Philosophical Progress Can Be Bad for You

Surprising Consequences for the Rationality of Philosophical Belief

About the Author

Bryan Frances is the world’s only intellectual wisdom coach. He’s a former professor of philosophy & logic, doing research & teaching at universities in the US, UK, Europe, Asia, and the Middle East. He teaches you how to become the wisest thinker in the room—which is different from being the most knowledgeable or having the highest IQ. Contact for a free session.

---

This post is a summary of a much longer article.

1. Epistemic Upheaval and Progress

In the bad old days, baseball pitchers were usually evaluated on their number of wins, win-loss percentage, earned run average, and number of strikeouts. Hitters were usually evaluated on their batting average, number of runs batted in, and home runs. No one who has kept up with developments in judging baseball performance thinks this is a valuable way to go.

The reason for the shift is that the field of statistical evaluation of baseball performance has gone through a period of epistemic upheaval. Thanks to the work of Bill James and others, we have far better statistical tools for evaluating performance in baseball as well as other sports. As a consequence, many old arguments regarding such performance are now viewed as almost primitive. What passed for a good argument forty years ago is often judged naïve today, by experts anyway.

The phenomenon of epistemic upheaval applies to philosophy as well. Much of the work from roughly fifty years ago concerning many philosophical topics is almost primitive compared to that of today. I will indicate what ‘epistemic upheaval’ comes to, in philosophy, through four real-life examples.

• Anyone familiar with the state of the art regarding vagueness will find little worth studying from before the 1950s, especially before the 1975 volume 30 issue of Synthese on the topic (not nothing: little). There was a subsequent explosion of good work that utterly transformed the area, particularly with Tim Williamson’s 1994 Vagueness.

• The same is true of material composition, with Peter van Inwagen’s 1990 Material Beings and certain other key works published around the same time.

• The philosophical study of consciousness was transformed in the 1990s and 2000s starting with David Chalmers’ 1996 The Conscious Mind.

• More recently, such transformation happened for the epistemology of disagreement, with initial works by myself, Thomas Kelly, Richard Feldman, and David Christensen.

Over the last fifty or so years, epistemic upheaval has also occurred with topics in the philosophy of perception, the philosophy of biology, and other areas.

I have used well-known examples in order to indicate what I mean by ‘epistemic upheaval’. It’s clearly tied to philosophical progress: regardless of how one construes ‘philosophical progress’, the four examples above definitely involved overall progress by the philosophical community on philosophical issues (‘overall’ in order to allow for some backsliding in addition to forward movement). I won’t offer a definition of ‘epistemic upheaval’, since it turns out that none is needed for my arguments.

In order to get a hint of the surprising argument of this post, pretend what it would have been like to be a philosopher of language working on vagueness in the late 1960s but somehow aware that very soon there will be improvements that make many of the arguments and positions being thought of and discussed in the 60s almost primitive in a comparative sense. Perhaps an oracle informs you of these facts about the near future. Naturally, you are excited to learn these things. After all, the investigation into vagueness is about to make significant progress! This is a cause for celebration and happy anticipation.

But then you go back to your study and look at the vagueness papers you have already published or the new ones you have been working on lately. You can’t help but ask yourself: what are the odds they contain good arguments and positions, given what the oracle just told you?

You should probably guess that the odds are pretty slim, provided (i) in those works you aren’t arguing for logical truths that you can easily symbolize and then derive, (ii) you are arguing for theses more ambitious than ‘That argument of so-and-so has such-and-such flaw’, and (iii) you recognize that you are not a genius when it comes to philosophy. Sure, your essays on the topic might be publishable and good or even excellent for the present time, in the sense they are about as high quality as other essays being published now. The oracle hasn’t told you anything that should make you think your work is worse than that of most other philosophers working on the topic now. Neither has she told you anything scandalous about the work being done by others at your time. But you know, from the oracle, that the odds are very high that there are serious flaws in your work on the topic—so serious that your work will be viewed, by near-future experts, as almost primitive.

Given that insight, what do you now think about the quality of your current arguments regarding vagueness, or some other topic that is going to go through epistemic upheaval?

2. Major Argumentative Flaws in Philosophy

Just because there is going to be great progress in a field or on a topic doesn’t suggest that all or most of the current thinking in that field or topic is mistaken. Even if progress in mathematics is extreme, fifty years from now we are still going to think that 14 x 14 = 196 and, more interestingly, there are infinitely many pairs of primes that differ by at most 600. This holds in philosophy as well. Even if over the next decade there is going to be all sorts of revolutionary work done in epistemology, for instance, we can be confident that the odds are quite high that each of the following claims won’t be revealed to be seriously problematic.

• Two people can believe the same thing but one person’s belief is irrational while the other’s is rational.

• Knowledge is objective in this sense: just because someone thinks she knows P doesn’t always mean that she really does know P.

• One can have a true belief P without knowing that P.

Even so, it’s informative to skim journal articles from fifty to a hundred years ago. Don’t go for the classics alone. Instead, open a bunch of journal issues at random and skim the ordinary articles that satisfy two conditions: they treat topics you know very well and that have subsequently undergone epistemic upheaval (comparable to the cases indicated in the previous section). If your experience is like mine, it won’t be difficult to find in the vast majority of articles major argumentative flaws such as the following six, which are worded as if by an honest if brusque journal referee.

Problematic Premise. Look at the second premise, P2. The author thinks that it’s unproblematic, as she asserts it with complete confidence, no defense, and not even much comment. [It might not even be stated in the article, if the author is really unaware of its role in her argument.] The author is apparently blind to the fact that there are excellent reasons to think P2 is false--reasons that she has no idea how to counter. In fact, we know today that P2 is highly contentious, as the arguments against it are serious. Now, I’m not saying premise P2 is false. All I’m saying is that she is relying on P2 as unproblematic and yet it certainly isn’t anywhere close to having that status. Sure, she could just respond by saying that she’s assuming the truth of P2. That’s fine, but that move certainly reduces the interest in her argument, as her conclusion switches from C to ‘P2 É C’.

Ambiguous Premise. There’s a serious problem with premise P4. The author seems oblivious to the fact that it’s crucially ambiguous in a philosophically relevant way. If you interpret it one way, then I agree with the author that it’s unproblematic, or at least highly plausible. But then under that reading it offers virtually zero support for her conclusion. If you interpret it the other way, then it would certainly support her conclusion—if it were true. But of course we know that it’s highly contentious when interpreted that way--for reasons that she, like us, simply cannot defeat at all. I’m not saying that P4 really is false. But it certainly stands in need of a defense! Or the author can just assume its truth, but then we have the familiar problem of the significant loss of interest in her argument.

Unsupportive. Perhaps all the premises are true, but the argument isn’t even inductively strong, let alone deductively valid. The premises don’t do shit to support the conclusion!

Ambiguous Conclusion. The conclusion is ambiguous in a philosophically relevant manner. Interpreted one way, the supporting argument is great, but the conclusion is philosophically boring—even for the author. Interpreted the other way, the conclusion is definitely worthwhile, by just about anyone’s lights, but the argument doesn’t support it (either validly or inductively), for reasons I already went over elsewhere.

Mess. The argument is a total mess: it’s hardly clear what the conclusion or the premises even are. There are multiple ways of developing it but not in any way that remains at all faithful to the original and gives us premises that aren’t highly contentious and that offer decent support for a worthwhile conclusion.

Tweak. My criticisms wouldn’t mean much if the author could simply tweak her argument so that it was either valid or inductively strong, had premises that weren’t highly contentious, and had a worthwhile conclusion. But I see no plausible way to do that--and more importantly, the author certainly gives no indication that she knows of any such way.

The list is not exhaustive! But it captures a good portion of the flaws we actually encounter in philosophical arguments. Call these six plus closely related ones, Standard flaws.

In saying that the main arguments of those articles suffer from Standard flaws, I am not saying that the articles are bad, relative to the past or even the present. Argumentative quality doesn’t fix the worth of a work of philosophy. For instance, merely articulating an interesting thought experiment can be an enormous advance. And even a definitely bad argument can be highly valuable, if, for instance, it is both highly original and it can be repaired by someone else to become much stronger and retain a philosophically interesting conclusion.

But often enough, when we publish a paper that contains arguments, we think or at least hope the arguments are valid or at least inductively strong, contain premises that aren’t highly contentious, and have philosophically worthwhile conclusions. It’s fine to have an argument for conclusion C that relies on an assumption A that is highly contentious, as when you are not really presenting an argument for conclusion C but for the qualified conclusion ‘If A then C’. But in this case, you don’t want your (other) premises to be highly contentious.

With regard to the perusal of old, pre-upheaval articles, I’m saying that they had those flaws when they were published. For instance, it’s not as though the arguments had unambiguous premises back then but have ambiguous ones now. Decades ago almost no one was in a position to figure out that those arguments had Standard flaws, but the flaws were there anyway, waiting to be revealed.

3. The Unfortunate Consequences of Epistemic Upheaval

After learning that one or more of the Standard flaws is true of the large majority of the primary arguments in articles published before a period of epistemic upheaval, I am pretty much forced, by my own rationality and awareness, to reason inductively that one or more of those flaws is probably true of the primary arguments in the papers I’m writing now, provided my current papers concern topics that are going to undergo epistemic upheaval. After all, I’m just an average philosopher, and I’m not going around arguing for logical truths or overly modest theses of the form ‘Smith’s argument has the following weakness’.

Perhaps there is no current upheaval with respect to topic T that I am writing on today, and there isn’t any strong sign of one occurring soon. But given that this is philosophy, isn’t it highly probable that there will, in the next fifty years or so, be a period of upheaval for T (again, comparable to the cases mentioned in section 1)? If so, then it’s highly probable that that farther-future upheaval will show that it’s highly probable that my work on T is flawed in the sense of having one or more of the Standard flaws.

Whether the upheaval is going to happen in the next few years or in the next few decades or even century matters not at all: either way, one or more of the Standard flaws is very probably true of my work now. We can complain about how progress in philosophy is not nearly as impressive as it is for physics, chemistry, and biology, but what we do know is that we philosophers are simply outstanding at finding serious flaws in arguments and claims. Hence, it’s highly probable that in the future the philosophical community could, given a bit of time and effort, show that some of the Standard flaws are true of my work.

If I have appreciated this probability fact just articulated, then what should my attitude be towards my current and past work—or even future work? It may be well worth publishing, given the current state of the art, but should I believe in the soundness or other cogency of its arguments?