All contemporary science evidentially depends on folk psychology. For instance:
- it is assumed that other scientists tend to say what they observed, and "observation" is a term of folk psychology; the evidence that they tend to say what they observed is based on evaluation of their motives;
- it is essential to the modern scientific enterprise that one be able to assume that technicians are doing what they claim, rather than producing equipment that displays fraudulent results; but the evidence that they are doing this is based on theories about the technicians' motives;
- the concept of reporting depends on folk psychological concepts; if nobody has intentions, nobody reports any results; but the reports of others are central to contemporary scientific practice.
17 comments:
Thus, those who are scientifically motivated to attack folk psychology are cutting the branch on which they sit.
Sure, but if the theory of gravity depends on folk psychology, what reason do they have not to cut? ;)
Much as I enjoy some other stuffs on your blog, these assertions are completely unconvincing, Prof. Pruss.
The evidence that other scientists tend to say what they observe (as opposed to what they guess, speculate, fabricate, etc) is that the results they report are replicable, or validly/cogently inferred from replicable results.
Replicable, convergent experimental results are also evidence that technical equipments used in the experiments are reliable as they are claimed to be. That they are reliable is the best explanation of why they produce replicable results.
As for reporting, what has intention to do with that? My watch reports the time, in any plausible sense of the term.
Alex - I take it that a few scientist will make the move Quine does in _From Stimulus to Science_, to try to reduce the notion of observation as something involving anything mental. I don't think it works, but it is at least worth acknowledging that many reductive physicalist (and eliminative materialist)are aware of this problem.
Gill -- I take it that Quine would think this argument is, at the very least, troublesome.
Gill:
In-principle replicability absent actual replication does no epistemic work.
Many scientific studies never get replicated, because either the author is trusted, or it's too much trouble to replicate, or it's too expensive to replicate. Except in the case of something really groundbreaking, there is very little payoff to replicating someone else's work: you don't get to publish a paper that says "Yeah, I did what Jones did, and got the results she did"; it doesn't help you get to tenure; etc.
Moreover, it wouldn't affect my argument if research always gets replicated by other scientists. Jane Scientist herself will typically have replicated only a tiny portion of the research she's relying on. The rest she has to take on trust, on the basis of some claim like: "It is unlikely that scientists X, Y and Z would all collaborate to lie in this way." But that claim depends on folk psychology.
As for evidence about technical equipment, I think that as the equipment gets more sophisticated, we have more reliance on people. Let's say Jane uses an oscilloscope in her experiment. She is tacitly relying on none of her technicians hacking the firmware of the oscilloscope to give the wrong results when and only when she is doing this experiment. A clever technician could do this. But, as we say, why would the technician bother? But that's folk psychology.
As for reporting, watches' reports are parasitic on the intentions of watch-makers and/or watch-users.
Besides, scientists need to distinguish between reports and jokes. The distinction between a report and a joke very likely relies on intentions.
And, finally, there is the issue of fraud detection.
And even when a scientist has replicated a result, in her later reliance on the result, she is depending on her memory, and that involves folk psychology, too. Can't get away from it!
Prof. Pruss,
So at least we are agreed that the fundamentals of science are backed up by actual replication. But three more points (the last is the most important):
Point 1. Results can get replicated (confirmed is a better word) or found unreplicable quite inadvertently. X pursues his own project relying on Y's result, and it works out the way he expected. In some relevant sense X has "replicated" Y's result. If it doesn't work out as expected, X may come to doubt Y's result, though he may not. But with enough non-replication like this (esp. by independent researchers), people may finally try to actually test Y's result.
Which leads to point 2. I don't know how often doubts lead to actual attempt to replicate, but it's obviously a possibility. This possibility alone, plus scientists' (often well-founded) belief that they have far more to lose from their fraud or negligence being detected than to gain by getting away from it, should provide them sufficient incentive to be honest and conscientious. And scientists know that each other knows this. So they have reason to believe that each other are honest and conscientious. And they are often right.
Well, I have used a lot of folk psycho concepts above. Now kick them away.
Point 3. People who don't like folk psychology don't like its concepts. They don't deny there is a reality (brain states, behavioral patterns, or whatever) the folk theory is trying to describe. Just that "intention", "belief", etc. are not the right vocabulary. So, OK, granted that scientists can't do without thinking such things as "X is intending to inform us of his observations". But for science to function as it does, all we need is that the reality that "X is intending …" is trying to capture, is really there. It doesn't follow that the folk concepts themselves must be the right vocabulary.
If Galileo would have made a similar mistake if he thought: "Mmm, no astronomer can do without the crude telescopes we have, so astronomy depends on the crudeness of the telescope."
The "If" at the beginning of the last para shouldn't have been there.
Re the watch: the watch's reporting the time isn't parasitic on the intention of the watchmaker. The sun can do that as well.
The reporting is parasitic on the user's intention, I should think. But if that was what you had in mind, I would expect you to have said "if no one is taken to be reporting, then nobody reports anything". But anyway, that's not the issue. The issue (sorry if I'm repetitious) is which matters to science: that there is some reality of which folk psychology is crudely true, or the crudeness of folk psychology.
The watch "reports" the time only in an extended sense of the word. It's not an unreasonable extension.
To take the sun to report the time is stretching it even further.
In any case, whether it's the maker's intention or the user's intention, it's still an intention, and that's what matters. (I think that in general the line between a user and a maker is not very well defined. Think of Ikea furniture.)
Even if reporting requires what folk psychology calls "intention", I don't see how this helps your position. Maybe reporting requires something, but is folk psychology right to call it "intention"? That is the very question. If folk psychology is wrong, then scientific practice (even if it depends on reporting) does not depend on people having intentions to report.
That is to go back to my main objection to your view: Point 3, 12:17PM.
So, the scientist's reasoning is something like this:
1. Other scientists say that they have observed that p.
2. Other scientists have an incentive not to lie in this regard.
3. So, probably, other scientists have observed that p.
But if folk psychology is wrong, their scientific beliefs rely on false, or maybe even nonsensical, premises. That sure looks like something that would make the epistemic status of their conclusions dubious.
Or put it more positively: Each time things work out correctly in science--each time things fit--scientists have confirmation of a prediction of folk psychology. So folk psychology is, in the words of my colleague Trent Dougherty, the best confirmed theory we have. For just about every time we confirm any other scientific theory, we thereby also confirming some prediction of folk psychology.
"Even if reporting requires what folk psychology calls 'intention', I don't see how this helps your position?"
Are you suggesting that the predicate "is an intention" applies to an object that isn't an intention? Doesn't that violate the truism that "is an F" (in English) applies to x if and only if x is an F?
I think there is a nice move available in the vicinity, though, and maybe it's what you really mean. There are cases where reliance on a false theory T0 does not undercut scientific knowledge of T, because all one really needed in one's argument for T was the empirical adequacy of T0 (in the van Fraassen sense). For instance, we can imagine that some pre-Copernican chemist relied on a sundial for timing some chemical experiment, and her reliance on the sundial was predicated on the Ptolemaic theory of the solar system. In that case, all she needed was the empirical adequacy of the Ptolemaic theory, and not the truth of the Ptolemaic theory, so that the falsehood of the Ptolemaic theory does not undercut her chemical conclusions. So maybe the scientist doesn't need the truth of folk psychology, but only its empirical adequacy. Another theory with the same empirical consequences would do just as well.
That is a nice move, but an analogue can be made about every scientific theory. When we have no alternative theory that makes the same empirical predictions, and the empirical predictions are highly confirmed, we should assume the theory to be true. Otherwise we have no hope of scientific realism.
Of course, maybe you're not a scientific realist. In that case, my argument won't bother you.
I enjoy asking Eliminative Materialists if they believe their theory is true.
Gill, in your first comment, you say you "enjoy" other parts of the blog. Are you sure that's the right word? Then you say this post is "unconvincing"... again, another word that's going to need to go. I would continue this little game for the rest of your commentary, but, unfortunately, the size of my coment would crash blogger.com's servers.
Prof. Pruss, the "nice move", as you kindly call it, was the move I meant to make. I was saying that the false theory that is folk psychology can still be, and (I believe!) is empirically adequate for the purpose of organizing science conferences, etc. And this empirical adequacy is all that scientists (and the rest of us) need to rely on. Empirical adequacy is the phrase. Thank you very much.
But, scientific realist or no, I suppose one is no obliged to regard as true a theory that is just empirically adequate for the purpose of organizing conferences? Another consideration is how well that theory coheres with the rest of what we believe to be true. Folk psychology does not do well very on this score. (Of course, opinions differ. Suffice it to say that Dennett at el. are not yet proven wrong to their satisfaction.)
Nick, just because I don't think folk psychological concepts are picking out well-behaved theoretical categories, that doesn't mean I think that there are no well-behaved theoretical categories to pick out in the vicinity. In fact I don't. (So, I'm not sure if I deserve the dubious title you conferred on me, but that doesn't matter.) Ok, I have used 'think' twice -- a third word that probably has to go. But why not? I gather questioning engineers why they build bridges using Newtonian physics is not much fun.
What sort of lack of coherence bothers you?
I don't think you would describe the relation between physics and folk psychology as easy. That's the lack of coherence I have in mind.
1. Why is that more a problem for folk psychology than for physics?
2. The relationship between non-folk psychology, or sociology, or mathematics, and physics isn't "easy" either.
Is the first question rhetorical, or do you really want me to go over the whole list? (Short answer: Why isn't there a Nobel Prize or something for folk psychology?)
As for your suggested companions in guilt, let's me just say that they look very different (and maybe they are different): unlike folk psychology, neither math nor sociology nor non-folk psychology threatens to burden our ontology with an extra kind of entity or state or process. To the extent that this is not true, that's because, in the case of sociology and non-folk psychology, they are not entirely folk-psychology-free. In the case of math, well, it will destroy my world view if Platonism turns out true. But maybe it won't.
I thought we were arguing about whether scientific practice depends on (the truth of) folk psychology. What's the situation with debate now?
There is a Nobel prize for folk psychology. It's called the Nobel prize in literature. :-)
As for the relation between folk psychology and physics, there is a general problem about the relationship of all the special sciences and fundamental physics. They all seem to posit entities that fundamental physics knows nothing about: organisms, cells, volcanoes, globular clusters, etc. What makes the entities of folk psychology more problematic? Is it simply that there is no hope of a reductive story? But it's an oddly asymmetric coherence requirement that for theory B to cohere with theory A, theory B needs to reduce to theory A.
As for the debate on whether scientific practice depends on folk psychology, I think we've roughly reached agreement on the claim that scientific practice requires the empirical adequacy of a large chunk of folk psychology.
Given the absence of a plausible empirically alternative to folk psychology, I think it's quite reasonable to say that this shows that scientific practice currently depends on folk psychology. One might, of course, have a hope that there will be a better future theory whose empirical import agrees sufficiently with folk psychology. But that's just a hope or a conjecture rather than a scientifically confirmed belief.
However, I have some doubts whether any replacement for folk psychology that is empirically adequate will satisfy you. The part of folk psychology that science needs includes claims like:
(p)(If C(x,p) and x asserts that p, then probably p is true).
Here, C(x,p) says something about how whether p holds is within x's observational capabilities and x has no good motive to lie or be self-deceived about p. Now, while the range of quantification need only include those p that are observable, it needs to include an indefinite variety of them, because trained scientists constantly come to be capable of new kinds of observational judgments (moreover, we can expect new ways for our senses to be enhanced), and there are all sorts of grammatically and logically complex ways for them report these. So it is prima facie very plausible that any replacement psychology that does the job science needs done will need replacements for propositional attitudes for a wide array of propositions, and I think this is pretty close to the stuff people find problematic about it.
Moreover, it seems likely that the empirical adequacy of folk psychology *includes* psychological claims of a sort that the opponents of folk psychology don't like. For we directly observe (using "observe" in a non-factive sense) assertions, motivations, etc., and hence claims about assertions, motivations, etc. are going to count as part of the empirical import of the theory in van Fraassen's sense, I expect.
Post a Comment