The Enlightenment, the OED, and the History of Concepts, with Ngrams

In the fall of 2010, the online Oxford English Dictionary revised its entry for “Enlightenment.” Since 1891 the definition had read as follows:

1. The action of enlightening; the state of being enlightened …. [I]mparting or receiving mental or spiritual light.

2. Sometimes used [after Ger. Aufklärung, Aufklärerei] to designate the spirit and aims of the French philosophers of the 18th c., or of others whom it is intended to associate with them in the implied charge of shallow and pretentious intellectualism, unreasonable contempt for tradition and authority, etc.

For quite some time, admirers of the Enlightenment have had problems with the OED’s definition. As Peter Gay noted, it has the dubious distinction of collecting “most current prejudices in one convenient place” [Peter Gay, The Party of Humanity (Princeton, 1959) 263].

In contrast, the treatment of the Enlightenment in the OED’s new definition is quite well-behaved:

The dominant European intellectual culture in the 18th cent. which typically emphasized freedom of thought and action without reference to religious and other traditional authority, proposed a deistic understanding of the universe, insisted on a rationalist and scientific approach to the understanding of human society, the law, education, the economy, etc., and had as an important aim the development of new theoretical methods and practical reforms for these areas; (also) the period of time during which this climate of thought was dominant.

I had, as they say, a dog in this fight — admittedly, it was a little, yappy one, but a dog nevertheless: back in 2003 I published an article in the Journal of the History of Ideas examining some of the peculiarities in the OED’s definition and trying to figure out how it got to be the way it was (there’s an open access copy here). The article argued that not only was the definition rather flaky, it had also misquoted one of the three source quotes and misinterpreted another. It got the third quote right, but while a .333 batting average is respectable in baseball, one hopes for something better in a dictionary. In addition to addressing most of my gripes, the revised definition in the online OED is kind enough to refer readers curious about the history of the earlier definition to my JHI article (which means that, at least until the next OED revision, my place in History is secure).

The point of this post is not to gloat (well, not just to gloat) but (1) to say a few things about the new definition, (2) to discuss how much easier it is today to do what I was attempting to do back in 2003 (i.e., seven years before the arrival of the Ngram), and (3) to explore some of the implications of my adventures with the OED for the enterprise known as “the history of concepts.” This post will focus mainly on point 2 and offer a few thoughts on point 3. A sequel will address point 1 and have something more on point 3.

Learning to Write Like a Scottish Hegelian

Looking back on how I wound up writing the JHI article in the first place, what strikes me now is how unsystematic I was. While it had been clear for quite some time that a lot was wrong with the OED’s definition, I’d assumed—as I suspect most other readers of the OED assumed — that the definition offered an accurate, if lamentable, reflection of the way in which “enlightenment” was used by its critics during the nineteenth century. Since I was (and remain) interested in eighteenth-century discussions of the question “What is enlightenment?” and their later implications (this blog is an attempt at working through this obsession), I figured that the OED definition might provide some hints on how these discussions might have made their jump from German into English. And because I knew that Hegel was quite interested in these discussions (he’d copied Moses Mendelssohn’s answer to the question into a notebook he kept while in high school and he carried this notebook with him throughout his life), it was encouraging to see that two of the three sources quotes in the OED came from James Hutchison Stirling’s The Secret of Hegel (1865), the first book on Hegel to appear in English —or, at least, as close to English as a Hegel-smitten Scot could produce.

I’d already done some preliminary searching on WordCat and other places, trying to determine when “the Enlightenment” first began to turn up in book titles. The earliest seemed to be  John Grier Hibben’s The Philosophy of the Enlightenment (1910) and its opening pages betrayed what struck me as a good deal of insecurity over whether his readers would be familiar with term he’d used in his title. Within the space of two pages he referred to the period as “the Enlightenment, or Aufklärung,” as the “philosophical century,” as “the age of illumination, or enlightenment,” and, finally,“the age of reason.”  Clearly, here we had a man who was trying to cover all the bases.

Hibben, like Striling, was a Hegelian. He taught at Princeton and went on to become its President (right after Woodrow Wilson) and Princeton had his papers. I spent a couple of hours rummaging through them, which included some of his correspondence with his publisher, hoping to turn up some deep thoughts on the title he had chosen, or at least some indication that initial drafts or proposals might have employed a different title. No such luck. Having struck out with Hibben, I resigned myself to reading Stirling’s Secret of Hegel (and here is the place to say that this is not something I’d recommend).  When I did, I was surprised to see that the quotes the OED extracted from the book weren’t doing what the OED said they were doing. He used “enlightenment” not to refer to a particular historical period (i.e., the second of the old OED’s two definitions) but, instead, in something close to the OED’s first definition. Odder still, when he wanted to talk about what we would call “the Enlightenment,” he either retained the German Aufklärung or, in a few cases, employed the term “Illumination.”

At this point, I settled on something approximating a plan of attack:  to figure out when the English stopped using Aufklärung to refer to “the Enlightenment” and shifted over to using “the Enlightenment,” what I needed to find was a place where an English speaker would be forced to produce a word that did what Germans did when they used Aufklärung in the way that Hegel had started using it in his Berlin lectures on the philosophy of history and the history of philosophy (for any Hegel fans out there, I don’t think the section in the Phenomenology on die Aufklärung is about “the Enlightenment,” but let’s leave that for another time). I figured that the most obvious places to find this would be in translations of works by Hegel and histories of German philosophy. Fortunately, there weren’t that many of them and it was simple enough to figure out where the German term would appear in the text that was being translated.  It was while attempting to track down early English translations of German philosophical texts, that I became aware of the existence of the German Museum, a short-lived British periodical that translated a remarkable amount of late eighteenth-century Enlightenment texts from German into English, among them, a translation of Moses Mendelssohn’s answer to the question “What is enlightenment?”.  Fortunately, the rare book room at Harvard owned one of the few surviving copies of the journal and working my way through the journal confirmed my growing  suspicion that the preferred translation for Aufklärung appeared to be “mental illumination.” The subsequent fleshing out of my argument was greatly aided by David Armitage’s suggestion that I might want to spend some time poking around in the Anti-Jacobin Review to see what terms were being thrown around there and by Darrin McMahon’s lesson on the jargon used by French anti-philosophes.

It was on the basis of this shaky foundation that I wrote an article maintaining that it was not until sometime around the close of the nineteenth century that the term “Enlightenment” came into use in English as a way of referring to a discrete historical period. I figured that if I was wrong, someone would let me know and I consoled myself with the thought that I was doing what Karl Popper said we ought to do: make bold conjectures and see if anyone shoots them down.  So I shipped the article off to the JHI, waited a year or so while they reviewed it and, after they’d signed off on it, I forwarded a copy of it to the OED.  I eventually received an email letting me know that they were on the case.

Enter the Ngram

Ten years later all of this would have been so much easier. For one thing, Google’s long march through the libraries means that a host of nineteenth-century texts, including the German Museum were now available online, which means that I can now read this strange periodical on my iPad while commuting into my office.  (Pause for a moment to reflect on how strange and wonderful this is).  And then, in the winter of 2011, Google said “let there be Ngrams!” and, behold, there were Ngrams. And they were good — or at least good enough to do a job like the one I was trying to do … and do it very quickly.

Thanks to the happy accident that English, unlike German, does not necessarily precede its nouns with articles and gives writers the option of capitalizing or not capitalizing those nouns, all I would have needed to do is plug “the Enlightenment,” “the Illumination,” and (since Stirling credited himself with having introduced a German term into English) “Aufklärung” into the search fields and hit the search button. Out would pop things like this:

Aufklärung,Ill, and Enl

The Ngram seems to lend support to my rash conjecture. There are very few uses of “the Enlightenment” until after the turn of the twentieth century (note the little bump that coincides with the publication of Hibbins’ book),  And then the ascent begins. The flat lines for “the Illumination” is a bit more surprising: I would have expected it would be popping up more often than it does (the sequel to this post will explain why that expectation is silly). Probing inside the results led me, momentarily, to think that I’d manage to refute my own thesis: the Ngram gives a link to an 1822 translation of Eucken’s history of philosophy, which used “the Enlightenment” in precisely the way that I insisted it wouldn’t be used until the twentieth century. But it turns out that Google’s metadata was faulty: the publication date was, in fact, 1922. Faulty metadata also accounts for some of the scattered appearances of Aufklärung that the Ngram picks up: it appears that Google classified certain German publications as “English books.”

The History of Concepts in the Age of Digital Reproducibility

So (with apologies to Walter Benjamin), what are the implications of my adventures for work on the “history of concepts” in the age of digital reproducibility?

First, and, most obviously, things have gotten a lot easier.  Imperfect though the Ngram may be, it would be crazy not to run a few of them before setting out on the sort of research that I did.

Second, the ease with which Ngram can be thrown together forces some reflection on  what the “history of concepts” allegedly involves. Reinhart Koselleck once insisted that a “history of concepts” should not be confused with a “history of words,” since a concept can be designated by any number of different words [see Koselleck, Futures Past 85-6]. For example, the entry on “Aufklärung” in Geschichtliche Grundbegriffe deals with “Age of Reason” as well as “Enlightenment.”  Koselleck might want to argue that while the OED got the history of the word “enlightenment” wrong, this error is of little importance for the historian of concepts (though it surely did matter to the OED, which is very much in the business of tracing the history of words). Koselleck might go on to argue that, while Stirling did not use the word “enlightenment” to designate the concept that we have come to designate as “the Enlightenment,” he had another word that did more or less the same job: “illumination.” Once we make the necessary translations, it should be clear enough what Stirling was up to.

My difficulties with this argument center on the “more or less.” For, in an important way, this interpretation fails to capture what Stirling was doing: it pays insufficient attention to the degree to which the choice of one word rather than another sometimes matters. Quentin Skinner has rightly (at least in my view) emphasized that the best evidence that “a group or society has entered into the self-conscious possession of a new concept is that a corresponding vocabulary has been developed, a vocabulary that can then be used to pick out and discuss the concept in question with consistency” [see Skinner, Visions of Politics I:160]. I suspect that the reverse might also be the case. The plethora of alternative terms (“Aufklärung,” “Illumination,” “Age of Reason”) that Stirling and others employed to designate what we now call “the Enlightenment” suggests that he might not have understood the period in the quite the same way as we do. And his vacillation about what to call this thing that had happened (or might still be happening) is very much in keeping with the patterns of usage that prevailed during the period whose grip he was hoping to escape. Late eighteenth-century texts manifest an obsessive concern that certain terms have been misused, a conviction that certain words can no longer be employed in the way in which the authors of theses texts would like to use them, and a zeal for tracing connections between the different words that are in play. It is as if an entire vocabulary has become suspect.

More on the implications of such suspicions next time.

About James Schmidt

Professor of History, Philosophy, and Political Science Boston University
This entry was posted in Uncategorized and tagged , , , , . Bookmark the permalink.

7 Responses to The Enlightenment, the OED, and the History of Concepts, with Ngrams

  1. Pingback: Further Thoughts on “the Enlightenment,” the OED, the History of Concepts | Persistent Enlightenment

  2. Pingback: Deeper into the “Shallow Enlightenment” (Ludwig Tieck, George J. Adler, and Herman Meville) | Persistent Enlightenment

  3. Dave Mazella says:

    Ha! Well, Skinner’s contribution has been important, but from my perspective as an onlooker in lit studies, there were a bunch of approaches in “political thought” that seemed quite similar to me, at least at the surface level: Koselleck’s conceptual history; Pocock and the Cambridge School; Foucault’s discourse analysis in his Governmentality phase; and then finally Skinner’s rhetorical approach. I ended up using them all, one way or the other, in my Cynic book, but I was very surprised by the fact that these different writers and their followers seemed not to attend to one another, and their work was read very selectively in lit studies “theory” circles. So my view is that there were some missed opportunities in the 90s and 00s that could be followed up with the electronic goodies we now have access to.

  4. Dave Mazella says:

    Aha, now I need to go read Gitelman and Rosenberg, thanks. I don’t really have a problem with datedness of the Koselleck project. It is what it is, and represents a particular historical moment. What I’m interested in seeing, though, is some theoretical work that works in the intersection between “big data” and Koselleck’s approach, which never seemed to quite coalesce into a methodology. My suspicion is that if an appropriate methodology could be formalized, it could become quite a powerful model for humanities scholarship.

    • At some point, I need to go back and rethink what I wrote about the Geschichtliche Grundbegriffe business back at the close of the last millennium in History of European Ideas 25 (1999) 9-14 (there’s an open source preprint here). It was a hastily composed contribution to a roundtable at the American Political Science meetings organized by Mel Richter that included contributions from David Armitage and others, which were also included in the issue, along with some final thoughts by Mel. In the article, I suspect I come off as more Skinnerian than Skinner. Maybe that’s not a bad thing?

  5. Dave Mazella says:

    Thanks, James, for giving us this behind-the-scenes look into conceptual history as it could be practiced in 2003 and 2013. Ten years have made an enormous difference, at least in terms of access. I agree that the easy availability of n-Grams and Google books, not to mention ECCO and now Gale’s new 19c collections, make an enormous difference in terms of scholarly labor, but what I’m not seeing is a methodological advance that acknowledges the huge new repositories of sources we now have. Or am I missing something?

    One of the obvious differences between the conceptual history of Koselleck’s era and our own is the ready availability of sources in multiple genres. Koselleck and co. tended to work very closely with “classic” texts of political thought, or at least whatever texts in political thought were available to them in their research libraries. Now we can trace words across genres, within genres, across brief and extended periods, and so forth. So I am hoping that more work on conceptual histories could be done to compare the differences between conceptual formations in literary as opposed to philosophical genres (to the extent these can be distinguished). This could account for some of the persistent differences in usage you and others have found.

    • Thanks, Dave, for the comment and for all the good work you’ve been doing for so long over on The Long Eighteenth.

      Certainly, those of us who are working on eighteenth-century texts are at a disadvantage in doing the sorts of things that our colleagues who work on later periods are able to do, largely because we’re having to deal with texts that have been scanned, but not digitized, or which resided on commercial sites (like ECCO) that limit the sorts of access we can get, or both. Daniel Rosenberg has a good discussion of some of these problems in his contribution to Lisa Gitelman’s collection Raw Data is an Oxymoron (just out from MIT Press). But there are a few good examples of really smart attempts producing striking data visualizations using the sources we do have. For an example, see the amazing stuff that Dan Edelstein and his colleagues are doing at the Mapping the Republic of Letters project, which ought to serve as a model for pushing “big data” projects back into earlier periods.

      Like everyone else, I have lots of reservations about the Geschichtliche Grundbegriffe venture, particularly the rather heavy handed application of the Sattelzeit business (but, damn, those volumes are impressive looking!). I hope to post something on this later. I suspect that later generations will look back on it the way we look back on Charles Babbage’s analytical engine (with an army of graduate students playing the role of Babbage’s levers and gears). I agree with your suggestion that more attention to the migration of terms across genres would help, particularly for what I’m trying to do. It strikes me that Bookworm can serve as a sort of “proof of concept” of how this might be done (and here, I suppose, I should confess that my interest in this project is not solely professional!).

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.