Language Log

There's a little lake near here called Sob Lake. I only recently learned the etymology of this name. According to Akrigg and Akrigg's British Columbia Place Names, the lake was originally named by a survey party. Finding the homesteader who lived nearby obnoxious, they recorded their opinion of him by naming the lake "S.O.B. Lake". The authorities in Victoria, however, felt that this was improper and bowdlerized it to "Sob Lake" by removing the periods.

In a post last February I wrote about Yale professor of language and literature Thomas R. Lounsbury (1838-1915), whose 1908 book The Standard of Usage in English bucked the priggish prescriptivism of the era. More recently, Arnold Zwicky hailed his English Spelling and Spelling Reform as "a bracing, sharp-tongued book" full of "elegant rants." Lounsbury also played a major role in the history of language study at Yale, along with his more famous colleague William Dwight Whitney. In my February post I wrote that Thomas Lounsbury was not (as far as I knew) related to the anthropological linguist Floyd Lounsbury (1914-1998), an expert on North American and Mesoamerican indigenous languages who taught at Yale several decades later. It turns out the two language-loving Lounsburys were indeed related, but rather distantly: they were third cousins thrice removed.

According to Bruce Bower, "Evolution's Ear", ScienceNews, 8/30/2008

In a new study, anthropologist John Hawks of the University of Wisconsin–Madison finds that eight hearing-related genes show signs of having evolved systematically in human populations over the past 40,000 years. Some alterations on these genes took root as recently as 2,000 to 3,000 years ago.

“Hawks makes a compelling case that not only is human evolution ongoing in the past 10,000 years, but it has sped up,” says anthropologist Clark Larsen of Ohio State University in Columbus.

Seven genes identified by Hawks produce proteins that make stereocilia and the membrane that coats them. The eighth gene assists in building middle ear structures that transmit sound frequencies to the inner ear.

As far as I can tell, this work hasn't come out in written form yet. John Hawks has a terrific weblog ("John Hawks weblog: Paleoanthropology, genetics, and evolution") and he mentions the Bower article in a brief note ("Hawks featured in Science News"), but all he says about it is "This is a really nice article, and I wasn't expecting it to come out, so please go read it!"  Which you should do.

The New York Times has a piece by Ellen Barry entitled Barriers that are steep and linguistic about linguistic aspects of the situation in Georgia, which quotes both me and Johanna Nichols, who unlike me is an authentic expert on the languages of the Caucasus. As newspaper articles go this is actually pretty good, but I thought it might be useful to fill in some of the details.

Nicholas Lemann ("Conflict of interests: Does the wrangling of interest groups corrupt politics — or constitute it?", The New Yorker, 8/11/2008) is promoting an Arthur Bentley revival:

In a year saturated with political conversation, can there be any topic that has not yet been discussed? Well, here’s one: 2008 is the centenary of a curious and mesmerizing book that was long considered the most important study of politics and society ever produced by an American—“The Process of Government: A Study of Social Pressures,” by Arthur Fisher Bentley. The reason its big anniversary hasn’t been celebrated is that “The Process of Government” is an ex-classic, now sunk into obscurity. The reason it should be celebrated is not just that it deserved its former place in the canon but also that it is uncannily relevant to this Presidential election. […]

The University of Chicago Press brought out “The Process of Government” in 1908, to almost no notice. In 1911, Bentley quit Chicago and newspapering and moved to the small town of Paoli, Indiana, where he remained until his death, in 1957. He produced a series of increasingly abstruse books (sample title: “Linguistic Analysis of Mathematics”), and his renown grew steadily. His closest intellectual companion was John Dewey—a published collection of their correspondence runs to more than seven hundred pages—but Bentley’s papers, at Indiana University, also contain letters sent to him over the years by, among many others, Albert Einstein, Thomas Mann, Sidney Hook, Estes Kefauver, and B. F. Skinner.

If any book is more obscure than Bentley's 1908 The Process of Government, it must be his 1932 Linguistic Analysis of Mathematics. At least, I'd never heard of it, and it's not cited in e.g. Paul Benacerraf & Hilary Putnam, Philosophy of Mathematics, 1984; or Dale Jacquette, Philosophy of Mathematics, 2002; or George Lakoff and Rafael Núñez, Where Mathematics Comes From, 2001; etc.  But reading Lemann's article made me wonder about it, and a quick internet search showed that Leonard Bloomfield reviewed this book, along with Bentley's 1935 Behavior, Knowledge, Fact, in Language 12(2): 137-141. (This issue is dated as April 1936 – June 1938, suggesting a more leisurely sense of time than journals generally have today.)

My parenthetical remark that Sir William Jones is incorrectly viewed as the discoverer of the Indo-European language family and founder of modern historical linguistics provoked the question in the comments of why I said this.

If your experience with an expression has been limited to a particular context, you're likely to assume that the meaning it has in that context is its "true", "real" meaning. If you then come across it in other contexts, you might well assume that its occurrences there are somehow connected to the uses you're familiar with. This belief might have nothing to do with the facts of linguistic history.

Case in point: a recent discussion of uses of the term non-dual (or nondual) on Jerry Katz's Nonduality Blog, which treats nondualism, "the understanding or belief that dualism or dichotomy are illusory phenomena" (as the Wikipedia article puts it). Katz explains:

I monitor in a non-scientific way the appearances of the words nondual, nonduality, nondualism, in blogs and the press. I don’t think there is any question that in the last ten years there has been a significant increase in the use of those words in the mainstream press. I also think there is no question that the increase in the awareness and usage of those terms has occurred within the field of spirituality in general.

But what I have realized is that the term non-dual is being used increasingly in ways unrelated to spirituality, philosophy, expressions of reality, or even science. I’m not going to speculate on what that means or whether what I’m seeing is real phenomenon. I’m only pointing it out.

I found all the following usages of non-dual in the last two months. It seems like an explosion in these findings. I wonder if it presages a greater spiritual explosion.

(Note the probable instance of the Recency Illusion, signaled by "increasingly".)

I'm puzzled. The reason is that I've just read Merja Kytö, "Be/have + past participle: The choice of the auxiliary with intrasitives from Late Middle to Modern English", pp. 17-86 in Matti Rissanen et al., Eds., English in Transition: Corpus-based Studies in Linguistic Variation and and Genre Styles, 1997.

The content of Kytö's chapter doesn't puzzle me — it explains very clearly how English changed from be to have as the marker of perfect aspect in intransitive verbs. This change is easy to see in bible translations, where for example in 1 Samuel 26:20, the King James Version of 1611 gives "the king of Israel is come out to seek a flea", where the 1978 New International Version of 1978 gives "[t]he king of Israel has come out to look for a flea".

And the timeline is also pretty clear. Based on tracking the use of be/have + past participle in a corpus of about 2.7 million words spanning the period from 1350 to 1990, Kytö demonstrates that "in the late Middle English period, the use of have increases gradually, gains in momentum in the late 1700s and supersedes the use of be in the early 1800s".

What puzzles me is why this process seems to have escaped the censure of prescriptive grammarians. Here's a change that "[gained] in momentum in the late 1700s", just when the likes of Robert Lowth and Lindley Murray were in bloom. Did anyone stand up against the rising tide of have for marking the perfect in intransitives? If so, their delaying action was ineffective and quickly forgotten.

As reported in the New York Times and Time Magazine, Yale law librarian and quotation-hunter extraordinaire Fred Shapiro has uncovered evidence undermining the long-held attribution of "The Serenity Prayer" to the American theologian Reinhold Niebuhr. Niebuhr's family originally claimed that he composed the prayer in the summer of 1943, but Shapiro has uncovered variations on the theme going back to 1936 in various American publications. (The first printed attribution to Niebuhr is actually from 1942.) Shapiro lays out his evidence in the Yale Alumni Magazine, followed by a rebuttal by Niebuhr's daughter Elisabeth Sifton.

What's particularly fascinating about Shapiro's documentary evidence is how the early citations all fit a general formula and yet show a divergence in phrasing reminiscent of the Telephone game. Regardless of how much claim her father ultimately has to originating the prayer, Sifton is correct to point out that "prayers are presented orally, circulate orally, and become famous orally long before they are put on paper." It's clear that by the time the prayer found its way into print in the '30s and '40s, the oral transmission of the meme was already well under way, as illustrated by the mutations it underwent in the retelling.

I'm in Paris for Acoustics 2008, and Edouard Geoffrois invited me to come a little early to attend the 10th annual Séminaire DGA/DET "Traitement de la parole, du langage et des documents multimédias" ("Processing of speech, language and multimedia"), held at the École Nationale Supérieure de Techniques Avancées (ENSTA). In this case, "attend" turned out to mean "give two talks at", and one of my assigned topics was "Human Language Technologies in the United States" (and yes, after a brief excuse in French, I'm ashamed to say that I gave the talk in English…)

For this survey, I decided to start with some historical background, and so I went back to the famous ALPAC report. This was a report to the National Academy of Sciences by the Automated Language Processing Advisory Committee, entitled "Language and Machines: Computers in Translation and Linguistics", and released in 1966.

I'm nearly sure that the people who submit comments on this and other weblogs are all intelligent, knowledgeable and thoughtful. Often, this is plain in what they write. But sometimes, in order to maintain my commitment to this belief, I'm driven to conclude that a commenter has recently tunneled through the barrier separating us from some parallel — and fairly distant — universe. Yesterday, one such transdimensional pilgrim contributed this observation:

Transformational grammar hasn't been accepted as plausible for decades, except, for some reason, in English departments.

Now, I know several of the inhabitants of several English departments in American colleges and universities. And if these organizations are full of partisans of transformational grammar, they're keeping a very low profile. In fact, in my (admittedly limited) experience, English departments are among the last places on campus where you're likely to find any indication of interest in any form of linguistic analysis whatever.

It's not only in the United States that linguists have failed in their responsibility to educate the public. As Geoff Pullum explained yesterday, the English Teachers Association of Queensland (Australia) recently published a teachers' guide to grammar that was "full of utter howlers". And some of the discussion of the controversy is not much better. For example, Graham Young wrote today in a blog post at the National Forum ("Grammar's taught to grammarians", 6/14/2008):

The Romans, driven I suspect by their infatuation with standardisation (which palls in comparison to ours, but they caught the disease first), invented grammar. It didn't exist before them, people just spoke languages.

This short passage contains several implicit indictments of my profession's educational failures.

Yesterday, I described Olga Feher's demonstration that species-typical songs emerge, over several generations, in an isolated colony of zebra finches founded by birds raised in isolation ("Creole birdsong", 5/9/1008). I compared this pattern to Derek Bickerton's "bioprogram" hypothesis, first put forward in his 1981 book Roots of Language, and discussed again in his 2008 book Bastard Tongues ("A Trail-Blazing Linguist Finds Clues to Our Common Humanity in the World's Lowliest Languages"). As the Wikipedia article on the "language bioprogram hypothesis" explains, Derek's idea is that

when the linguistic exposure of children in a community consists solely of a highly unstructured pidgin[,] these children use their innate language capacity to transform the pidgin, which characteristically has high syntactic variability, into a language with a highly structured grammar.

I also mentioned some of the subsequent debate over the bioprogram theory of creolization, quoting from an encyclopedia article by John Rickford and Barbara Grimes. Some of this debate has focused on whether the process of regularization in creole languages is complete in the first generation of native learners, or takes several generations. I observed that Bickerton's general idea ought to be consistent with a multi-generational emergence of a cognitive phenotype, where the species-typical pattern results from the accumulation of learning biases over several iterations.

However, some of Bickerton's critics have seen multi-generational creolization as evidence against his hypothesis. And to my surprise, it seems that he agrees with them. In an interesting comment on my post, he wrote:

Mark, you say that "Where social learning is involved, perhaps it's normal for the phenotype to emerge over multiple generations." And you may well be right, since social learning has nothing to do with creolization. How can you "socially learn" something for which you have no model, which didn't exist until you made it?