from long time practitioners of X. They still need to be tested / discussed.

    Richard,

    what Eamon said.

    > philosophers have traditionally put too much emphasis on deductive arguments from intuitively obvious premises, and not enough on inference from evidence to the best explanation. <

    I think there is a tendency of overplaying the role of intuitions in philosophical reasoning. And inference to the best explanation is something that philosophers have used and thought about at least since the big induction debate between Mill and Whewell.

    > Philosophy is bound to be "soft", because the evidence doesn't speak clearly and needs a lot of interpretation <

    The use of the term “soft” seems to imply a comparison with science, which I think is misguided. Philosophy is a different type of activity. Yes, philosophers most certainly need to keep in mind the available empirical evidence. The bright ones have always done so.

    > On some of the most distinctively philosophical questions, once we unbewitch ourselves we see that the question is meaningless, and there is no question to be answered. I take as an example the question, "Do numbers exist?" <

    Ah, of course my opinion on that is that you picked the wrong example, particularly given my sympathy for mathematical Platonism (http://goo.gl/V5FcK). btw, contra your statement, I think a lot of mathematicians do not think at all that that question is vacuous.

    ReplyDelete
  8. Excellent article as usual, Massimo.

    One thing that I am curious of is whether you will do something similar with the continental side. Do they have their two dogmas, as it were?

    Also, I am curious to your thoughts on the Sep/Oct issue of Philosophy Now, dealing with the articles on metaphilosophy.

    ReplyDelete
  9. Chris,

    > I am curious of is whether you will do something similar with the continental side. Do they have their two dogmas, as it were? <

    Ah, yes I think they have a good number of “dogmas.” Maybe I’ll get specifically to those in a future post.

    > I am curious to your thoughts on the Sep/Oct issue of Philosophy Now, dealing with the articles on metaphilosophy. <

    Frankly, it could have been done better. I can’t believe that one of them was based on the long ago discredited “bicameral mind” theory. And I’m not much of a fun of experimental philosophy, though I need to pick up that ball again too and write about it.

    Phoenix,

    > I read the entire post - compliments to you - but when an item is footnoted in such a way as to suggest that the academic enterprise which underpins it may be useless then I wonder at possible exemptions. <

    I’m surprised you interpreted it that way. You know I am also a scientist, right? My point was simply that scientistically oriented critics of philosophy shouldn’t be too complacent, since every academic house is in need of some serious re-ordering and re-thinking, at least from time to time.

    ReplyDelete

Note: Only a member of this blog may post a comment.