How Should Philosophers Write?

Some Unpopular Opinions

When I started out in philosophy, as a PhD student, I read an enormous amount of philosophy. I found most of it fascinating, partly because I hadn’t read anything like it before, since I started out in physics. But eventually, after I became a professor, I got to the point that I pretty much dreaded reading most philosophy.

There are reasons for this change that aren’t relevant here, since they are idiosyncratic. But one reason is relevant and familiar to most people who have read philosophy for at least a few years: it’s just not fun to read philosophy. It’s work, all the way through. When I run into philosophy professors who like reading philosophy, I think to myself “Jesus Christ, we have different tastes”.

A recent post by philosopher Steven Hales, “Why Must Philosophy Writing Be So Bad?” is a rant about the quality of philosophy writing. Now, it’s a rant, by his explicit admission, so it’s probably meant to be a bit exaggerated, a bit over the top. But it expresses well what many readers think.

Although I agree with some of his criticisms, I want to offer a different opinion on this matter. In fact, I want to tentatively argue that one of the main things Steven and many others decry should be increased, not decreased.

Just to be clear: I will be addressing philosophy written for PhD students and professors. There are other audiences, for which the rules are different.

*****

One of Steven’s criticisms:

Here’s another beef. Articles are sooooooo looooooogggg. Get to the point already and shut the hell up. Back when I was a wee tot in grad school, one of my senior professors gave this advice: a philosophy paper should be at most 20 manuscript pages. At 30 it should be a world-beating article. At 40 just write a book. Back then, in the days of monospaced fonts, a page was about 250 words. So he was saying to keep it under 5000 words. Oh boy has that ship sailed. Papers now are routinely 30 pages or more in manuscript. A 10,000 word paper is medium length. Articles in supposed short-form journals like Thought or Analysis are more like the average-length article from 40 years ago.

Yes, articles are longer now. Steven has a view as to why this increase in length happened:

Why? Because authors think they need to address every teensy-tiny, picayune objection from every possible reader, all of whom will approach the paper with maximum malevolence. So half the paper is stage-setting, 40% is trying to forestall possible objections (or answer picky objections from the journal’s referees), and 10% is saying something new in needlessly jargony, technical language. Lord almighty, just tell me what topic you’re on about (10%), give me your new idea and positive arguments (80%) and stop me from making dumb retorts (10%).

Yeah, I get it. Here is my guess as to what’s behind Steven’s remark, at least in part:

If you take an article that has loads of stage-setting and objections, you, as the reader, could easily rip through the article as editor and remove the stage-setting that wasn’t needed and the objections that are too weak or otherwise irrelevant. When you were finished, the article would be much shorter, which would be a huge improvement.

I’ll get to the bit about objections in a minute. First, stage-setting.

I sympathize with the complaint about stage-setting being excessive. And it’s not just in the introductory sections of papers that we find excessive stage-setting; you can often find it peppered throughout a book or article. I recently started to read a book on the problem of evil, and I quickly realized that at least 90% of it was a waste of my time, since that material was either elementary, totally familiar, or addressed to readers with insane views (joke). (No, the book wasn’t advertised as being intended for students.) For the most part, the vanishingly few people who are going to read your paper or book already know the stage pretty well. There are exceptions to that rule, but not a lot.

I am proud of one article I published not because it was any good but because it has ZERO stage-setting. From the very first sentence, I got right to the point.

But I disagree about the complaint about objections. Sure, it’s annoying to wade through a bunch of objections that you personally don’t find terribly strong and thus a waste of time. However, the objection issue not nearly as straightforward as Steven’s remark suggests—although, again, please remember his post was a rant, by design and advertisement, so it’s not intended to be picked apart in the way I’m doing here! I’m using his post as a foil.

Before I begin to express my own views, I should say that I don’t have firm opinions here. I could easily be wrong about a lot of what I’m going to say. It’s not a simple topic. But quite frankly, I have better shit to do and I want to get this off my chest, okay?

I have six reasons why I think the complaint about objections is overblown.

First reason. If you want to get your paper published in a professional philosophy journal, then you’ll increase your odds of getting it accepted if you anticipate and treat the objections that a referee might make. So, you preemptively address as many as you can think of, even if they strike you as wrong-headed.

Steven is aware of this first reason. The other reasons have nothing to do with publication, however.

Second reason. What one fully competent philosopher thinks is a “teensy-tiny, picayune objection” (Steven’s words) another fully competent philosopher thinks is a serious objection requiring a response. It’s not as though the only people who would think that this particular objection here is serious, are clueless. The task of separating the serious from non-serious objections is not at all easy.

After the quote from Steven above I talked about how easy it is to act as editor and cut out unnecessary stage-setting and objections. But a different reader, with different views, and just as competent as you are, would say that you’ve cut out some of the most important parts—and you failed to cut out some of the least important parts. If they acted as editor, the result after deleting parts of the original paper would be quite different from what you got (unless of course this other reader shares most of your views).

So, the advice to shorten articles by addressing only the serious objections (and keeping only the essential stage-setting) seems like a good idea but I think it’s quite difficult to pull off successfully. Hence, we feel as though we’ve got to address objections even when we think they aren’t great.

Third reason. I suspect that many philosophers today are more appreciative, compared to the vast majority of philosophers in the past, of the phenomenon of thoroughgoing, long-term disagreement amongst fully competent philosophers regarding the quality of views, arguments, and objections. Most of us have been struck with the gravity of the following realization:

For virtually any philosophical position we are tempted to endorse, we can find many contemporary philosophers of distinction who not only disagree with us but understand quite well our reasons for our position—and they are roughly no more biased than we are.

The defensive response “Well, they just don’t understand my reasons, or they’re biased” is immature wishful thinking. I know of at least one senior philosopher who trots out that response, and he’s known to be wildly overconfident in his opinions. Offering that response to philosophical disagreement is a sure sign of immaturity. At some point early in one’s career, you get to have a few choice encounters with disagreeing philosophers who are willing to patiently address all your supporting arguments. One then discovers that the defensive response is silly, provided one has the intellectual courage to do so.

The realization italicized above makes many of us more cautious than the vast majority (not: all) of philosophers from the distant past. So, many of us take objections more seriously now, even if the only person who has raised them (thus far) is ourselves.

Fourth reason. The standards for philosophical argumentation have been slowly getting higher over time, thank God. This has occurred much more in some subfields (e.g., metaphysics, the philosophy of logic, and the philosophy of language) than in others. As our standards go up, we feel as though we need to make sure that all objections are dealt with. This reason overlaps the previous one.

Fifth reason. I suspect that philosophers today are more aware than philosophers from the past that misinterpretation is incredibly common and difficult to prevent. Addressing objections is a convenient and often effective way of clarifying one’s points. So, again, philosophers today are more willing to articulate and respond to objections.

Sixth reason. For some philosophical issues, the participants in the debate know that every possible position is counterintuitive in fairly extreme ways, due to the nature of the problem being addressed. Think of the liar paradox, the sorites paradox, or the problems of material composition in metaphysics (although the problems I have in mind aren’t limited to those topics). For these problems, appeals to what’s “obvious” or commonsensical doesn’t have as much weight. Objections are thereby taken more seriously.

*****

Another one of Steven’s criticisms:

What I’m complaining about is the faux precision, the endless abbreviations for an author’s idiosyncratic definitions of quasi-formal expressions, the attempts to make things look mathy, as if a few symbols sprinkled into a paper sharpen the vagueness of life. I’m not saying there’s no role for formal logic or specialized locutions. But I will say that these are needed much less than people seem to think.

I agree with some of this, for sure. I’m not sure what he meant by “the vagueness of life”, but I’ll ignore that.

1. Sometimes you can tell that an author is throwing in extra material not because it helps the paper but because it makes the author look like (a) they are impressive because they know math, (b) they are impressive because they know the science relevant to the philosophical issue being examined, (c) they are impressive because they’re rebellious, (d) they are superior to some famous person who has written on the topic, (e) they are a member of a currently fashionable clique, (f) they are virtuous, (g) they have the right political views, etc. That makes academic writing similar to a great deal of other nonfiction, right? Including Substack posts?

2. On occasion, the author’s abbreviations or names for key claims is mystifying. Why the hell did they choose to call that thesis “TRD-AT3”?? (Fictional example!) Then again, if you’re going to be repeatedly referring to multiple premises from multiple arguments, it’s not easy to come up with a system that is kind to the reader. You can use numerals for the premises of one argument and capital letters for the premises of another argument, but then the reader has to go back and forth constantly to figure out what the hell premise 3 was, what premise D was, and so on.

3. Authors occasionally write things like “Person P at time T assertively used sentence S in order to express proposition R”. In some cases, the paper could have done just as well without the P, T, S, and R stuff. Yep.

But quite frankly, I’d much rather there be more precision, not less, in philosophy. Yes, some of it will be annoying if not downright awful. You’re right, and it sucks. But on the whole it will vastly improve philosophy. I am not fully confident in this assertion, but I suspect it’s true!

There’s a cost-benefit evaluation to do here. The cost of adding lots of precision is that it’s painful to read. I don’t want to minimize that. Even so, I want to argue that the benefits exceed the costs in most cases.

But first: yes, for some papers, precision is a total waste of time. For instance, sometimes we’re writing a paper (that’s one of mine) in which the topic hasn’t been addressed in much detail and we frankly have only some vague, introductory thoughts about it that we think are worth publishing so that others can run with them. We think that publication is worthwhile because we know, or at least suspect, that only a few others have put much thought into the topic. So, it’s worth just putting the topic out there in the philosophical community, in rough form.

There’s nothing wrong with more tentative, rough papers like that. However, they should be rare exceptions, for two reasons. First, it’s better to have the patience to dig deeper and offer more than vague, introductory thoughts. Second, most of us are publishing on issues that have already seen highly developed work.

I have two main reasons for wanting there to be more precision in philosophy: it can prevent more misinterpretation, and it improves the quality of philosophizing.

In general, in order to make sure I (the reader) don’t miss any hidden premises in the author’s primary argument, I want the argument to be deductively valid. Hidden premises are often contentious, not innocent, so they shouldn’t stay hidden. In order to make sure the argument is deductively valid, I want to be able to symbolize it, even if neither I nor the author actually bothers to symbolize it. (One typically needs first-order logic plus propositional modal logic, but sometimes one needs second-order modal logic.) But if I am going to be able to symbolize it, it certainly helps me when the author throws in things like variables and other forms of precision.

For instance, I want to know what the quantifier domains are. Is the author’s claim about all people, just adults, just highly educated adults, adults that aren’t mentally disabled in some way, or what? Is she talking about all sensory experience, of all animals, or just humans, or just neurotypical humans, or just vision, or what? I also want to know what the intended modal strength is. The author might be offering a general claim that is supposed to provide necessary and/or sufficient conditions. Okay. But maybe it’s weaker, meant to capture just paradigmatic cases. Or maybe it’s stronger, meant to apply to other, non-actual possibilities. In that case, I want to know what the modal strength is.

So, the precision helps us understand what the premises and conclusions actually are. And of course, I need these precision devices in order to evaluate the premises and conclusions, after I know what they are!

This isn’t just a matter of helping out the reader. When forced to add precision, the author’s philosophizing will improve dramatically, in unforeseen ways.

Yeah, I know that some philosophers don’t think this is necessary. They have a point: it’s not necessary for publication purposes. However, one’s quality of philosophizing can massively improve if one bothers to take the time to be significantly more precise. Just try it.

When I was teaching in Hong Kong, a visiting philosopher gave a talk in normative ethics. I remember that during the talk I couldn’t figure out why the speaker thought her argument, if sound, ruled out the view she was arguing against. I just didn’t see how her argument went against the view at all, even if her argument was perfect. Sure enough, in the Q&A, my colleague Dan Marshall expressed the very same confusion. He and the speaker went THREE rounds trying to clarify things. We never figured out how the argument went against the view.

Now, the fault might have been with Dan and me, not the speaker. After all, neither Dan nor I had any expertise in ethics, whereas the speaker had plenty. But for God’s sake, it should be elementary to show people that your argument goes against a certain view! She thought it was clear. She was wrong about that, because she didn’t bother to be precise.

I recently finished revising a draft of a book on the philosophy of color. Practically the only reason the manuscript is any good is because I went through the trouble of making sure every argument was deductively valid and each key claim clarified. I did this by making sure I knew quantifier domains, modal strengths, and so on. I actually went to the trouble of symbolizing many of the arguments, just to be sure I wasn’t missing anything. Often enough, I was missing something. By doing that precision work, all sorts of possibilities, both helpful and detrimental, were revealed that I wouldn’t have thought of otherwise.

I know a philosopher who has the same MO for every topic, whether he’s trying to write his own thing or analyze someone else’s: put everything in logical notation. Every freakin’ important claim gets symbolized. He can publish pretty much anywhere he wants, and everything he writes gets published in a top journal. His record is absolutely outstanding. And he’s not a genius at all. Adding precision is like a superpower in philosophy. I could do the same thing, despite not being gifted, but I don’t give a shit about publishing or “making a contribution” to the profession.

I’m sure I’m missing a lot of relevant important points, but that’s my two cents for now!