As Dan Edelstein once observed, scholars have gotten into the habit of using Kant’s 1784 to the question “What is enlightenment?” as a convenient “one-stop shop for defining the Enlightenment.”¹ There is a tendency to assume that because Kant was obviously an important eighteenth-century thinker, his attempt to explain just what was taking place at the close of the eighteenth century played a significant role in how the period was understood. Since I am as guilty of that tendency as the next person, I was somewhat taken aback when I first read Gisbert Beyehaus’ 1921 article on Kant’s “program of enlightenment” and came across his complaint that previous scholars had tended to treat Kant’s answer in a “step-motherly” fashion.² At the time, I wrote this off to the tendency that scholars have to make it seem as if they are breaking new ground, even in those cases when they are moving down well-trodden paths (and, in any case, there was enough in Beyerhaus’ article that was new to me to get me interested in the context of Kant’s answer).

Recently it occurred to me that the Google Ngram might provide a quick (albeit dirty) way of exploring the reception of Kant’s essay. Matters are made somewhat easier by the fact that the famous opening paragraph of Kant’s contains a combination of words that is unlikely to occur in any other text unless it is quoting or discussing Kant’s article: the famously paradoxical formulation “selbstverschuldeten Unmündigkeit” (I’ve discussed the problems of translating this phrase in an earlier post). Further, Kant places this phrase at the very start of the essay, in a paragraph that has tended to be the most frequently quoted part of the essay. So, I figured, plugging selbstverschuldeten Unmündigkeit into the Ngram Viewer might provide a rough sketch of the reception of Kant’s essay. Here are the results:

Readers who are familiar with the peculiarities of the Ngram may notice that I’ve left Google’s default smoothing on. Turning off the smoothing results in a somewhat different — and I suspect more accurate — picture of the discussion of Kant’s essay, with intermittent, but relatively intense, interest in the essay in certain years (note the peaks at 1799 and 1839):

Looking at the samples Google provides confirms that most of the texts are, indeed, quoting the opening of Kant’s essay (in other words, “selbstverschuldeten Unmündigkeit” is not a term that turns up very frequently in contexts other than a discussion of Kant). But it also indicates that the peaks we are seeing in 1799 and 1839 are, in part, being driven by editions of Kant’s works, as opposed to actual discussions of Kant’s essay. Since I’m more interested in the latter than the former, it would be useful to screen out the occurrences of the phrase that occur in republications of Kant’s essay. Fortunately, the Ngram provides a way of doing this.

It is possible to construct an Ngram that subtracts the occurrences of one phrase from another and plots the result. The trick, of course, is to come up with another phrase in the essay that, while unlikely to occur in  texts from the period other than Kant’s essay, will probably not be turning up on the Ngram unless Kant’s entire essay is being reprinted. After playing around with a few possible contenders and checking to make sure that they weren’t phrases that were likely to turn up in contexts other than editions of Kant’s essay, I came up with “Privatgebrauch nenne ich denjenigen” (which appears at the point where Kant introduces his distinction between public and private uses of reason, but is not an interesting enough passage to stand on its own as a quotation).   Here is the result:

Since it may be easier to make sense of this by actually looking at the Ngram itself (where Google gives you the ability to highlight different lines on the graph), it would be worth opening this link in a separate window and playing around with it as you read what follows (obviously, this is easier if you have a really big monitor). The green line shows us the result for all of the uses of “selbstverschuldeten Unmündigkeit” that occur in texts that do not also include the phrase “Privatgebrauch nenne ich denjenigen.” It is my suspicion that this provides us with discussions of Kant’s essay, as opposed to editions of Kant’s essay. One way of appreciating the differences between editions and discussions is to go to the Ngram itself and move the cursor between the graph for “selbstverschuldeten Unmündigkeit” and the one for “selbstverschuldeten Unmündigkeit – Privatgebrauch nenne ich denjenigen.” Notice how the subtraction of the term dampens the peaks at 1799 and 1840? What is being eliminated here are Kant editions.

I am perplexed by the negative result on the “selbstverschuldeten Unmündigkeit – Privatgebrauch nenne ich denjenigen” line right before 1800. If it is accurate (and I doubt it is), this would indicate that there are more occurrences of “Privatgebrauch nenne ich denjenigen” than of “selbstverschuldeten Unmündigkeit,” hence the tipping of the line into negative territory. While I suppose it could have been the case that was a flurry of discussion of Kant’s public/private distinction, I doubt it. What is more likely, I suppose, is that we probably should not trust what we are seeing for the 1799 results at all. If we simply limit ourselves to the post 1801 data, what we see is a discussion of Kant’s essay that is more or less dormant until shortly before the peak at 1840, has a lesser peak at 1870, and then begins a persistent rise around the beginning of the twentieth century. That rise around 1900 conforms rather nicely with what we’ve seen in earlier discussions of the gradual fading of the various pejoratives that once preceded the work Aufklärung, a fading that I’ve suggested might mark the point when Aufklärung stopped being a contested term denoting a process that could be explained in a variety of ways and began to be used to denote a (no less contested) historical period.

Of course, nothing that I have said here speaks to Beyerhaus’ complaint that, prior his essay, scholars had treated Kant’s essay in a “step-motherly” fashion.  But it does cast some light on a more basic question:  just when did scholars begin to deal with Kant’s essay at all.

1. Dan Edelstein, The Enlightenment: A Genealogy (University Of Chicago Press, 2010) 117. ↩
2. Gisbert Beyerhaus, “Kants Programm der Aufklärung aus dem Jahre 1784,” Kant-Studien 26 (1921) 2-16, 2. ↩