One of Coyne’s big pushes in “Faith vs Fact” is the idea that science, broadly construed, is the only “Way of Knowing”. The problem, however, is that in order to justify that he needs to, well, define what it means to know in such a way that that can work. As with most of the terms in his book, Coyne eschews the actual fields that study it and work with it, and instead runs to the dictionary to yank his definition out of that, tweaking it with some supplemental material to allow him to make his point. The failings of that approach are on full display when he talks about “Ways of Knowing”, because he essentially defines knowledge in such a way that science is the only thing that could produce knowledge, because he limits the domain of “true” to the “empirical” world, and denies that, in general, to mathematical truths and philosophical truths. Unfortunately, Coyne needs at least the latter to be true in at least a meaningful sense in order for his points to work out, because if I can say that his definitions of knowledge and truth are not, well, true and therefore ignore them, and if Coyne cannot say that he “knows” that those definitions are correct, then anyone who wants to deny that science is the only way of knowing can, in fact, simply dismiss his claims and insist that, by their definition, faith still counts as a way of knowing. The fact that his definitions are, in fact, fairly self-serving only heightens the temptation to simply dismiss what he’s saying as an attempt to win the debate by definition, not by evidence or argument.
And the sad thing is that if he was a little more open to philosophy, he could have avoided all of these issues, because the field of philosophy that studies knowledge — epistemology — has already provided a pretty decent rough-and-ready definition of knowledge that he could have used: knowledge is justified true belief. To remind you of the details of that:
S knows that p iff:
S believes that p.
S’s belief that p is justified.
p is true.
Now, there are potential issues with all of these — as philosophy is well-known for generating problems for any theory that it comes across [grin] — but for the most part this works. It seems a contradiction to say that you know something to be true but don’t believe it to be true. It also seems a contradiction to say that you know something to be true that isn’t actually true, no matter how strong the purported justification is. And the difference between simply believing something to be true and knowing it to be true seems to be tied to how strong a justification you have for thinking that it is true, and is something that Coyne himself it seems definitely wants to accept (because that’s how he tries to eliminate “faith” as a way of knowing). So the only thing to clarify here is what it means for something to be true, as Coyne definitely tries to limit it to statements of “fact” about the empirical world, while philosophers and mathematicians, at least, will want to extend it to include their fields. In general, though, the way I define truth is this: the statement accurately describes what is the case in a particular domain. So empirical truths describe the way things are in the empirical world. Mathematical truths describe what implications and statements are correct given a certain mathematical system. And most importantly, philosophical truths describe what is true about a set of concepts or philosophical positions, such as, well, what the definition of “knowledge” must be conceptually. This, then, allows Coyne — even if he rejects the standard definition of knowledge — to be able to point to a definition of knowledge that is “true”, and therefore can be justified to be true, and therefore is one that people cannot simply dismiss … which his view actually doesn’t, as “truthiness” is not a property of the empirical world.
In talking about science, it seems to me that Coyne does at least roughly align with Larry Moran’s definition of science, in that he seems to regard science as being critically involving rational thought, skepticism and empirical evidence. I’ve talked about that before, and pointed out how that leads me to say that there are three different ways of knowing: science, everyday reasoning, and philosophy. Everyday reasoning, it seems to me, rejects skepticism, but also produces the vast majority of the knowledge that we rely on, well, every day. Philosophy rejects empiricism, although unlike in the linked article I now actually believe that philosophy is, in fact, even more skeptical than science is.
Which is a good place to talk about skepticism, since in order to demonstrate that everyday reasoning is not skeptical we need to know what it means. Many assert that skepticism is critically about tailoring your beliefs to the evidence, but this is, I think, problematic because it essentially defines the skeptical approach as reasonable and then argues that it is the only reasonable one you could have. People relying on faith, for example, could easily insist that they do tailor the confidence of their beliefs to the evidence, but that the skeptic simply denies that the evidence exists, which leads to a massive morass that we probably didn’t want to jump into simply by defining a term. So I prefer this, and relate it to belief itself, directly: a skeptical approach says “Do not believe without good reason”, while a non-skeptical approach says “Believe unless you have good reason not to”. Now, neither approach means that you accept things that are “unevidenced”, or that have no evidence in their favour. And both, in some way, insist in appropriate evidence for the beliefs, especially if they want to make a knowledge claim out of it. So opposing a non-skeptical approach that way won’t work.
Let me use everyday reasoning, which I think is a non-skeptical approach, to show how this works in practice. To form a belief, in everyday reasoning we first have some kind of initial “evidence” for that proposition. We see something. Someone tells us something. We conclude something from what we already “know”. We get a belief from the culture. Whatever. At that point, we check that against our Web of Belief, against the things that we already believe or know to be true, and if there’s no contradiction we accept it. Contrast this with science, which wouldn’t stop there, but would instead definitely try to directly test it before accepting it, and would do so in ways that would try to prove it wrong. Everyday reasoning doesn’t try to prove things wrong, but instead adopts them, acts on them, and lets reality disprove it if it can.
When I say that everyday reasoning is non-skeptical, then, this is what I mean: everyday reasoning will not doubt — and therefore will not explicitly test — a proposition unless it has reason to. It does not reject skepticism outright, but instead instead rejects skepticism as a primary criterion for truth and, most explicitly, as a starting point for forming true beliefs.
The same applies to philosophy’s rejection of empiricism. It’s not that philosophy says that empirical data can’t produce knowledge, because philosophy concedes that it can and often does. And it’s not even that philosophy says a priori that empirical data can’t be used to settle traditionally “philosophical” problems, because the history of naturalized philosophy and empirically-minded moral philosophy belies that as well. No, philosophy rejects the idea that empirical data is required in order to know anything — even about the empirical world — because it can think of at least possible propositions that could be true and yet where empirical data is useless for determining its truth value, and of cases where even purely rational proofs could indeed prove truths about things that are typically considered empirical. This is why I say that philosophy is even more skeptical than science is, because while science accepts certain preconceptions built around the idea that there is a certain method that one ought to use when examining the empirical world, philosophy rejects even that: in a philosophical argument, it is quite reasonable to demand that someone demonstrate why they think their approach can actually get them to the right answer. Imagine someone in a scientific context demanding that the scientist show that the use of scientific method is actually reasonable in this case! But in philosophy, you can demand that the person justify their use of rational conceptual analysis here instead of science, and vice versa.
So, to me, there are three different ways of knowing: everyday reasoning, science, and philosophy. (Whether mathematics is another way of knowing or just a special case of philosophy is not a discussion that I want to go into today). But it can be asked, especially given science’s record, what we need the other two for? Science gives us truths and is very good at it, so why use the other two at all?
Everyday reasoning, because it is less skeptical than science, is wrong more of the time (although it is also self-correcting). But its big benefit — and why it is non-skeptical — is that it is fast. For any proposition that both everyday reasoning and science come to the same conclusion on, science will take a lot longer to get there because it will have to run far more tests to come to that conclusion. This means that if we need to act on that proposition in the world, we will have to wait far longer for science to tell us which way we should act towards it: as if it is true, or if it is false. But we generate a massive number of beliefs, and act on a massive number of beliefs every day. We just don’t have the time or resources to test every belief skeptically, nor do we have the time to wait for science to tell us what we should or shouldn’t do. If when I have a cold I eat chicken soup and feel better, I certainly feel justified in continuing to do that — and so making a claim that I know that chicken soup makes me feel better — without waiting for science to demonstrate that it really can … whether I’m actually right about that or not. If science, on testing it, finds reason to think that that is not true, then I have reason to doubt and adjust accordingly. But I don’t have to wait until it does to decide whether that belief is justified or not, and so to act on it or not in the world … which, since I don’t have the time to do that, is a very good thing.
As for philosophy, the biggest benefit of it are what we might call “meta” analyses: examining whether what a method entails and whether or not it works to achieve its goal. This is why we have so many “Philosophy of …” fields, as philosophy looks at the field from the outside, without using its own methods to study itself (which, even if it can be done, must be considered suspect from the start). In this way, it also challenges presumptions, which can reveal flaws and also that a field is rejecting a finding only because of presumptions and preconceptions that it need not make. And, currently, it’s also the go-to field for any propositions that might not be empirical, as science can’t address it. So, the normative claims of morality or even of epistemology. Again, philosophy does not presume these to be non-empirical or beyond the reach of science, but instead argues for it … and challenges that conclusion itself.
In short, we definitely seem to have three methodologies that are fundamentally different and yet all produce justifications for propositions that we definitely want to consider “true” in a strong sense. They are not all science, nor can they all be science without losing what makes science good at what it does. Thus, there are definitely more ways of knowing than just science. Note that I personally don’t think that faith is a way of knowing because I don’t see how it provides any justification for a proposition; it seems to be about coming to believe a proposition with a greater confidence than the evidence supports, not about providing a justification of that move. So we agree on that, at least that far. But to deny that there are any other ways of knowing is, to my mind, going way too far.