I’ll start by simply listing the ten most popular topics (out of the ninety) in each of the five eras. The popularity metric I’m using is the expected number of articles in that topic from that era. For each era-topic combination, I look at all articles in that era, and sum the probabilities that the article is in that topic. Then I rank the topics by this probability sum to get the following table:
|Idealism||Ordinary Language||Ordinary Language||Ordinary Language||Ordinary Language|
|Life and Value||Meaning and Use||Sets and Grue||Arguments||Composition and Constitution|
|Ordinary Language||Definitions||Promises and Imperatives||Theories and Realism||Arguments|
|Methodology of Science||Methodology of Science||Meaning and Use||Reasons||Modality|
|Definitions||Other History||Truth||Composition and Constitution||Knowledge|
|Physicalism||Life and Value||Moral Conscience||Causation||Evolutionary Biology|
|Propositions and Implications||Moral Conscience||Propositions and Implications||Modality||Truth|
|Dewey and Pragmatism||Universals and Particulars||Universals and Particulars||Population Ethics||Reasons|
The striking thing is that ordinary language philosophy comes out top in each era after World War II. This is related to the fact that it is as much a style as a topic, and the style never really went away.
We can also look at the bottom ten topics in each era, by the same metric. (The lowest ranked topics are at the top of this list. Any time I’m listing the lowest ranked anything, I’ll put the lowest at the top.)
|Formal Epistemology||Population Ethics||Vagueness||Psychology||Psychology|
|Population Ethics||Abortion and Self-Defence||Wide Content||Beauty||Heidegger and Husserl|
|Belief Ascriptions||Cognitive Science||Races and DNA||History and Culture||History and Culture|
|Cognitive Science||Wide Content||Cognitive Science||Races and DNA||Beauty|
|Frankfurt Cases||Vagueness||Models||Social Contract Theory||Faith and Theism|
|Medical Ethics and Freud||Races and DNA||Psychology||Ancient||Verification|
|Game Theory||Egalitarianism||Game Theory||Faith and Theism||Idealism|
|Abortion and Self-Defence||Feminism||Kant||Functions||Social Contract Theory|
|Norms||Models||Functions||Origins and Purposes||Dewey and Pragmatism|
That’s not too surprising. Formal epistemology was not a big deal in philosophy pre-1945; Heidegger and Husserl haven’t had much presence in these journals in this century.
But both of these tables, while somewhat useful, have a flaw. The ninety topics aren’t all the same size. So this table is telling us as much about how the model carved up different parts of philosophy, as it is telling us about trends in the field. Here’s a better approach. Take the probabilistic measure I just used, and divide it by the size of the topic as a whole. (That is, the expected number of articles in the topic across the whole data set from 1876-2013.) And use that to rank the topics. In effect, within each era we’re ranking the topics by what propoportion of the work on that topic was in that era. So it’s a measure of what were the distinctive topics within each era. Here is what we get.
|Psychology||Verification||Crime and Punishment||Wide Content||Models|
|Idealism||Meaning and Use||Promises and Imperatives||Population Ethics||Norms|
|Life and Value||History and Culture||Sets and Grue||Vagueness||Composition and Constitution|
|Other History||Heidegger and Husserl||Egalitarianism||Cognitive Science||Cognitive Science|
|Self-Consciousness||Beauty||Intention||Belief Ascriptions||Formal Epistemology|
|Physicalism||Definitions||Duties||Abortion and Self-Defence||Evolutionary Biology|
|Dewey and Pragmatism||Analytic/Synthetic||Radical Translation||Justification||Vagueness|
|Beauty||Dewey and Pragmatism||Speech Acts||Reasons||Races and DNA|
|Early Modern||Faith and Theism||Abortion and Self-Defence||Personal Identity||Modality|
|Kant||Marx||War||Radical Translation||Population Ethics|
And that was a little surprising to me. I had not realised how much the work on crime and punishment (in these twelve journals) was concentrated in 1966–1981. If you’d given me twenty guesses for what would be the distinctive topic of this era, probably the era I care most about in all of philosophy, I wouldn’t have guessed this. And I would not have guessed that promises and imperatives (which remember includes a lot of deontic logic), and sets and grue, would have been second and third.
In general, most of the topics turn up one to two eras later than the most famous works in those eras. This makes some sense; the secondary literature has to come after the primary literature. But it means that a story of the history of philosophy that concentrates on the great works will leave a misleading impression of when topics were being most discussed.
Let’s do the same thing but for the topics that have the lowest share of their articles turning up in a given era.
|Formal Epistemology||Population Ethics||Idealism||Idealism||Idealism|
|Population Ethics||Cognitive Science||Psychology||Psychology||Psychology|
|Belief Ascriptions||Egalitarianism||Wide Content||Beauty||Heidegger and Husserl|
|Norms||Wide Content||Vagueness||Life and Value||Verification|
|Cognitive Science||Abortion and Self-Defence||Other History||Verification||History and Culture|
|Radical Translation||Vagueness||Life and Value||Other History||Life and Value|
|Modality||Evolutionary Biology||Cognitive Science||Definitions||Meaning and Use|
|Frankfurt Cases||Formal Epistemology||Norms||History and Culture||Beauty|
|Egalitarianism||Composition and Constitution||Evolutionary Biology||Meaning and Use||Definitions|
|Models||Modality||Dewey and Pragmatism||Universals and Particulars||Faith and Theism|
For me, the middle column is the most interesting one. The last two columns are pretty much what we’d expect. And the first two columns are as much noise as signal. In the first column the absolute values are all low, and one or two articles that by coincidence shared some keywords with topics from a century later could move something up or down by several spots. In the second column, the fact that the model doesn’t see consequentialism and utilitarianism as having much in common drives the “top” result.
But in the middle we get an interesting mix. Idealism is over, so all the idealism-related topics are relegated. And the journals that are still publishing history are settling on just publishing work on the same few big names, so some history topics are relegated. But other topics have yet to start. It isn’t a surprise that before 1982 there wasn’t much work on vagueness or on evolutionary biology in these journals. (There are important articles in those fields from before 1982, but they don’t tend to be in these twelve journals.) And norms is as much a style as a topic. But I’m a bit surprised cognitive science hadn’t started getting more attention yet. And I’m very surprised that wide content is showing up here.
Remember that Putnam’s “Meaning and Reference” (Putnam 1973) is from the very middle of this era. And given the importance of twin Earth debates to philosophy of the last forty years, I would have thought it would have kicked off a huge debate. But it just didn’t. I’ve already discussed this when talking about the wide content topic, but it is just shocking to me how long it takes for the literature on Wide Content to really get going.