As is often the case when I see an either-or question like the one in the title of this post, my knee-jerk reaction is to say, “I don’t know, maybe both?”. (I tend to be very suspicious of dichotomies, suspecting that many more are actually false than is commonly assumed.) In this case, after further reflection, I would definitely say both.
The folks over at The Christian Humanist have put together a really nice podcast all about epistemology. I highly recommend you head over to their site (also linked in the blogroll on the left), subscribe to their podcast feed, and give it a listen. In one part of their podcast, they discuss the relationship of epistemology to modern science, specifically by arguing that modern science actually contains elements of both a “rationalistic” epistemology, and an “empiricist” epistemology, even though it is commonly assumed to be largely described by the latter.
I have to agree, but first I need to explain what is meant by the terms “rationalism” and “empiricism” in this context. Now, I’m certainly no expert here, but from what I’ve been lead to understand, rationalism is a theory of knowledge that tends to emphasize the faculties of pure reason, although the Wikipedia article linked to above appears to define this as “idealism”. Perhaps someone who knows more about these distinctions can chime in here. Rationalism, thus defined, would be concerned in a scientific context with the building of models and theories, logical and mathematical, for which to make sense of the data of science and to make predictions about what new observations and experiments might show. On the flip side of this is empiricism, which is a theory of knowledge that emphasizes observational data, gathered by our senses, and by extension, our scientific instruments and observations and experiments utilizing those instruments.
The question in my mind is, which one of these epistemologies, if either, is primary in modern science? I myself have tended to lean toward an empirical view of things as having the final say in matters of science, and I think many (dare I say, most?) scientists would agree with me. What I mean is, that at the end of the day, all of our rationalizing–that is, our theories and models–can be overturned by new observations and experiments. On the other hand, I also think that theories in science are absolutely indispensable for at least a few reasons: 1) They help us make sense out of patterns we see in nature, which is what gives form to science and keeps it from being a mere collection of facts, 2) they make predictions about what new phenomena we might uncover once we have the technological capability of doing so, and theories are thus at least in part measured by how successful their predictions are, and 3), they guide the development of new ways of observing and experimenting to begin with. In this way, theory feeds back on experiment, which in turn tells us how well our theories are doing. So, with this way of looking at things, one can argue that both theories of knowledge are needed in the modern scientific enterprise. I might also point out that this is one of the areas where philosophy can really help us scientists think more clearly about what, in fact, we mean when we talk about gaining scientific knowledge, and how we come about that knowledge.
What do you think?
P.S., I also made a comment on the blog summary of the podcast that gave some specific contemporary examples of the interplay of theory and observation/experiment, if anyone is interested.
Pure rationalism isn’t what I would call science, even though it’s an extremely helpful tool – in fact it’s indispensable for science. What distinguishes science from other endeavors is not its rationality, although being rational is certainly a minimum requirement for any scientific idea. Rather, what characterizes science is the TESTING of ideas (rational models), and their possible revision in light of those tests. Classical “natural philosophy” is mostly about rationalism – what transformed it into science was the recognition of the necessary requirement that those ideas fit observations. Mathematics is not science – in math, absolute proof is not only possible, but it’s commonplace: state the axioms, follow the rules, and a proof resolves the issue of the correctness of a theorem once and for all time. Scientific understanding, on the other hand, is always provisional, subject to re-examination at any time, and especially when new observations become available. I subscribe to Karl Popper’s view of things … an untestable idea (one not subject to empirical tests) is outside of the domain of science, without regard to its inherent rationality.
Chuck, I’m pretty much in agreement with you here. I think I hold–provisionally at least–to a Popperian view of things as well when it comes to deciding what is and is not science. However, I also recognize that scientists throughout the ages, and even in modern times, have not been in uniform agreement on this question. After all, on this view of testability being a necessary condition for science, vast areas of theoretical physics (take String Theory and M Theory as examples) would be on the cusp of being non-scientific, by definition. However, as I alluded to above, I sincerely doubt String Theorists would agree with this! Also, historically, for example, mathematics was counted among the sciences, and it’s really only in our modern age that it has begun to be excluded from this based precisely on this idea of empirical testability (or falsifiability, as Popper put it). Of course, none of us could do science without mathematics! So, while at the end of the day I tend to come down on the side of Popper here, I’m open to the idea that there is a gray boundary of science, so to speak, within which certain ideas and theories (such as String Theory) may be compelling and worthy of being called science, despite not being strictly (or at least practically) testable or falsifiable in the Popperian sense.
I’m sure that string theorists consider themselves to be scientists, despite their ideas being wholly mathematical. I’m confident Einstein thought of himself as a physicist, although when he first produced relativity theory, there was no known way to test his ideas. Relativity became part of science when it first was tested and its creator was not a participant in the test, although he surely was an interested outsider. He was satisfied with the ‘beauty’ of his mathematical construction and was confident it would be validated if it ever was tested.
However, the world is full of those who are capable of constructing mathematical theories who are equally convinced they have had a profound insight. I’ve encountered such many times during my career. They seem to resent the very idea that they would have to test and thereby validate their ideas before I would give them any consideration. For every Einstein that turns out to be right in his abstract speculations, there are many, many equally arrogant crackpots who have indulged in creating mathematical hypotheses and who demand that someone else put their ideas to the test or even into practical use!
Purely mathematical physics is, therefore, not science. The test against observations is what allows us to separate mere speculation, no matter how rational or sophisticated, from the realm of science. String theory uses mathematics in ways consistent with physical laws insofar as we know them, but goes well beyond what is known into the unknown and (at least for the moment) unknowable. String theorists (and other theoretical physics types) often speak of things like ‘beauty’ to justify their unfounded belief in their own theories. This is a profoundly unscientific argument, in my view. Mathematics is a powerful tool in the sciences but it is NOT science, per se, any more than the language we use to express our scientific ideas is science. And I dispute your assertion that science can’t be done without mathematics.
A quick note to add to the conversation. To my knowledge, Popper’s demarcation criterion has been thoroughly questioned (if not abandoned altogether by most scientists and virtually all if not entirely all philosophers of science). If Popper’s criterion is true (that it is the falsifiability of a theory that renders that theory scientific) then a very great deal of contemporary science is rendered unscientific. For example, the entire standard model would be thrown out because the standard model posits unobservables. Since unobservables cannot be falsified, the standard model cannot be tested, so any theory in the standard model would be rendered unscientific. (This ultimately points to the very big questions being investigated in the scientific realism debate, so I’ll stop here.)
Cactus Man … “abandoned altogether by most scientists”? I think this statement is simply not an accurate one. Show me the evidence that such a statement is accurate! There are many testable aspects of the Standard Model, as demonstrated by the apparently successful effort to detect the Higgs boson!! As I indicated, future work may uncover ways of testing the predictions of mathematical physics in a credible way. What makes you think you can speak for “most scientists”?
Hi Chuck. I apologize for not responding sooner. On a whim, I returned to this site and saw your response—I did not receive an e-mail indicating that I received a response (which some comment forms initiate). So, a quick side note to the webmaster, is it Dan?: does your comment form e-mail responses? If so, perhaps this one was marked as spam or something, though I don’t see it there either. Also, does your site support a comments RSS in addition to the main “posts” feed? I could stay apprised of new comments that way, too.
Anyways, back to you Chuck. I don’t have much time, so I’ll be somewhat brief here. I can qualify “abandoned altogether by most scientists.” Perhaps that should say “abandoned altogether by most scientists who are aware of the logical problems inherent in Popper’s description of science.” In other words, I don’t have something like a poll that would indicate just how many scientists are Popperian, but many scientists who have studied the philosophy of science have acknowledged the logical problem. In fact, I can’t think of even one scientist who has written in the philosophy of science who has overcome the logical problem—so it’s perhaps that that I had in mind when I wrote what I did above. But perhaps you could be the first! Who knows?
So, to continue: surely there are testable aspects of the standard model, but, and just here is the problem: what is testable is only the observable. The Higgs boson is itself not testable, and never will be, for it will always remain unobservable. There’s just no way around that. What we justify the existence of the Higgs boson with is observable, yet entirely circumstantial evidence. However, as I indicated above, this relates to the realism debate, which is ongoing. There is, though, even a more problematic challenge to Popper, something called the Duhem–Quine thesis. In brief, it states that no theory is strictly-speaking testable (in isolation). That is, theories rely on large scientific paradigms (perhaps we could call them sets of background knowledge). The problem: if we test a theory, and it fails, we might be tempted to say that we falsified the theory. But: it could be the case that what is false is not the theory but something in the background knowledge; and we can’t tell which by testing theories.
Taking all this together, scientists who are aware of these problems do not identify themselves as Popperian. Yes, of course, they still test theories (it’s a hallmark of science, of course!), but they do not claim the Popperian demarcation criterion of testability as the cardinal feature of science.
I’ll try to check back sometime to see if you have a response (unless I learn of another way to be alerted to comment responses here).
Hi, The Cactus Man,
Sorry I haven’t replied until now. I’m currently on vacation and before this have been pretty busy during the summer, and so haven’t paid as much attention to my blog as I should have! I’m not sure why you aren’t getting notifications of responses (so I’ve sent you an email for the time being). I will look into it and get back to you as soon as I can.
In any case, thanks for dropping by and leaving your comments! I very much appreciate them, even if I haven’t replied to them yet. I will do so soon!