2.24 Ordinary Language

Category: Ethics/Philosophy of Mind

Keywords: ask, certainly, really, surely, try, anything, quite, answer, else, answers, tell, think, put, saying, perhaps

Number of Articles: 1395
Percentage of Total: 4.3%
Rank: 2nd

Weighted Number of Articles: 1740.3
Percentage of Total: 5.4%
Rank: 1st

Mean Publication Year: 1962.6
Weighted Mean Publication Year: 1967.5
Median Publication Year: 1963
Modal Publication Year: 1956

Topic with Most Overlap: Meaning and Use (0.0379)
Topic this Overlaps Most With: Emotions (0.0819)
Topic with Least Overlap: Space and Time (0.00114)
Topic this Overlaps Least With: Models (0.02079)

A scatterplot showing which proportion of articles each year are in the ordinary languagetopic. The x-axis shows the year, the y-axis measures the proportion of articles each year in this topic. There is one dot per year. The highest value is in 1884 when 10.0% of articles were in this topic. The lowest value is in 1893 when 3.5% of articles were in this topic. The full table that provides the data for this graph is available in Table A.24 in Appendix A.

Figure 2.60: Ordinary language.

A set of twelve scatterplots showing the proportion of articles in each journal in each year that are in the Ordinary Languagetopic. There is one scatterplot for each of the twelve journals that are the focus of this book. In each scatterplot, the x-axis is the year, and the y-axis is the proportion of articles in that year in that journal in this topic. Here are the average values for each of the twelve scatterplots - these tell you on average how much of the journal is dedicated to this topic. Mind - 6.7%. Proceedings of the Aristotelian Society - 11.4%. Ethics - 5.3%. Philosophical Review - 6.0%. Analysis - 4.0%. Philosophy and Public Affairs - 5.4%. Journal of Philosophy - 5.0%. Philosophy and Phenomenological Research - 4.9%. Philosophy of Science - 2.7%. Noûs - 6.0%. The Philosophical Quarterly - 7.6%. British Journal for the Philosophy of Science - 3.1%. The topic reaches its zenith in year 1884 when it makes up, on average across the journals, 10.0% of the articles. And it hits a minimum in year 1893 when it makes up, on average across the journals, 3.3% of the articles.

Figure 2.61: Ordinary language articles in each journal.

Table 2.59: Characteristic articles of the ordinary language topic.
Table 2.60: Highly cited articles in the ordinary language topic.

Comments

The biggest topic in the model is also one of the hardest to classify. I’ve called it ordinary language philosophy, and you can see from the graphs that it peaks in the years we associate with ordinary language philosophy, but beyond that it gets hard to say precisely what it is.

It is such a big deal in the years after World War II, especially in the Proceedings of the Aristotelian Society, that I had to change the scale for the graphs by journal. And yet there are very few articles that are unambiguously in this topic.

Part of what’s going on here is that what the model is finding is a style as much as a content. Take a look at the keywords at the top of the page: ask, certainly, really, surely, try, anything, etc. These aren’t philosophical topics, but they are a way of talking about philosophy. And they were a distinctive enough style that the model, which is always on the lookout for correlated word usages, declares it a big topic.

To get a sense of just how distinctive the style is, take a look at the frequency of various words that were popular with the ordinary language philosophers. These graphs are taken straight from the JSTOR data. So they cut out all the one and two letter words, and some stop words that JSTOR filters out, but they don’t remove all the other words that I took out. So the fractions on the y-axis show how often the word appears as a function of almost all the words in the article, not just of the words included in this study.

The dashed lines are roughly speaking the average rate that the word appears across the whole study. But not exactly. More precisely, it is the average across the 138 years of the rate of the word in each year. I thought that was a little more representative of the role of the word in the study, rather than just having the huge number of articles in recent years swamp the averages.

A scatterplot showing the frequency of the words ask, surely, try, put, tell. The word ask appears, on average across the years, 375 times per million words, and in the median year, it appears 359 times per million words. Its most frequent occurrence is in 1961 when it appears 635 times per million words, and its least frequent occurrence is in 1876 when it appears 187 times per million words. The word surely appears, on average across the years, 349 times per million words, and in the median year, it appears 348 times per million words. Its most frequent occurrence is in 1883 when it appears 653 times per million words, and its least frequent occurrence is in 1890 when it appears 154 times per million words. The word try appears, on average across the years, 309 times per million words, and in the median year, it appears 308 times per million words. Its most frequent occurrence is in 1964 when it appears 527 times per million words, and its least frequent occurrence is in 1889 when it appears 109 times per million words. The word put appears, on average across the years, 489 times per million words, and in the median year, it appears 472 times per million words. Its most frequent occurrence is in 1878 when it appears 706 times per million words, and its least frequent occurrence is in 1910 when it appears 363 times per million words. The word tell appears, on average across the years, 227 times per million words, and in the median year, it appears 228 times per million words. Its most frequent occurrence is in 1963 when it appears 382 times per million words, and its least frequent occurrence is in 1882 when it appears 103 times per million words.

Figure 2.62: Popular words in ordinary language.

A rise is visible in the frequency of each word in the years after World War II. In some cases this is restoring the level that existed back in the 1870s and 1880s, but in all cases there is a steep fall from the mid-1960s onwards. Let’s see what happens if we restrict this to the journal that seems central to ordinary language philosophy, Proceedings of the Aristotelian Society. (Note that the averages on this graph are averages for just this journal, so they are higher than the overall averages.)

A scatterplot showing the frequency of the words ask, surely, try, put, tell in the journal Proceedings of the Aristotelian Society. (All stats from now on just refer to that journal.) The word ask appears, on average across the years, 494 times per million words, and in the median year, it appears 465 times per million words. Its most frequent occurrence is in 1941 when it appears 1007 times per million words, and its least frequent occurrence is in 2004 when it appears 158 times per million words. The word surely appears, on average across the years, 405 times per million words, and in the median year, it appears 397 times per million words. Its most frequent occurrence is in 1954 when it appears 767 times per million words, and its least frequent occurrence is in 1892 when it appears 0 times per million words. The word try appears, on average across the years, 454 times per million words, and in the median year, it appears 417 times per million words. Its most frequent occurrence is in 1966 when it appears 1375 times per million words, and its least frequent occurrence is in 1891 when it appears 70 times per million words. The word put appears, on average across the years, 575 times per million words, and in the median year, it appears 566 times per million words. Its most frequent occurrence is in 1962 when it appears 999 times per million words, and its least frequent occurrence is in 1891 when it appears 105 times per million words. The word tell appears, on average across the years, 277 times per million words, and in the median year, it appears 260 times per million words. Its most frequent occurrence is in 1957 when it appears 615 times per million words, and its least frequent occurrence is in 1920 when it appears 41 times per million words.

Figure 2.63: Popular words in ordinary language in Proceedings of the Aristotelian Society.

And again you see much higher frequencies in the 1950s and 1960s.

There is a question about whether I should have manually filtered out something like this whole topic. I could have added ask, quite, else and so on to the stop words that I filtered out. If I’d been really aggressive about this, I could have gotten rid of this whole topic. And maybe I should have done this. There are a few topics that are like this. I could have added argument as a stop word and gotten rid of the arguments topic. I could have added concept as a stop word and gotten rid of the concepts topic. There is an interesting question about where to stop, and how tightly to focus on topics as opposed to tools (like arguments or concepts), or styles (like ordinary language). I do think that if I were starting over I would try to strip some of the keywords here out, but it’s not obvious to me what’s right.

There are a lot of highly cited articles in this topic. That’s largely because there are a lot of articles in this topic. Notably, a lot of the highly cited articles here are well after the peak of ordinary language philosophy. The highly cted papers are by a fairly disparate set of authors. But one striking thing is that, to my eyes at least, there are a lot of very good writers represented here. So we see papers, for instance, by Bernard Williams, Frank Jackson, Stephen Yablo and Philippa Foot. On the other hand, there is also a Wittgenstein paper here, and I never thought of Wittgenstein as a great stylist, so maybe it’s just a coincidence. But mostly I think most writers would be better off if they tried to adopt the styles of the midcentury greats.

Finally, note that I’ve put this topic in two categories. While the papers here to cover a huge range (there are fourteen hundred of them after all), there do seem to be two big clusters in ethics and in philosophy of mind. I’ll go over how I split topics in two when I come to another topic, sets and grue, that is even more binary than this one.