Brian Weatherson (Michigan) has a really interesting project featured at Daily Nous applying methods of 'topic modeling' to the history of philosophy. One of the many interesting findings Weatherson presents is that journal articles have gotten progressively longer in recent years, indeed longer than they have ever been. Here's the relevant graph:
In brief, while the average paper (50% percentile) was just less then 10 pages at points in the 1950's-60's, it is now double that. Similarly, while papers upwards of 30 journal pages were rare outliers throughout the 20th Century, now a full 10% of articles are upwards of 30 journal pages and outliers upwards of 35 pages.
Is this a problem? Are journal articles getting too long? Weatherson thinks so, writing, "This feels like a bad thing; articles are getting bloated, and we need to find a way to get them back to a reasonable length." For my part, I'm a bit torn for reasons outlined below. But I'm really interested to hear what you all think!
Here are a few brief thoughts of my own. On the one hand, I personally don't feel like journal articles are getting too long and bloated. First, I've by and large enjoyed the fact that a fair amount of philosophy articles seem to be getting longer and more ambitious. One of my biggest frustrations in graduate school was that I was routinely told (and heard) that "a journal article should make one small point." This seemed to me to turn philosophy into little more than "making small moves" akin to moving a pawn one space forward in a chess game. While I think work making small moves can be valuable, as a disciplinary norm it seemed to me to pretty stultifying, making philosophy on the whole rather boring and discouraging the kind of more ambitious work that attempts to take larger steps forward. The way I see it, pluralism is good. By all means, let's have short papers. Indeed, I've suggested that journals should make more space for short 'reply' pieces, not less. But let's not revert to a norm that papers 'must be short.' Long and short works tend to do different kinds of valuable things: 'lumping' and 'splitting' (roughly, systematizing vs. precisifying). And some of us are better at the former than we are at the latter, and vice versa. Instead of saying, "papers should be shorter" or "papers should be longer", how about this: journals should be open up to papers of various sizes, and reviewers and editors should seek to ensure that papers are no longer than they need to be to accomplish what they need to (i.e. make an important contribution to the literature). This seems to me like the kind of norm we should all want.
That being said, I do have one reservation (hence my earlier comment about being a bit torn). In brief, I think journal articles these days may be unnecessarily long and bloated in one particular respect. Just yesterday, I finished what I think is a nice, crisp 5,000 word paper, far shorter than the kinds of papers I normally publish. Alas, later in the evening, it immediately occurred to me that I should probably include a "replies to objections" section that is almost certainly going to add 2,000 or so words to it. Why? More or less, because I imagined some reviewers may have particular objections, and because it is expected that authors address plausible objections. However, are these actually good reasons to make a paper longer? As Mike Huemer writes here, this specific feature of the review process seems to incentivize longwindedness:
If you make a point too quickly, referees will think it must be superficial. Thus, you need to add a bunch of discussion, even if it’s not actually helpful. If you discuss some misunderstandings of your point, and some bad reasons why someone might disagree with it, it’ll give the referee more confidence, even if those misunderstandings and bad reasons aren’t things he would have been tempted by. For this reason, it may help to discuss your idea with other philosophers.
Why should a journal article need to raise and respond to possible misunderstandings and objections? Why not instead simply ask, "Is this article's argument interesting enough to publish?", and then let the subsequent literature explore possible objections? As Huemer adds:
Now, sometimes a paper should be rejected because the argument is no good. But that doesn’t just mean any argument that you think is mistaken. If the argument is one that some intelligent and reasonable philosophers would find plausible, then that should be good enough.
Of course, with that weak standard, many more papers would be accepted . . . but the journals don’t have enough room for more papers. Solution: more papers should be rejected for reason (e) under III.A.1: uninterestingness. I would like to see more controversial arguments, but fewer semantic arguments, fewer small technical points, fewer “meta” arguments about the state of the discussion or the relationship between so-and-so’s theory and some other theory…
When refereeing a paper, the question to ask is not, “Is this convincing?” The question is, “Does this add to the discussion? Would people like to talk about it?”
This seems to me exactly right. If papers should be shortened at all (as Weatherson suggests), let's shorten them by changing this norm in particular. Instead of expecting authors to 'address possible objections', let's just ask: is this paper's argument interesting enough to publish? If this became the norm, I think the result would almost certainly be shorter papers, as well as (potentially) more subsequent discussion in the literature.
But these are just my thoughts. What are yours?
Leave a Reply to Robert A GressisCancel reply