I have been following Vincent’s spat with Edward Feser (A Catholic philosopher with some reactionary views – his blog) over whether Feser’s own “cosmological argument” has the merit Feser seems to think. Here’s Vincent’s latest post on the matter.
Not being able to post at Uncommon Descent, I thought I might catch up with Vincent at Feser’s blog but I seem to have worn out my welcome. In case anyone decides to pop in from Feser’s blog, I thought I’d offer this thread for discussion. And please regard it as an open thread. Nothing will be considered off-topic. Usual rules apply!
Keith:
I think something needs to be added to that phrase: my understanding is that it requires translation and/or bridge laws of the terms themselves.
Some philosophy defines reduction of science A to science B as the ability to logically deduce the theories of science A from science B. You cannot do logical deduction of one set of theories from another unless you can map the terms between them.
To translate terms, you need to understand their technical meaning in a science. But how does you define a term such specialized terms? One answer is roughly that a term is defined by the way all theories of that particular science use that term Which is why understanding that science could seen be a requirement for understanding the terms and then translating them.
People are talking past each other in the discussion because they have not agreed on what they mean by “reduction”.
First, I think that acceptance of that “regulative ideal” is helpful in justifying methodological reduction, that is trying to formulate explanations of phenomena observed at one level of a science by using mechanisms or explanations involving another. (Although I suspect the need for that type of justification would be of interest only to philosophers).
But my basic difference from your point of view is that I don’t think it is helpful to define reduction solely as logical deduction. Restricting reduction to logical deduction of one theory from another is not what science does. Discussion of whether reduction works needs to consider what scientists actually do.
Successful sciences uses opportunistic, non-deductive reduction, such as reduction by mechanism, so I think epistemic reduction is possible as part of successful scientific explanations without having to reduce whole theories and without logical deduction.
Given that, and assuming you are a scientific realist about the entities in theories of science, then those could be seen as cases of successful ontological reduction.
Furthermore physics would tell you the basic constituents of reality. If you agree with L&R that other sciences are looking at the emergent patterns of the entities of physics, then you have that global regulative ideal you talked about.
Thanks for the pointer. I’ll need to do some background reading to fully appreciate your points.
I’m not surprised that you distinguish the necessity of brains for minds from the sufficiency of brains to explain minds.
But I see brains as necessary and sufficient for minds in some sense.
For example, I think that brains/bodies do provide the necessary and sufficient proximate causes for the actions of minds/persons, of course with the proviso that those proximate causes are embedded in a chain of causes involving interaction with the world. So I am saying brains/minds are also sufficient for proximate, causal explanations of minds/persons.
It is not so clear that will work to explain meaning/intentionality with only brains. How could one do so in the light of Twin Earth type arguments?
Unfortunately, I have some errands to run now, so I cannot provide my answer.
I see this as a potential example of the type of opportunistic reduction I referred to above: that is, there is a research program in cognitive science to explain the behavior of minds/people by reductive neurocomputationalism.
There is an old programmer’s joke regarding the use of regular expressions to solve problems in text processing. (In programming, regular expressions are a powerful but complex and error-prone tool for text processing).
The joke:
1. You have a text processing problem.
2. You decide to solve it using regular expressions.
3. Now you have two problems.
I find myself in an analogous situation.
1. I have trouble understanding a philosophical issue.
2. Someone posts a complex technical piece of philosophy which explains the issue.
3. Now I have two problems.
Note that this is not a complaint, simply an observation.
One thing struck me on initial reading: surely any complete description of reality would be in mathematics, not natural language? So the basic issue would be about philosophy of mathematics, and in particular that subset of mathematics needs to provide a successful, complete physics.
If course, it still remains to be understood what could be meant by accepting the reality of emergent phenomena described by ordinary experience or by other sciences.
Surely not.
Consider the reduction of phlogiston theories to more modern chemistry or of eliminationist attempts to reduce psychology to biochemistry. What gets reduced are the events described by the first theory: you’d need to believe in the existence of the events to be explained but not the theory explaining it.
But those holding the first theory often just respond “You don’t understand! Tou’re missing the EXPERIENCE!” (or the phlogiston). On the Kuhnian/Feyerabendian view, such languages are not completely translatable.
Bruce is right: there can’t be a very sensible discussion of reduction until there’s a common sense of what “reduction” means.
Why mathematics? Do you mean an ideal language of the Tractarian type? That wouldn’t be mathematical only. It would have empirical predicates.
The way I was taught this stuff, phlogiston chemistry is eliminated by the discovery of oxygen. Phlogiston just gets eliminated from the ontology, just as ether did about a century or so later.
“Eliminative materialism,” as a position in philosophy of mind (i.e. the Rorty-Churchland-Rosenberg sense) is the view that our practice of positing propositional attitudes should be replaced by positing propagations across neuronal populations as the site of semantic content. (I’ve actually studied under the Churchlands and read a lot of their work, so I think I have a pretty good grasp on what they are actually claiming — unlike most people who hear the phrase “eliminative materialism” and attribute to the Churchlands whatever view is evoked by hearing that phrase.)
It’s worth pointing out that a historical motivation for the development of eliminative materialism — initially by Feyerabend and Rorty, then picked up by Churchland — was their keen awareness that Nagel-style reductions are extremely rare in the history of science apart from the history of physics itself.
If that ever happens to one person’s practice in the next 500 years, I’ll give you $500 (American).
I’m not foolish enough to take that bet! I was only pointing out what the Churchlands’ view actually is. I don’t accept it myself.
Please relay my offer to them. :>}
Bruce,
What else could the translations and bridge laws apply to, if not to the terms?
Sure, but why is that a problem? Why, for example, should KN be skeptical of the reduction of chemistry to QM merely because the terms are theory-dependent?
Right, but the issue walto raised is whether it is necessary to believe a theory in order to understand its terms. I can’t think of any such cases. Can you?
walto,
Phlogiston theories haven’t been reduced to modern chemistry — they’ve simply been discarded.
Right, and modern chemists believe neither in phlogiston itself nor in the theories that invoke it. They don’t attempt the reduction, because there’s no point in doing so.
For that argument to succeed, you’d have to demonstrate that there’s something scientifically relevant that can only be perceived by those who truly believe the theory, and not by those who merely accept it arguendo.
What is and is not “scientifically relevant” will be a function of the scientific theory in play, no?
ETA: Right, and modern chemists believe neither in phlogiston itself nor in the theories that invoke it. They don’t attempt the reduction, because there’s no point in doing so.
Exactly. As I said, they don’t really reduce the theories so much as the the putative objects or events to items found in THEIR theories. But one who believes in mental events or phlogiston may deny that the reduction has been successful on the grounds that the reducers don’t really understand the terms the older theorists used to refer to those putative entities.
walto,
Yes, but I’ve never encountered a scientific theory that required genuine assent, and not just arguendo assent, in order for its relevant scientific content to be understood. Can you think of any?
keiths:
walto:
The purpose of reduction isn’t to unify good science with bad science. Given that modern scientists reject phlogiston theory, they don’t expect it to be reducible to modern chemistry or physics.
The reductions they attempt are between successful theories.
At which point they can describe precisely what the reduction fails to capture. If they can’t do so, or if they insist that it is ineffable, then their assertion is no longer scientific.
What is “relevant scientific content”? (See below.) Also why are we restricting to scientific theory reductions here? What philosophers are more commonly interested in is reductions of “folk theories” (chairs and minds) to scientific theories–like one of Eddington’s tables to the other one. Reduction of modern chemistry to modern physics is a different matter AFAIK.
That’s generally right, I think. They reduce the putative objects of theories–whether true or false. But they need not be either true or scientific. Scientists want the stuff that worked in the old theories to be replaceable by stuff that works in their new ones. But “reduction” isn’t always necessary for that, I don’t think.
Not necessarily. Stuffing all of Ptolemy into Copernicus or Newton into Einstein and Planck was a matter of reducing theories that were discovered to be unsuccessful for some purposes. Same for elimination of phlogiston. And the question of whether OL “theories” of the world are successful is controversial. I mean, I think they are, but the Churchlands disagree.
One person’s entity requiring scientific explanation (take “subjectivity” e.g.) is another person’s ghostly ectoplasm. That’s why I like the Churchland’s test. It makes it completely clear when I will owe them $500. (Never.)
walto,
Content that helps us judge a theory’s scientific worth, i.e. its explanatory and predictive power. For example, atoms (and molecules) are central to Boltzmann’s statistical mechanics, but it was not necessary for physicists to genuinely believe in the reality of atoms in order to evaluate statistical mechanics as a theory. They could accept the existence of atoms arguendo. Indeed, Boltzmann’s success played an important role in eroding the widespread skepticism regarding their reality.
It’s a pragmatic matter. Folk theories are notoriously unreliable in general, so reduction is more likely to be successful for theories that have been vetted scientifically. That said, any theory is a candidate for reduction, if someone is willing to invest the effort.
keiths:
walto:
They focus their attention on successful or promising theories. They aren’t trying to reduce phlogiston, and in fact they’ve abandoned it entirely. Neither the theories that employ it nor the concept itself are considered scientifically valuable these days. If you’re going to attempt a reduction, it makes far more sense to start with a successful, accepted theory — unless your primary goal is to bolster the scientific credibility of the candidate theory by showing that it reduces to a widely accepted one.
keiths:
walto:
Ptolemaic astronomy and Newtonian physics were both highly successful theories! The difference is that Newtonian physics successfully reduces to a limiting case of modern relativity theory, and is retained in practice, whereas Ptolemaic astronomy doesn’t reduce and has been abandoned altogether.
keiths:
walto:
Which is why it’s good scientific practice to relate your theoretical entities to observables. The holy grail for anti-reductionists is an observable phenomenon that cannot be reduced, even in principle, to fundamental physics.
What made you break your rule about not posting just to argue for that one?
Anyway I try to clarify limited point in a reply to Walto.
Keith
I was just trying to help the discussion by explaining the type of reduction I thought KN was referring to; there is more on that approach in the SEP and IEP if it is something that interests you.
Regarding the discussion between you and walto on terms, understanding, and reduction: I don’t even think scientists do or would want to do reductions in that philosophical sense, which seems to be part of what Walto is assuming for his argument. So I am not sure I understand the basis for the discussion.
If it is just a matter of scientists using old theories: of course, scientists and engineers use old theories all the time (eg building bridges, thermodynamics instead of stat mechanics) as good enough in the required circumstances. I think being able to use the theory has to say something about understanding it.
But that’s pretty obvious and so not likely Walto’s issue.
I was trying to make a limited point in the context of an initial skim of the excerpt you posted.
I understood a key point you were discussing to be whether metaphysical realism required one to believe that there was one true and complete description of the world.
It struck me that at least some of that description would require mathematics and that might affect the nature of the concepts you could describe. So my initial thought is that philosophy of mathematics must enter into the discussion.
But, as you say, you still have to go from the mathematics to predictions about the world, which may mean you can ignore the mathematics.
I thought that eliminativism was referring to scientific theories, not everyday language.
It seems to be that a bet with 500 years time frame might be a worth taking.
After all, if we are both alive in 500 years, then it is quite likely because we have both been uploaded to AI’s, and it is a reasonable bet that eliminativism in the scientific sense would need to be true for such AI’s to be created!
It’s not a rule. It’s a case by case decision. In this instance, I thought it important.
Walto hit the key point with his remark about empirical predicates. Mathematics does not refer to anything in the real world.
I’ll settle for “common” and not worry too much about “common sense” in a philosophy discussion.
It seems we now have at least four conceptions of “reduction” in play:
1. Reduction as a historical process within one science, like relativity replacing Newtonian mechanics or modern thermodynamics replacing phlogiston theories. Two subcases:
a. The replacement is a generalization, as with relativity and Newtonian mechanics.
b. The replacement is an elimination, as with phlogiston theory.
(A further subargument seems to have developed about whether one can truly understand “terms” of the older theory in each case unless one believes that older theory.)
2. Reduction between two sciences as in psychology to neuroscience.
3. Reduction between the “ordinary objects”, every day view of the world and the scientific view (independent of any particular science, I think).
4. Epistemic reduction versus ontological reduction. I may be the only one who thinks this distinction is important.
I think it would help the discussion to separate these cases or at least explain why you think they don’t need separation.
This is not at all obvious. Yes, science uses mathematics in its descriptions of the world. But I see that as mostly a matter of convenience rather than as a matter of necessity.
When I wrote that off-the-cuff reply to Walt, I think that I was subconsciously motivated by the following type of rough argument
1. A complete description of the world is a scientific description (or has a large component that is a scientific description).
2. Science is in principle reducible to physics.
3. Physics requires mathematics.
4. Mathematics is “unreasonably effective” when used in physics, which is saying that somehow the world is describable by mathematical concepts.
5. The (parts of the) any two separate complete description of the world (eg by us and some alien species) in mathematical physics will hence involve the same (or at least mathematically equivalent) concepts.
I realize all of these statements are quite questionable, although I would have thought that #3, the need for mathematics in physics, would have been among the least questionable premises!
Intrigued by references to “Ladyman and Ross” (I see it is avilable as a PDF here), as I often do when seeing book recommendations, I check out reviews and critiques.
I see Cian Dorr has written:
Look at any well-written paper in analytic philosophy and you will see arguments aplenty; if the author has not done your work for you by making a list of numbered premises, he or she has probably done enough that you could make such a list without having to exercise too much creativity. The arguments may very well be valid: you will be convinced that the conclusions are true if the premises are. So far, so good, you think. Now, what about the premises — the claims that are not the conclusions of any argument? Where did they come from? (The premise factory?) You will look again to see what your author has to say in favour of them. Sometimes you will find an appeal to some expert authority. But pretty often — perhaps especially often in metaphysics — you will find your author saying something to the effect that the premise is intuitive, or “supported by intuition”, or that its negation is “counter-intuitive”.
And concludes:
I find that conclusion depressing, if true.
I wonder how closely Ed Feser fits the description of analytic philosopher?
I missed this when Neil started a new thread. Thanks for the pdf link.
The associated blog looks interesting. Is it worth reading regularly? And does he make a habit of posting books?
I think it is true, and I also find it depressing. Much the same can be said about analytic epistemology — heavy on the “intuitions,” light on connections to cognitive science.
I suspect — and this is my own, very idiosyncratic impression, based on my experiences of analytic philosophy in the past thirteen years (I started grad school in 1997) — that analytic philosophy started off by taking science very seriously — in the era of logical positivism. When the core ‘dogmas’ of logical positivism were rejected in the 1950s (for very good reasons!), and the consequences of that became crystal-clear to all concerned, philosophy faced a dilemma: either cease to be philosophy altogether (with Rorty and Philosophy and the Mirror of Nature, 1979) or return to pre-critical, pre-pragmatic, pre-scientific metaphysical realism (with Kripke and David Lewis). Since the former would have meant an end to philosophy as a discipline, the discipline as a whole chose the latter.
My grad training was at one of the last bastions of old-school analytic naturalism, where the Churchlands were keeping alive the spirit of Quine and Sellars, but where there are also small pockets of anti-naturalism grounded in German Idealism and phenomenology. Since then I’ve tried to forge my way in analytic pragmatism, but with more attention to natural science (esp. biology, which was my undergrad major) than most other analytic pragmatists.
I would say that Feser is very close to analytic metaphysicians, in that he thinks he can get his science on the cheap — “yeah, sure, I accept all of empirical science, but there’s this other thing, metaphysics, that’s foundational to science” — without actually taking the time to learn any science or think seriously about its implications for philosophical reflections on how cognition happens. (Doing that is “scientism,” after all, and that’s Really Bad.)
Feser has a contribution in New Scholasticism Meets Analytic Philosophy (reviewed here). That suggests to me that he’s practicing a very similar style of philosophizing — heavy on the intuitions, light on the data, and pretending that the former count as the latter.
By the way, I do know of a few Thomists who disagree with Feser about whether the accommodation between science and metaphysics can be purchased so easily. Even though Feser has a loud blog-presence, he’s not the only voice in the chorus.
Don’t be depressed. Be inspired (to do something about it).
The trouble with that, is that it makes science prior to philosophy. And if science is prior to philosophy, then the scientists are better equipped to do that kind of philosophy that derives from their science.
To me logical positivism seemed to be a kind of scientism, subject to many of the criticisms that we see of scientism. If L&R want science to be prior to philosophy, they are probably going to run into similar problems.
If philosophy is to be important, then it has to be prior to science. And it cannot be successful if it tried to do this by making up premises (overusing intuition), for that makes it look too much like religion. So it needs to find another way of being prior to science.
To illustrate, mathematics is prior to science and still important. I’m not suggesting that philosophy should become mathematics. I’m just pointing to that as an example. Philosophy needs to find a way, other than mathematics, that it can be prior to science but yet be highly relevant.
Perhaps the future rôle for philosophy is in the study of “oughts”, leaving the study of “isses” to science but using the knowledge provided by scientific endeavour for more realistic premises. Science has singularly failed at giving clear answers to “what we ought to do”.
Answer: No.
Quote from the blog:
“I have thought up a method for FTL communication. Not sure if it works, probably doesn’t and i’m wrong about some fysical [sic] detail, not being a fysicist [sic] and all”.
Given that, it’s not clear what use he could make of the copy pdf of L&R’s book that he provides.
It’s not clear what you find depressing. If it is the state of metaphysics given L&R’s criticism, then I think the following reported conversation (in an iai broadcast) between a philosopher and a famous neuroscientist is relevant:
Philosopher: I think 95% of neuroscience papers are junk.
Famous Neuroscientist (after some thought): No, I’d say it is more like 98%.
Here’s one example that coincidentally appeared in my Feedly this AM (although it is not neuroscience):
False Positives in Psychology
The fact that most published stuff is considered wrong or of no enduring value by experts in the field is not restricted to philosophy.
I’m not sure if that means you should not be depressed or if it means that everyone should be depressed.
I can’t imagine how any largely rational person might think Feser’s blog would be valuable. I notice too, that some of the links he has to other philosophers (most of them similarly crazy) don’t work.
ETA: The L & R book looks interesting–especially the stuff on structural realism. I have some correspondence with Psillos on that subject (especially with regard to Schlick) here: http://www.scribd.com/doc/203211505/Psillos-Horn-Correspondence-on-Schlick
OTOH, I don’t think I agree with the main thesis of their book–at least as explained in their preface. (Prolly should read at least some of it before saying that, though.)
My quote was not from Feser’s blog, in case there is any confusion there — it is from the blog at the same site as the pdf of the book.
Ah, I did misunderstand that. I had a quick look at Feser’s blog, and threw up in my mouth.
Neil Rickert,
I think there are some ways in which philosophy can be a priori (if that’s what you meant by “priori”?) while respecting science, in part because there are a priori features of scientific theories themselves, and philosophy can explicate those features. And the elucidation of the conceptual frameworks that we use within the manifest image is also important, though more as a kind of anthropology than as directly characterizing objective reality.
Naturalizing metaphysics, epistemology, and ethics does not put philosophers out of a job — on the contrary! But I don’t think that metaphysics, epistemology, and ethics that wholly ignore the natural and social sciences is an intellectually responsible activity.
Everyone should be depressed, but some should be more depressed than others.
Yes, Feser is a real piece of work. He first came to my attention a few years back when he was leading a campaign against the APA’s statement of non-discrimination. Specifically he objected to the inclusion of “sexual orientation” in that statement. You can pretty much guess why.
I wish I were making this up, but Feser really does think that the only reason why anyone would believe that homosexuality and abortion are morally permissible is because they have adopted an utterly and obviously false theory of the nature of reality. The only correct theory of the nature of reality, he thinks, is Scholasticism, and that shows why abortion and homosexuality are intrinsically and categorically immoral.
I think I said “prior” rather than “priori”, meaning that there are things philosophy can study that science depends on.
Yes, that would be a good example.
His political views do seem pretty reactionary, judging by his endorsing the SCOTUS decision on Hobby Lobby and his objections to Obamacare.
Feser thinks that a progressive makes the error of selecting one’s metaphysics on the basis of one’s political commitments. To the extent that that’s true, he fails to see that he (and all the theistic reactionaries at Uncommon Descent) all do the exact same thing.
Kantian Naturalist,
Science stops being science when it attempts to answer questions like “Is that all there is?” or “Why should we do stuff that makes people feel good?” or “Can an artwork be good even if nobody likes it?” or “Does that kind of reduction or explanation actually eliminate the reduced items?” or “Why does consent matter if a course of action will produce more utility?” Etc. Etc.
We can say that such questions are meaningless or are just mis-stated claims about people’s feelings (which means that everybody is wrong about almost everything they think say, do, and argue about every day of their lives), or we can put them under the provenance of philosophy so that philosophers can muse (endlessly) about them, making precious little progress in spite of centuries of blabbing and dickering. But they make for crap science because they simply are not empirical questions.
These views indicate why I don’t think I’d like the L & R book too much. Metaphysics isn’t eliminable without taking out a ton of important stuff with the bathwater.
Having looked a bit more at L&R and also having started reading the follow-up volume Scientific Metaphysics, I think that L&R would not say that such questions are meaningless, but that objective reality is silent about them. On their view, all the metaphysics that there is about objective reality is the metaphysics based on science, because only scientific practices reliably filter our false beliefs from our true beliefs, regardless of how intuitive the former and counterintuitive the latter.
I’m still on the fence about this — I worry that throwing the manifest image under the bus is not a defensible attitude, all things considered.
I’m a masochistic manifest mofo, myself. 🙂