Philosophy of science

“The philosophy of science is as useful to scientists as ornithology is to birds.”

I believe the preceding quotation, often attributed to the great physicist Richard Feynman, captures a common sentiment among scientists. Philosophy of science is thought to be useless to practicing scientists. Why bother with philosophy when you could just do science? What exactly is philosophy of science, and why is it considered, by philosophers, to be a legitimate area of inquiry?

Philosophy of science, proper, deals with normative, epistemological, and metaphysical questions that concern all of the sciences. Here are five such examples:

o   What are the criteria for theory confirmation?

o   What makes a theory scientific?

o   What rational basis is there for making inductive inferences (reasoning from past cases to future ones)?

o   What kinds of metaphysical commitments do our best scientific theories have (e.g. should we believe the Higgs Boson exists?)?

o   Does science tell us about reality or only about appearances from the perspective of human beings? (Would space aliens develop scientific theories that are at odds with our own?)

Over the past few years, there has been a vocal group of scientists (e.g. Neil de Grasse Tyson, Peter Atkins, Lawrence Krauss) who think that reflecting upon on these kinds of questions are (1) irrelevant to their line of work, (2) trivially easy to answer, or pointless to ask, either because (3) there is no answer or that we could never know the correct one. A more general critique to philosophy of science as a whole is that it is alleged to be uninformed by contemporary science (4). In this post, I will address each of these contentions in turn.

(1)               Philosophy of science is irrelevant to practicing scientists

I think it’s true that scientists can go about their work without ever thinking about the philosophical assumptions that they make, but I think it's much less clear whether they should. But I do not know of any philosophers who think scientists need to do philosophy of science to become competent experimenters or to develop novel hypotheses to test. Hence, I think scientists who say that they don’t need to do philosophy of science are attacking a straw man. That isn’t to say that philosophy of science is of no use to scientists. On the contrary, I think it would be a good thing for scientists to be aware of the philosophical import of their views. Being aware of the assumptions one makes is the first step in determining whether or not those assumptions are warranted. 

The physicist Lawrence Krauss advances a pretty weak argument for why scientists philosophy is of little to no use for scientists. His reasoning? Because scientists, who do their jobs just fine, don’t read philosophy journals. The problem with Krauss’s argument is that it doesn’t tell you anything about whether the scientists who do keep up with philosophy of science are better or worse off. It could be philosophy of science is of use to scientists, but that most scientists have been missing out.

(2)             Philosophical questions about science are trivially easy to answer

The first thing to say about this is that these questions are not easy and those who think that they are probably haven’t thought much about them. If it’s so obvious, it should be easy to develop arguments and reasons for why theory A is preferable to theories B, C, and D. Without doing the work—in showing why their own assumptions are rationally defensible—it prevents them from entertaining alternative views that may be more plausible, given a different set of starting assumptions. Moreover, some of their scientist peers probably have the same attitude about their own views (i.e. they’re just obvious), preventing a proper debate about the issues to be had. Philosophers are in the business of revealing hidden assumptions in order to subject them to careful logical and conceptual analysis.

(3)              Philosophical questions about science are meaningless or impossible to answer

How do we answer the questions raised by philosophers of science? What all of these questions have in common is that you won’t solve them by doing science. Perhaps that is why scientists don’t really like to raise these questions. Their area of expertise is in addressing questions that could, in principle, be answered through experimentation and observation. But those are not the only ways to settle a debate. Scientists may not know how to answer normative, metaphysical, or epistemological questions, but it doesn’t follow that no one does. Philosophers have been making progress on these questions by clarifying the concepts and theories that have been advanced in trying to answer these questions. For example, in determining whether a theory is scientific or not, philosophers have put forward a number of views, each one building upon the successes and improving upon the faults of their predecessors. Scientists like to defer to the philosopher Karl Popper, in that science, unlike pseudoscience, is in the business of falsifying hypotheses, not confirming them. To just accept what Popper says at face value is to assume that no progress has been made on the question since the 1960’s. However, there are more recent views that are, in my view, much more plausible and nuanced (c.f. Lakatos 1978, Thagard 1978, Pigliucci and Boudry 2013).

(4)             Philosophers of science have not kept up with the science

Stephen Hawking has alleged that philosophy of science is a dead enterprise because philosophers have not kept up with modern science (Hawking and Mlodinow 2010). It is interesting to note that, in his book, Hawking provides no evidence to support his claim.  

Contrary to Hawking’s claims about philosophers, many have extensively studied the science relevant to the questions they work on. Taking a look at some of the prominent philosophers working on the foundations of physics, David Albert and David Wallace have PhD’s in physics, while others like Tim Maudlin have BA’s in physics and have displayed extensive knowledge of the field in technical papers and books. So while scientists concentrate more of their interests on conducting experiments and carrying out research, it is not as if philosophers of science are unaware of their projects. Philosophers keep themselves up to date with the scientific literature, all while continuing their work on solving the philosophical questions that have been lingering in the background all along.

Ever since science split off into all of its specialized sub-branches (e.g. biology, physics, psychology), new philosophical questions, specific to the subject matter of each particular field, began to surface. In turn, philosophers have become increasingly specialized to deal with such questions. To give three quick examples, philosophers of physics grapple with how to interpret experimental findings (e.g. double slit experiment) relevant to understanding quantum mechanics, philosophers of biology sort through controversies over how to understand the level (e.g. individual vs. group) at which natural selection takes place, and philosophers of psychology work on clarifying and analyzing the concepts used to explain cognition and behavior (e.g. ‘mental representation’). While scientists also grapple with many of these questions from time-to-time, the philosopher works full-time on these issues.

Conclusion

Just like there has been a division of labor among scientists with respect to a range of subject matter, there has been a division of intellectual labor among the kinds of questions that are raised in each field. Before the 20^th century, most scientists, or natural philosophers, reflected upon both kinds of questions. Even within the 20th century, some of the greatest scientists (e.g. Einstein) carefully studied and reflected upon the philosophical issues related to their areas of study. This, rather recent, division of intellectual labor is not necessarily a bad thing, but it has led to a kind of rivalry between scientists and philosophers that is, in my view, mistaken. Science and philosophy go hand in hand in explaining the world and our place in it. Where science leaves off, philosophy takes over. This says nothing about whether one is more important or explanatorily privileged than the other. It is just to say that these are different endeavors, each sharing the same aim: knowledge and understanding.

Works cited

Hawking, S., & Mlodinow, L. (2010). The Grand Design: new answers to the ultimate question of life.

Lakatos, I. (1978). Science and pseudoscience. Philosophical papers, 1, 1-7.

Pigliucci, M., & Boudry, M. (Eds.). (2013). Philosophy of pseudoscience: reconsidering the demarcation problem. University of Chicago Press.

Thagard, P. R. (1978, January). Why astrology is a pseudoscience. In PSA: Proceedings of the Biennial Meeting of the Philosophy of Science Association (Vol. 1978, No. 1, pp. 223-234). Philosophy of Science

What Fodor got wrong: Distilling theoretical issues in philosophy of biology into plain talk

In 2011, the philosopher Jerry Fodor riled up philosophers and scientists alike with the release of a book, co-written with a cognitive psychologist, criticizing Darwin and his theory of natural selection (Fodor and Piattelli-Palmarini 2011). Fodor is a card-carrying atheist, and accepts most of what modern evolutionary biology teaches: the glaring exception being the theory of natural selection. If I read Fodor correctly, this is roughly what he argues in his book, “What Darwin got wrong”: The theory of natural selection is—by itself—empty or unexplanatory. This is because, for any two linked traits (e.g. pumping blood and thump thump noises), the theory does not tell you which was selected for. In the case of the heart, evolutionary biologists know perfectly damn well that it was selected for pumping blood (Fodor thinks even his grandma could have known that!). However, the evolutionary biologists are drawing from lots of empirical evidence and fields of study (e.g. functional biology) that fall outside the reach of the theory of natural selection. Since Darwin’s theory does not provide an answer to the selection question (i.e. pump or thump?), natural selection—understood as a scientific theory—does not account for the very phenomenon it tries to explain (i.e. the selection of traits). Fodor thinks that trying to provide a selection mechanism to save the theory is a fool’s errand, since there’s likely no such mechanism. Therefore, Fodor concludes that we should give up on the theory all together. There are two burning questions: 1) how does Fodor reach this conclusion? & 2) what is the significance of Fodor’s claim (if he’s right)?

§1: The theory of natural selection

Before I get to Fodor, I will quickly summarize some basic concepts with the aim of further illuminating Darwin’s theory. Those who are familiar with the theory can skip or skim most of this section. Darwin’s theory is predicated on two theses: 1) common descent and 2) the transmutation of species. Roughly, these state that 1) all life on earth shared a common ancestor (in the distant past) and 2) lifeforms have changed—or evolved—over time. The second thesis can be equated with evolution, understood as a natural phenomenon. The theory of natural selection aims to explain how evolution is possible. Amongst laypeople, it is commonly asked whether or not evolution is true, or whether the theory of evolution has been proven. However, both of these questions are ill-posed.

I think that the question of whether evolution is true or not commits, what philosophers call, a category mistake. If one understands evolution to be the mere phenomenon of creatures changing or evolving over time, then such a question would be akin to “Is gravity, or precipitation true?” Gravity is a natural phenomenon, one we are all familiar with. In addition, there are theories of gravity (e.g. Einstein’s general relativity theory) that explain how the phenomenon works. It would be strange to say that gravity—the phenomenon itself—is true. Likewise, it is improper to say that evolution is true. Phenomena cannot be true or false; they just occur or not and can observed or detected. The main disconnect between gravity and evolution is that virtually all of us directly observe the former but not the latter. Evolution—like other natural phenomena (e.g. formation of islands, glacial movements)—is typically too slow or subtle for most of us to notice in the real world. However, scientists observe evolution occurring (both directly and indirectly) on a regular basis (e.g. the evolution of the flu virus).

Taking up the second question—of provability—is again making a category mistake. There are no instances of proven scientific theories because they aren’t the sorts of things that can be proved. It would be like asking, “What is the marital status of a triangle?” Proofs are perhaps the standard in disciplines like mathematics, but not in science. Scientific theories perform a number of functions. Explanation is perhaps the most important role. Of explanations, we can ask whether or not they are adequate, plausible, well supported by evidence and so on. It is often remarked that “evolution is just a theory”. But this ignores the wide spectrum of theories, some of which aren’t supported at all by evidence (e.g. Dogs are cleverly disguised Martians), and our scientific theories supported by an overwhelming amount of evidence (e.g. relativity theory, quantum mechanics, plate tectonics). It would be like saying that “So and so is just a basketball player”, discounting all of the diversity of skill level that could fall under such a description. No reasonable person would lump their next door neighbor, who plays basketball for fun by himself on weekends, to an NBA hall of famer. Likewise, no reasonable person should put well-supported scientific theories into the same boat with wildly speculative and empirically unsupported theories.

The theory of natural selection is typically broken down into three key ideas that aim to explain how species evolve over time: variation, inheritance and selection. The first two ideas are fairly straightforward. First, in order for creatures to change over time, there needs to be some way for differences (e.g. anatomical) between creatures to accrue. An organism that gave birth to a carbon copy of itself would result in no species change. Second, mere trait changes would not get very far if there wasn’t a way for the offspring to inherit those changes. We now know that both variation and inheritance can be explained at the level of the gene, and that genes can be altered through random mutations which result from a number of causes (e.g. UV rays, natural sources of radiation on Earth, genetic copying mishaps). Genetic variation combined with a mechanism of inheritance (i.e. genes) provides a way for creatures to change over time.

The final piece of the puzzle is to account for why certain creatures survive and pass along their genes, while others die out. Darwin made a distinction between two kinds of solutions. The creatures can be selected (to survive) either artificially or naturally. Artificial selection is perhaps best exemplified by the evolution of the domestic dog. All domestic dogs—including the Chihuahua and Great Dane—shared a common wolf ancestor roughly 30,000 years ago (Skoglund 2015). Modern domestic dogs vary so significantly because of humans, whom artificially selected the nicest and prettiest wolves. Natural selection involves no selecting agents such as humans. Natural selection occurs by a number of means, though natural disasters (e.g. forest fires, tsunamis), disease, and predators are the big three. Typically, the creatures that survive long enough to reproduce are the ones that are best adapted to their environment. Sadly, since most environments change over time (e.g. due to climate change, asteroid impacts, etc.), most creatures will die out. Roughly 98% of all creatures that have ever existed have gone extinct. The survivors—including us humans—did not just get lucky; all extant life on the planet survived for a reason, namely, we all more-or-less inherited the traits necessary to survive in our respective environments.

§2: How Fodor gets there

The previous section laid out Darwin’s theory of natural selection in a nutshell. To make the issues more engaging, one might now imagine Fodor’s reaction after sitting in on such an overview of the theory.

Fodor: Artificial selection makes a lot of sense; it’s the natural selection part that gets me. When a dog breeder selects the dogs he wishes to breed, there is both an obvious selection mechanism at work (i.e the dog breeder) and an obvious way to determine which trait is being selected for; you just ask the guy! However, when it comes to natural selection, there is the question, for any pair of linked traits (e.g. pumping blood and thump thump noises), which gets selected for? There is first the issue of there being no human breeder (e.g. God) to interrogate, as well as the issue of what a natural selection mechanism—that picks out one trait and not the other—would even look like. In every instance where you have linked traits, it seems as if ‘nature’ has to select both. It is obvious to everyone that there is a fact of the matter about which trait is being selected for (the heart!), but there doesn’t seem to be a way for Darwin’s theory to tell us which, absent some kind of natural selection mechanism or a law of adaptation.

Upon learning about evolutionary theory, really no one (other than Fodor) raises such problems. Why is the theory so commonsensical to scientists and laypeople alike, but so incoherent to Fodor? I think there is a straightforward answer, and one that does not appeal to name-calling. Fodor has some pretty sophisticated views about the nature of scientific theories, laws and explanations. The reason why scientists don’t really engage with his claims is because they probably don’t have any well-developed views on the nature of explanation and science, both of which typically fall under the heading of ‘philosophy of science.’ The reason why most philosophers of science have not “seen the light” is because Fodor’s views on these matters are not widely held today. For example, Fodor endorses Hempel’s deductive-nomological (DN) model of explanation (Smith 2008). Without getting into the details, this view on explanation does not work very well for Darwin’s theory, an observation made well before the publication of Fodor’s book (Hempel 1965). For those uninterested in evaluating the merits of the competing philosophical views on explanation, here is an (overly simplistic) argument you might find helpful.

What Fodor says about natural selection essentially hinges on the truth of Hempel’s view on explanation. If Hempel’s view is wrong, then Fodor is wrong. According to most contemporary philosophers of science, it is doubtful that Hempel’s view works (see Strevens 2006 for an overview). Therefore, Fodor is (probably) wrong about Darwin’s theory. 

§3. What if Fodor’s right?

Given that Fodor concedes that evolutionary biologists have the tools answer the selection question, the science can proceed as it has for some time. However, I think that the scientists would have to concede that they aren’t answering the selection question via the theory of natural selection. Additionally, they would have to admit that the theory of natural selection adds nothing to their scientific understanding of selection effects. What would then become of the theory? One might think that the theory might still be perfectly useful as a metaphor for introductory purposes. Perhaps eventually, an even better metaphor might replace Darwin’s own (Fodor and Piattelli-Palmarini 2011, ch. 9). However, one might still be unimpressed with Fodor’s argument. So what if we don’t derive truths about selection effects from Darwin’s theory? Darwin laid the groundwork for fields that now do manage to get at those truths. If all that we have to change is how we should talk about Darwin’s theory, no one is really going to care. But I believe Fodor’s claim (if right) would actually be somewhat significant. You wouldn’t be able to call Darwin’s theory a theory any longer. Would this cause celebration amongst creationist circles, or some embarrassment amongst evolutionary biologists? Possibly so. Would it change anything about how the actual science proceeds? Probably not.

Works cited

Darwin, C. (1897). The origin of species by means of natural selection, or, The preservation of favored races in the struggle for life. Vol. 1. International Science Library.

Fodor, J., & Piattelli-Palmarini, M. (2011). What Darwin got wrong. Profile books. Chicago         

Godfrey-smith, Peter (2008). Explanation in evolutionary biology: Comments on Fodor. Mind and Language 23 (1):32–41.

Hempel, Carl (1965). Aspects of Scientific Explanation and Other Essays in the Philosophy of Science. The Free Press.

Neander, Karen (1991). Functions as selected effects: The conceptual analyst's defense. Philosophy of Science 58 (2):168-184.

Skoglund, P. (2015). "Ancient wolf genome reveals an early divergence of domestic dog ancestors and admixture into high-latitude breeds". Current Biology 25 (11): 1515–9.

Strevens, M. (2006). Scientific explanation. Encyclopedia of Philosophy, second edition. DM Borchert (ed.). Detroit: Macmillan Reference USA.