Why do you endorse Platonism about numbers, or idealism, or skepticism, or consequentialism, or any other a priori justified (you think) philosophical opinion? I suggest that you look not to the philosophical truth of that view, with which you are somehow better in touch with than those with whom you disagree. Rather, look to your own biases, personality, and background.
The "strong program" in sociology of science holds as a methodological precept that in explaining the rise and fall of scientific theories one may not appeal to the truth or falsity of those theories. So also, I recommend, in thinking about the origins of your own and others' philosophical views, you would do well not to think about the truth of those views. You would also do well not to think in other terms that imply success or insight, such as "seeing the weaknesses" in a predecessor's view, or "recognizing" troubles or advantages of a substantive, philosophical sort, or "proving" or "establishing" anything.
Where philosophy blurs into science or math, this constraint becomes blurry. I say this because I think -- unlike the cartoon version of sociologists of science -- that in science and math features of the world (or of mathematical structures) start to apply strong pressures on views. The farther toward a priori philosophy, the less influence from such pressures. The farther toward concrete science or formal mathematics, the more the influence from such pressures.
No a priori claim is so patently false that no philosopher would endorse it. No constellation of views is so bizarre and self-contradictory that no philosopher would sign up for those views, were her background motivations and culturally-given assumptions right. If there are any a priori philosophical truths among the factors influencing our philosophical theorizing, their influence is modest.
Thursday, June 24, 2010
Psychology of Philosophy: The Strong Program
Posted by Eric Schwitzgebel at 1:38 PM
Subscribe to:
Post Comments (Atom)
45 comments:
"I say this because I think -- unlike the cartoon version of sociologists of science -- that in science and math features of the world (or of mathematical structures) start to apply strong pressures on views."
And why do you think this?
It seems likely that your own methodological advice requires you to give the explanation in terms of your own realist prejudices.
In case of 'platonism' among mathematician, I looked at it here. From the feedback I think I can say that it fit's quite well the way, leading mathematicians perceive their doing. An interesting new film on that, here (2nd talk) a fascinating talk of a leading math.-physicist
Why, Professor Schwitzgebel, of course that's just what you would say...
I quite agree that to completely understand why a belief is held, we often (or always?) have to look beyond the mere truth of a belief. And just as you say, that scruple holds even more true the further away from empirical verifiability the believing gets.
But it hardly follows from those observations that the truth is irrelevant in analyzing the sociology of belief. The converse of David Bloor's claim "there are no context-free or super-cultural norms of rationality" is that there are no colorably "rational" norms of belief that are completely unconstrained by facts. (A related point: the discovery of new, successful theories has profound effects on the very socio-epistemic conditions that strong-programists claim are determinative of belief in the first place. Cf. heliocentrism and evolutionary theory.)
Also, what P.D. and Toby said.
Thanks for the comments, folks!
P.D.: I think it because the empirical evidence suggests it. (But why am I an empiricist at all? That's a different question!) In what way does the empirical evidence suggest it? Well, the too-short version is: Over time, people with diverse psychological motivations appear to converge toward a view.
T: Many of the best 20th century logicians and philosophers of math were not Platonists. It's a mistake to portray it as the consensus.
Toby: Naturally! Of course I am ready to whip out arguments and I have the illusory phenomenology of being driven by insight into the philosophical truth....
Michael: What you say seems broadly true to me, except in a priori philosophy. It's not clear whether there are the right kind of facts to do the constraining work; or if there are, they seem to be largely free of influence. Take Platonism: Why does it rise and fall? Why is person X and Platonist and person Y not? I don't see a causal story here from the truth or falsity of Platonism to this pattern of endorsements.
You're on your way to being a Nietzschean now!
So first you say:
"I recommend, in thinking about the origins of your own and others' philosophical views, you would do well not to think about the truth of those views. You would also do well not to think in other terms that imply success or insight, such as "seeing the weaknesses" in a predecessor's view, or "recognizing" troubles or advantages of a substantive, philosophical sort, or "proving" or "establishing" anything."
P.D. Magnus asked you about the origins of your view here. What an excellent opportunity to follow your own advice! However, you say:
"I think it because the empirical evidence suggests it."
Ah, so the evidence "suggests" your view, i.e. "suggests" that your view is true. But aren't you here thinking about the truth of your view? Aren't you thinking here in terms that imply success or insight? Aren't you recognizing advantages of a substantive philosophical sort? Don't you think that the empirical evidence you refer to "proves" or "establishes" the view you hold, to at least some degree?
I think it's pretty obvious that the answer to at least one of those questions is "yes." But if so, you're failing to take your own advice. That's bad news bears! Curse those social pressures that moved you toward self-inconsistency! ;-)
"I think it because the empirical evidence suggests it."
The problem is that drawing substantive conclusions from evidence always involves ampliative inference. A slightly more sceptical empiricist might refuse to accept what convergence there is as a sign of anything. A slightly more credulous empiricist might interpret disagreement and differences among developed beliefs as evidence for a pluralist metaphysics. The question is not just why you are an empiricist, but rather why you interpret evidence and draw conclusions in the way that you do.
Eric, you write,
It's not clear whether there are the right kind of facts to do the constraining work; or if there are, they seem to be largely free of influence
But surely there is convergence on some apriori philosophical positions. There are hardly any Meinongians around, for instance, so there is clearly (even!) metaphysical pressure away from certain views, and there are hardly any genuine modal realists around. On the other hand there is hardly anyone who would claim that non-human animals have little or no moral standing, so there is pressure away from ethical positions too. How many philosophers think that the thin libertarian morality of, say, Morals By Agreement is even approximately true? I'd say one, if Gauthier still believes it. So there's pressure away from theories of justice as well.
Of course there are crazy outliers who will believe/defend just about any view you name. But that's true in the sciences, too. Right, there are a few more in philosophy, but there is also considerable pressure in philosophy away from such views.
Hi Eric, I meaned mathematicians, esp. leading number theorists and algebraic/arithmetic geometers. I am pretty sure that the described mindset is very common and in many cases stronger than expressed in my blog among them. Have contemporary philosophers written anything of interest on that issue? The few texts I saw did not give me the impression that the authors know what they talk about.
I would be interested in hearing what you expect the results of such a program to do. Do you think that the results could help to answer the questions that philosophers have been asking? Do you just find those "a priori" questions to be meaningless or unanswerable?
Also, wouldn't such a strong program require a certain a priori methodological justification? Or will the proof be in the proverbial pudding? Or is Psychology of Philosophy self-justifying?
well, sometimes scientists show up at our door looking for some sharpish world-view to bang with visual memories in search of a model. Sometimes you get Einstein reading Mach, or Turing reading Frege. Unfortunately, other times you get Friedman reading Ramsey ...
Eric, I'll endorse what you're saying in the original post except for calling it the psychology of philosophy. I prefer to call it the psychology of opinion mistaken for philosophy. Most people who call themselves philosophers are not philosophers at all but academic hacks.
@P.D. Magnus - You should be nice and try to understand Eric as doing philosophy instead of psychology -- unless you're doing psychology instead of philosophy.
@T - Platonism refers to the presumed transcendental objectivity of what is counted, but mathematics doesn't count anything in particular: it forthrightly excludes all objective predication from its pure operations. So if leading mathematicians believe their discipline leads us to Platonism, I am left with the impression that they don't know what they're talking about.
Eric, I took you to be endorsing the strong program generally, but I see now that I didn't mark your second paragraph well enough.
Anyway, Platonism is an interesting case, and I'm inclined to agree that there's no causal story that would explain Platonism-belief on the basis of Platonism-truth - not least because I think Platonism is untrue. (!)
On the other hand, I do think there is a causal story for Platonism-skepticism that at least appeals to truth-correlated methodological principles, like parsimony or abduction, principles that have proven reliable under empirical pressures. Part of the abductive analysis I have in mind is comparing the kinds of psychological pressures that would motivate Platonism with those that would motivate skepticism about Platonism, and assessing which are more plausibly active. (Possibly unfairly, and definitely without any hard sociological evidence to back me up, I'm inclined to view Platonism as a quasi- or crypto-religious theory that appeals predominately to those harboring otherworldly preoccupations. Thus, Platonism-belief tends to betray a yearning for something more than what the natural world has on offer.)
I'm not sure I'm clear on what your view is about the scope of the "strong program" claim. In your original post, you seem to suggest that it applies to those philosophical doctrines that are "a priori justified (you think)."
But in your reply to P.D. you seem to suggest that the line of demarcation is between questions on which there is consensus and question on which there is no consensus.
If your claim is that the strong program applies, intrinsically, to philosophy but not the sciences, then I dont see what your argument is. If your claim is that the strong program applies to questions on which there is no consensus, then I guess I dont see what is "strong" about such a program.
The strong program in sociology of science was, after all, a hypothesis about the causal origin of, precisely, consensus in science. It could be read as saying, "if you want to understand why there is consensus around a particular scientific hypothesis, dont look to the truth of the hypothesis. look to social factors that might explain why people might reach agreement that have nothing to do with epistemic or semantic characteristics of the hypothesis."
But if your "strong program" applies only to issues for which there is no consensus--then what analogous thing, exactly, is being denied? If there is no consensus, what is it that epistemic or semantic factors could have explained in the first place?
What you seem to be saying is that when there is consensus, explain that consensus via epistemic and semantic properties of the hypothesis, and when there is divergence of opinion, explain that in terms of the failure of norms of rationality to point to a correct answer.
What is "strong" about such a claim?
@Eric @Michael Drake - Regarding causal explanations detached from the presumed truthfulness of Platonism, there's more to look at than forces that might strike you as positive recommendations for endorsing that system. You could certainly endorse Platonism out of sheer spite for its detractors, which might include yourself if you chose instead to follow your natural inclinations. You could also endorse it with complete indifference to your natural inclinations whether they compel your endorsement or not. The truth of Platonism then becomes a matter of "being true to" to that system, however difficult, rather than of that system either "corresponding to" reality or having some pragmatic application.
Unfortunately my biases, personality, and background lead me to think that this theory is false. ;-)
Among mathematicians, 'platonism' refers usually to a mindset and mentality (as described in the link given earlier), not to ontological statements, just to clearify that. Attitudes like that have of course no infuence into which theories or concepts are accpeted, but have a very huge impact on which concepts and theories are looked for, which concepts and proofs are taken as satisfying. Aside the amazing success of 'platonism'-induced research programs, the experience of doing mathematics stabilises and justifies 'platonistic' attitudes. Interestingly, the concept of 'platonism as mentality', in contrast to it as set of ontological opinions, fits very well to it's ancient roots.
Anon 12:41: I am surely partly guilty as charged. And yet please note that to the extent I am citing empirical evidence I am not making an a priori philosophical claim - so I don't think I'm being as baldly contradictory as you seem to suggest.
P.D.: Right. I suspect it's an intermediate case.
Mike: I think convergence is better evidence than near-unanimity. Near-unanimity can be explained by the presence of similar psychological factors, convergence less easily so. But even convergence can sometimes be explained sociologically, so it's really the broad pattern across issues that is telling. And across issues I see no pattern of convergence.
I'm not sure whether I should recognize that some positions will be unpopular and unattractive on the a priori philosphical merits. If I were to accept that, then I might qualify the claim by saying the advice is sounder when considering live, attractive positions. On the other hand, maybe to the extent some positions are widely unattractive it is due to some combination of shared psychology and empirical and math or math-like pressures.
T: I meant Platonism ontologically, so we may have been talking at cross-purposes.
J. Vlastis: I don't think the strong program will answer traditional philosophical questions directly, though in some cases it can play a debunking role. Traditional philosophical questions must be answered by some combination of empirical evidence and pragmatic choice or they remain beyond our grasp.
That's a methodological claim, one that I think is partly empirically justified and partly a reflection of my own psychology (to step back a bit). So, too, I think are other methodological claims.
Benj: A helpful source of evolutionary noise!
Badda: Thanks for the helpful comments. I don't think of my colleagues as hacks, though -- simply a bit unrealistic about what is behind their (often interesting, beautiful) views.
Michael: Those are interesting point about parsimony and abduction. On parsimony: This is surely a psychological pressure at least in part. Whether simplicity is related to truth, also, is a vexed question in philosophy of science. Even if simplicity does predict truth for scientific theories, the generalization to a priori philosophy is dubious.
On abduction: Yes, I think the psychology of philosophy can sometimes play a debunking or skeptical-genealogical role.
Eric: I think it is an empirical fact that a priori philosophy does not converge, while science does. Thus, my claims are limited to the former. Before about 1600 C.E. I would apply the same methodological recommendation to most broad scientific theories and cosmologies.
Stephen: I'm glad. Without people like you, where would all my good subjects be?
Brian: I wrote my undergrad honors thesis on Nietzsche. I found Beyond Good and Evil so stimulating I could only read it in short bursts. Zhuangzi and Feyerabend were my other favorites.
Surely, those facts, and my current predilections, admit of psychological explanation.
Hi Eric,
Hope this is not redundant...
Can you say more about why you think that, "in science and math features of the world (or of mathematical structures) start to apply strong pressures on views"?
By what mechanism is this "strong pressure" applied?
@Eric - I think it's more interesting (and truthful) to think of your colleagues as hacks and to endorse their views anyway. That way you're unburdened from any personal hangups which compel you toward their views as if by historical necessity, and you're free to think about them more creatively
Brad: Among my reasons for making that assertion is that science and math seem to compel convergence among people with different backgrounds and initial attitudes, though I don't think convergence can be the whole story. The main mechanism that produces convergence in science appears to be the proper sort of sensory exposure to the world, under conditions expected to provide evidence for or against a theory (though again this is too simple to be the full story).
In math (and the math-like aspects of logic), my inclination is to think that something like rule following is at the heart of the matter. It must be relatively clear what counts as following or not following the rules, and it must also be the case that by following the rules one can settle questions that were not antecedently obvious, in a way that can be repeated by other people applying the same general set of rules.
Badda: The problem is that "hacks" sounds too negative to capture my attitude. I hope that viewing them positively is no hamper to skepticism about them. For example, I enjoy reading David Lewis but am not much tempted by most of his views. (Not that he's necessarily an example of the kind of "hack" you had in mind.)
@Eric - Why would you want to capture your attitude? Seriously. Are you doing philosophy or psychology?
@Eric - Point being, when you try to capture your attitude, you're really just surrendering to a psychological reckoning of your biases. Screw that. You're a philosopher. You don't need to answer to psychologists.
@Eric. Thanks for the clarification. I think that there is an especially important role for the kind of program that you espouse in uncovering hidden assumptions of many theories. However (and I suspect that you agree with me here) this cannot be the whole story. We are still left with, I think evaluating the truth of these claims when it matters most. At least the way that I read Nietzsche on this is that he wants to expose unwanted assumptions necessary for certain beliefs and moral attitudes. But I think that in order to even be able to do this, you need to do some of that nasty "a priori" stuff that you don't want to do in order to be able to somehow ground the results in science/mathematics (or at least show their coherence). If such a strong program is going to be successful, it must be to a certain extent self-reflective both with regard to empirical and a priori findings.
Badda: Maybe so!
J. Vlastis: Those are thorny issues. I still haven't entirely settled my attitude toward the a priori -- partly it depends on what the most useful way is to conceive of the "a priori". If appealing to the "a priori" just means that you have to start somewhere with some assumptions, then of course we need it. Obviously, most philosophers think the a priori involves something more substantive and philosophically deep than that -- something immune to empirical refutation -- and that's where I start to get suspicious. Even mathematics and formal logic are, I think, better thought of as involving optional rules in a symbolic dance or game (such that different sets of rules might, as a matter of empirical fact, be more usefully applied to some domains than others) than as robustly substantive truths. But is that a view I hold on a priori grounds? And is it a priori that *if* the empirical facts are such-and-such, then such-and-such a formal system will match in such-and-such respects? Or do we take the pragmatist dance/game maneuver to the higher level too? What a mess!
I have not yet unsnarled this knot to my own satisfaction. That's one of the reasons I spoke of the strong program as a "methodological precept" in the original post.
Conc. "in science and math features of the world (or of mathematical structures) start to apply strong pressures on views": A good example is described in this survey talk. It is about work on the Langlands program, a typical 'platonic' program, which e.g. had lead to the proof of 'Fermat's Last Theorem', and how one tries to adapt it's formulation to a special situation. (Your colleague at your university John Baez could tell you more such cases in mathematics and math. physics.) This special situation turns out to have (not entirely) unexpected properties which defy a direct transfer of the 'usual' methods. Such programs are a bit like the Apollo program of the Nasa: The `platonic`insights first show like sensorial inputs that there is something, e.g. the moon, and on this further work on how to get there (esp. where 'there' is) is based. Some initial successes on the first steps strengthen this, missing or misinterpreted informations cause wrong estimates, e.g. the founding members of the Royal Society estimated to be able to travel to the moon within a decade, by Montgolfieres, which become corrected in the process of work. Personal attitudes play just a small role during that work, but in the sensitivity and judgement involved in perceiving 'platonic' concepts and turning them into something meaningfull (falling apples were seen be everyone, only Newton saw 'gravity' then). As said, ontological ponderings are not of interest in that process, like in the space programs: If the Apollo spacecraft would have found nothing to land on, or a moon of green cheese, that would have not changed the attitude towards reality.
As a baby philosopher, I really have nothing of worth to contribute the discussion, but I at least feel like pointing out that no one has addressed Prof. Switzgebel's question, which I read as a personal question about why we endorse the ideas we endorse knowing we cannot simply toss the question off and say "because X is true / (more) correct." B.L.s comment reminded me of a quote he put in a recent paper about Nietzsche and moral skepticism:
"They all pose as if they had discovered and reached their real opinions through the self-development of a cold, pure, divinely unconcerned dialectic…while at bottom it is an assumption, a hunch, a kind of “inspiration”—most often a desire of the heart that has been filtered and made abstract—that they defend with reasons sought after the fact." (BGE 5)
Now I understand that we're not necessarily talking about moral philosophy, so take the comment for what's its worth.
Maybe because I am still an idiot, and very passe and continentally-oriented, and a hack, thank you, I can answer the question. But at the end of the day I simply find materialistic thinking boring and cold; I like playing with solipsism because we are in our heads, after all; finally, I begin from the position that my existence is a beautiful and special thing, because it is to me. I think "philosophy" is largely justifying and validating my own bullshit, and I might very well be a shithole philosopher, but in that case I'm definitely in good company... Reading the critical responses has been interesting, but I'm much more interested in having commentators actually attempt to answer E.S.'s question. Or did everyone else actually come to their positions via that cold, pure, divinely unconcerned dialectic? If so, how did you pull that off?... And hate on psychology and recede into philosophical chauvinism, but Freud might have something to say about this massive, collective question-dodging...
Conc. Nietzsche: If you are interested in Nietzsche, you surely wondered about what he had in mind when discussing the presocratics as model. Here a video talk on someone who probably fits best to that type of thinkers. Really interesting too are his refections on reading, writing and creativity, here a discussion and comparison with more well known concepts.
@Anonymous - Solipsism is nice. We could use more solipsists in the world if it weren't for the fact that there can be only one. What does that say about a Freudian diagnosis of (y)our question-dodging? ;-)
You appear to advocate applying the methodological precepts of the strong program in sociology of science to philosophy (or at least those areas of philosophy that don’t traffic in science or math). Thus it appears that you endorse:
Methodological Precept (MP): When explaining why a certain philosopher holds or abandons a particular view V (on some non-mathy/sciencey philosophical topic), we should not invoke explanations that appeal the truth or falsity of V (or entail the truth or falsity of V).
In support of this, you contend:
Possibility of Absurd Views (PAV): “No a priori claim is so patently false that no philosopher would endorse it. No constellation of views is so bizarre and self-contradictory that no philosopher would sign up for those views”.
But it’s not clear to me how PAV provides support for MP. In fact, I am inclined to agree with (most of) PAV, but reject MP. After all, the mere fact that it is possible to endorse any philosophical view does not show that people never adopt/abandon their views in light of the truth-values of those views. In fact, it seems to me that the following situation frequently occurs:
At t1, a philosopher S holds an inconsistent view V (about some non-mathy/sciencey area) without recognizing that V is inconsistent. At t2, S realizes that V is inconsistent (either through her own reflection, or because the inconsistency has been pointed out to her). At t3, she modifies V in light of this discovery, adopting a variant of V – V’ – which is in fact consistent.
In order to explain S’s transition from V to V’, I made use of expressions that entailed the falsity of V. After all, “realize that” is factive; hence if S realizes that V is inconsistent, V is inconsistent. Furthermore, if a set of propositions S is inconsistent, it cannot be the case that every member of S is true. But, according to MP, we should never explain why a philosopher abandons a view (on some non-mathy/sciencey area) in terms of the truth or falsity of that view. Thus, it seems in order to defend MP, one needs to insist that the situation described above never occurs or that the explanation provided was somehow defective. This seems to me extremely implausible.
This gives rise to a more general question: do you think that we can ever explain why someone holds or abandons a non-mathy/sciencey view in terms of factive states or processes? For instance, can we ever explain the fact that a person believes q (where q is neither a mathematical nor a scientific proposition) on the grounds that a person antecedently knew p, and deduced q from p? If so, then it seems that at least some of our correct explanations of why people hold the views they do will be couched in terms that entail the truth or falsity of those views.
Finally, in response to Badda Being’s comment at 9:33 PM PDT, it’s simply not true that there can only be one solipsist in the world. A solipsist is someone who believes that solipsism is true – i.e. that he or she is the only person who exists. There is nothing to preclude the possibility of a world filled with solipsists – that is, a world filled with people, each of whom (falsely) believes that he or she is the only person who exists.
Anonymous: Thank you for your thoughtful comment!
I agree that the possibility of absurd views is not, by itself, much of a justification for the strong program. I did not take myself, in this post, to be presenting an argument for the strong program. However, I do think that the frequency with which philosophers appear to endorse absurd views when the psychological pressures on them are right, is a consideration in favor of the strong program. I think also, perhaps, that you mistake the intended force of the methodological precept: It is a recommendation not to try to explain philosophers' changes of opinion by appeal to truth, not a claim that no such explanations are possible.
On your last point: I think simple deduction has similar force to math, so to the extent that is what is driving the philosopher, the philosophers' views will have the same kinds of explanations as the mathematicians'. However, I think that is rarely what is going on behind people's most basic, substantive philosophical positions. For example, from P and P->Q we don't always derive Q. If -Q seems plausible, then we might instead reject P or P->Q. It is in that balancing of plausibilities that philosophical reflection loses the convergent force of mathematics.
Post a Comment