The Science of Language

A note from Bob Ladd:

Yesterday I received a complimentary copy of Intelligent Life, the Economist's foray into general magazine publishing.  One of the feature articles was entitled "The last days of the polymath?", with profiles of a few people who "know a lot about a lot" and ruminations on the age of specialisation.  The article includes a little box entitled "Living polymaths: who qualifies?", which lists about twenty people who were regarded as qualifying for that title in an informal office poll of staffers at the Economist and Intelligent Life.  The list includes a number of names that LL readers might have been expected to come up with, including Jared Diamond, Douglas Hofstadter, and Noam Chomsky (no Daniel Dennett, though).

For each person listed, the box shows their name, age, and nationality, as well as up to five "strings" (presumably as in "another string to his bow").  The strings are ordered by their importance in the individual's intellectual and career profile - so Hofstadter is listed as a mathematician first, then an aesthetic theorist, then an author.  Here, too, not a lot of surprises.

However, LL readers may be interested to learn of Chomsky's list of "strings".  He's listed first as a philosopher, second as a cognitive scientist, third as a political activist, and fourth as an author.  What, you might well ask, happened to "linguist"?  The fact that the Economist feels it appropriate to label Chomsky a philosopher is about as clear an indication as one could ask for that linguistics as a field remains invisible.

Elizabeth Daingerfield Zwicky writes to me, following up on my Maurice Sendak "half-sentence" posting (which I'll have more to say about in a while):

… if I knew how to encourage sentence fragments, I would go for that. Opal's sentences go on for*ev*er. And if I type them for her, and she's watching, and I try to put a period in, so there's a shorter sentence even though it starts with "And"? She says "No, that's not right, it's part of the same sentence. Didn't you hear the 'and'?" Fortunately she doesn't usually watch me type, allowing me to punctuate things as I see fit.

Two things here. First, Opal's attention to the conventions of writing, including her awareness of the stupid No Initial Coordinators advice about written English. Opal is 5, in kindergarten (which she started last month), and is writing on her own, but decidedly imperfectly (she is given, for example, to shifting to a new line when she comes to the edge of the paper, even if that's in the middle of a word), so I'm astonished that she even knows about NIC (and can refer, albeit indirectly, to it), much less cares so deeply about it. Where did she pick up this stuff? Certainly not from her family.

The other thing is Elizabeth's reference to "sentence fragments". Sentences with initial coordinators are not sentence fragments on that account (Elizabeth's sentence beginning "And if" is indeed a sentence fragment, but not because of the "And"). The problem is how to refer to sentences with initial coordinators in an unbiased fashion, without labeling them as a kind of error, as "sentence fragments". Yes, I know, sentence fragments are not in general ungrammatical, but the label has picked up an association with incorrectness. Similar problems arise in other contexts, for instance with regard to "dangling modifiers".

The usual labels are the ones that are widely known by the general public, but they are tainted by their association with error, and that makes it hard to talk about phenomena. In my own practice, I bounce back and forth between the usual labels and neutral, but unfamiliar, terminology, depending on the context.

In a comment on my post "Metapun", John S. Wilkins traced the phrase "missing link" back, conceptually if not literally,  to the  "great chain of being" metaphor featured in Alexander Pope's 1744 Essay on Man:

[…] On superior powers
Were we to press, inferior might on ours;
Or in the full creation leave a void,
Where, one step broken, the great scale's destroyed:
From Nature's chain whatever link you like,
Tenth, or ten thousandth, breaks the chain alike.

Although Pope didn't use the specific word-sequence "missing link" — nor the word missing at all –  the concept is implicit in the idea of the "link" that "breaks the chain".  This morning, Victor Steinbok sent in the result of some textual sleuthing, which suggests that the collocation "missing link" was in the air of  1860 Britain with reference to links in other sorts of metaphorical chains,  neither Darwin's chain of "descent with modification" nor Pope's chain of "full creation".  Victor also turns up an example, from a widely-read work originally published in 1844, that uses "missing link" in the context of patterns of taxonomic similarity in biology.

Here's Victor's note:

You probably know that I am little more than an amateur when it comes to word sleuthing. On the other hand, unlike high-tech enterprises, it is a field that is somewhat accessible to amateurs with an inclination to research (and I have plenty of research training in rather diverse disciplines) and a desire for accuracy.

Lacking affiliation and easy research library access, at the moment, this would be a complete waste of time without Google Book Search –and these are, by all accounts, notoriously unreliable in many respects. The greatest part that is missing is context–it is often hard to place the volume, even correctly identified and dated, within the context of its contemporaneous publications. In a different field, this would be small bother–one can simply look up more books on the same subject and get a rather exhaustive picture. Not so here.

So, with this preface, I want to attempt to dig further into the "missing link"–the subject broached earlier on Language Log and on Visual Thesaurus (hence the two recipients of this email). I find it interesting that the OED entry for the "missing link" has an explicit evolutionary component:

1. Something lacking to complete a series or to form an intermediate between two things, esp. in an evolutionary process; (Anthropol.) a hypothetical animal assumed to be an evolutionary link between man and the anthropoid apes, esp. as sought by early evolutionary biologists.

The second definition is simply a variation on the first. Given the list of citations, this is not surprising. All are dealing with hominid species supposedly bridging the gap between large apes and humans. To these there should be added at least one entry from 2009–there have been quite a few pronouncing the end of the "missing link theory" with the arrival of "Ardi". In fact, similar claims have been made repeatedly with the latest finds of early hominids throughout the decade.

But this is clearly not a sufficient reason to get excited. Adding recent citations does not involve any antedating.

Now, returning to the source of the definition–largely citations of texts related to evolution. Judging from this list, it would be easy to overlook the fact that "missing link" was quite a popular expression, say, in London, circa 1860. And it had nothing to do with evolution. In fact, GoogleBooks does provide evidence of other sources–for one, a book published in 1859 in London by that very title. The exact name of the volume was "The missing link: or, Bible-women in the homes of the London poor, by L.N.R." It did not take long for the literatti to figure out that "L.N." stood for "Ellen", so that the book had actually been written by Ellen Henrietta Ranyard.

The subject of the book was, predictably, "Bible-women"–actually, women who went to the London tenements, Bible in hand, to return the masses to religion. The book appears to have been quite a hit–so much so that the London Quarterly Review turned it into the lead for the July 1860 issue. And it crossed the Atlantic as well–the New Englander and Yale Review gave it a paragraph (p. 274) in its February 1860 issue, and returned to it again (p. 1113f) in its November 1860 issue as part of the review of Ranyard's other book (both mentioned in Vol. XVIII).

The Book and its Story–The writer of this book published some time ago a little volume, to which she gave the somewhat singular title, "The Missing Link." We furnished some account of it on page 274 of the present volume. It was a simple narrative of her efforts in sending female colporteurs, or "Bible women," among that wretched class of people in London, who swarm in "tenement houses" in such places as "The Seven Dials." The success which these female colporteurs met with in circulating the Bible was suh that her account of it has been rapidly but quietly working its way into public favor and notice; and in July last, the London Quarterly made "The Missing Link" the basis of its leading Article.

To say the least, London, specifically, and English readers, more broadly, had been well primed by this little volume. With that in mind, Hopkins's line

But then, where are the missing links in the chain of of intellectual and moral being?

may well take on a somewhat different meaning. Although the citation is sandwiched by two Lyell quotations that have an undoubtedly evolutionary meaning, the two are actually quite removed from each other. Meanwhile, Hopkins wrote his article in Frasier's precisely around the time when the book was making rounds, explaining how the women were providing the "missing link" in the moral development of the wretched classes.

My proposition is a mere speculation, but, I don't believe, we can simply ignore the context here. Of course, there are two more things to consider. First, Lyell's 1851 citation is still earlier than Ranard's book. So there is no primacy claim. Rather, I am suggesting that the use in the anti-evolutionary circles might have arisen from the blend of the two tomes published, quite coincidentally, in the same year.

Second–and this is more important–the use of the expression in the title suggest that it was not quite as "singular" as the Yale Review suggests. In fact, it may well have been a play on a more familiar use and that surely was not the one exemplified by Lyell. Note, for example another 1860 periodical (November) using the phrase in a way similar to Ranyard's:

The Arabs of the street and of the city are being gathered into ragged schools, the social evil is being grappled with at midnight, "the missing link" of woman's gentle hand is now bringing up from the dregs of society into the genial influences of regenerating love and truth the most hideous shapes of lost humanity.

Here, GoogleBooks fails. This is precisely the limitation that I mentioned earlier–simply having access to a number of scanned books cannot provide a live context. Yet, it does offer a glimpse, even though most of the sources in question are not actually available.

Obviousness is a term of art in patent law that describes a type of invention that would have been inevitably discovered through regular use and adaptation of existing patents and, therefore, does not allow for a patent of its own. Certainly, the combination "missing link" is obvious in this sense. (See, for example, The Head of the Family, by Dinah Maria Craik, 1852, p. 67)

By far the more interesting use of "the missing link" is in genealogy. In particular, WorldCat has a pointer to an 1810 title My ancestors to the missing link, or, The family tree of Pierson Worrall Banning (Chicago). There is a GoogleBooks entry, but it lacks the text. Yet, the meaning within the title appears to be rather similar to that used later in evolutionary arguments.

The link between genealogical use and evolutionary use of the expression should by no means be ignored! Widener (Harvard) holds another 1860 London volume that provides an interesting text. (p. 51)

The internal structure of the stem, and the character of the seed-vessels, show them to have been a link between single-lobed and double-lobed plants–a fact worthy of note, as it favours the idea of a progress in vegetable creation, in the line of an improved organization. It is also curious to find a missing link of so much importance in a genus of plants which has long ceased to have a living place upon earth.

The volume is the 11th edition (!) of Vestiges of the Natural History of Creation [apparently] by Robert Chambers. Note, in particular, that the text does not refer merely to a "link" but to a "missing link". The text also appears to have been adopted from earlier editions, as additional materials are [bracketed] (e.g., on p. 54). The text cites Murchison and Agassiz rather heavily. (There is a citation to Lyell's Elements of Geology on p. 88 and a reference to Darwin's Journals on p. 81.) It is interesting that when the author gets to "the origins of the animated tribes", the text turns to religion. Evolution is only brought up in the appendix ("Proofs, Illustrations, Authorities, etc."). But the same "missing link" phrase is found on p. 62 of the 1845 2nd New York edition, which claims to be "from the third London edition, greatly amended by the author".

This makes perfect sense–it is highly unlikely that the phrase "the missing link" would have been adapted as satire on evolutionary claims had its normal use not been already well established in paleobotany or a related discipline. The connection to genealogy makes perfect sense.

I've done little more than some preliminary work, in an attempt to trace the possible formation and evolution of the expression prior to its use in the debates on evolution. I really don't know if anyone else will care for extending the origin of the phrase back anywhere from 6 to 50+ years, even though the meaning is not precisely the same. In fact, there appear to be two meanings — one is the genealogical one, referring to some unidentified, near-mythical ancestor, the other the connection between the Great Unwashed and the "intellectual and moral development", i.e., religion and/or the Bible. Both of these appear to have merged in the debate on evolution, giving it the necessary ironic twist.

[Above is a guest post by Victor Steinbok.]

[(myl): Here's a description of Vestiges of the Natural History of Creation, from the publisher's blurb for a  recent reprint:

Originally published anonymously in 1844, Vestiges proved to be as controversial as its author expected. Integrating research in the burgeoning sciences of anthropology, geology, astronomy, biology, economics, and chemistry, it was the first attempt to connect the natural sciences to a history of creation. The author, whose identity was not revealed until 1884, was Robert Chambers, a leading Scottish writer and publisher. Vestiges reached a huge popular audience and was widely read by the social and intellectual elite. It sparked debate about natural law, setting the stage for the controversy over Darwin's Origin. In response to the surrounding debate and criticism, Chambers published Explanations: A Sequel, in which he offered a reasoned defense of his ideas about natural law, castigating what he saw as the narrowness of specialist science. With a new introduction by James Secord, a bibliography of reviews, and a new index, this volume adds to Vestiges and Explanations Chambers's earliest works on cosmology, an essay on Darwin, and an autobiographical essay, raising important issues about the changing meanings of popular science and religion and the rise of secular ideologies in Western culture.

This reprint, whose version of Vestiges is a facsimile of the 1844 original, has the same passage ("It is also curious to find a missing link of so much importance in a genus of plants which has long ceased to have a living place upon earth").  This seems to indicate that a taxonomic meaning of "missing link", indeterminate between creationist and evolutionary interpretations, was well established before the publication of Darwin's work in 1859.]

In response to my recent posts "The United States as a subject" and "When did the Supreme Court make us an 'is'?", Rich Rostrom sent the following essay, reflecting some research he did a few years ago.

One popular comment on the Civil War is that "before the war, people said 'The United States are…', whereas after the war, they said 'The United States is…". That is, the common idea of 'the United States' changed from plural to singular - from a collection of states to a nation.

This comment was given much play in the Ken Burns PBS series a few years ago.

However, some have questioned it, as being rather too neat, and not actually supported by evidence.

I decided to make an inquiry, by examining a number of texts from the periods before and after the war, to see if there was a change in usage. I went to the Avalon Collection at Yale University for my sources.

The documents in the Avalon Collection include the Inaugural Addresses of the Presidents, Annual Messages of the early Presidents, and many of the treaties made by the United States.

I searched through these documents for all usages of the phrase "United States", and noted from context whether the phrase was a plural or singular. This could be determined by the form of an associated verb: "are"/ "is", of course, but also "have"/"has", or "do"/"does". An associated possesive is also indicative: "their"/"its".

All the indicative usages I found are quoted below.

The phrase did not appear as often as one might imagine - it is missing entirely from, for instance, some of the inaugural addresses.

In many other cases, it appears as a part of a longer phrase, such as "President of the United States" or "Constitution of the United States", or in a manner that does not distinguish singular from plural. For instance in the phrase 'the United States will…', 'United States' could be either singular or plural. It seems that some authors preferred to avoid any indicative usage.

Of all the documents I looked at, only 20 had an indicative usage. Some had more than one. Monroe's two inaugurals had nine plural usages. He seems to have been more decided on the point than anyone else.

The first singular usage appears in Jackson's Proclamation Regarding Nullification of 1832; there is also a plural usage there. The next singular usage is in the Gadsden Purchase Treaty of 1853, where there are three such (and no plurals); a big change from the Treaty of Guadalupe Hidalgo of 1848, where there are five plurals.

The next item (in chronological order) is Seward's treaty with Russia for the purchase of Alaska, in which "the United States agree to pay…" - after the War, but still plural.

I mentioned the various treaties. Most of these are routine treaties for commerce, navigation, trade-marks, naturalization, and so on. The boiler plate at the beginning of some of these lists the persons concerned in the making of the treaty: the President, the American negotiator, the foreign minister of the other power, and the other state or monarch. In several of these treaties, the boiler plate begins "The President of the United States of America, So-and-So, their Envoy…" etc.  That is, the envoy or minister of the United States in plural. This usage appears as late as an 1896 treaty with Argentina.   Some of the treaties had no indicative usage; none had singular usage.

The last four documents are the inaugural addresses of McKInley, Taft, and Hoover. All references in them are singular.

So while it is true the change in question took place, it is not at all clear that it took place at the time of the Civil War.

Here are the quoted usages.

Fourth Annual Message of George Washington, November 6, 1792 …

on the part of the United States or their citizens…

First Annual Message of John Adams, November 22, 1797

… vessels and merchandise taken within the limits and jurisdiction of the United States and brought into their ports…

Third Annual Message of John Adams, December 3, 1799

…the engagements contracted by the United States in their treaties with His Britannic Majesty…

First Inaugural Address of James Monroe, MARCH 4, 1817

During a period fraught with difficulties and marked by very extraordinary events the United States have flourished beyond example.

Situated within the temperate zone, and extending through many degrees of latitude along the Atlantic, the United States enjoy all the varieties of climate, and every production incident to that portion of the globe.

With such an organization of such a people the United States have nothing to dread from foreign invasion.

It is particularly gratifying to me to enter on the discharge of these duties at a time when the United States are blessed with peace.

Second Inaugural Address of James Monroe, MARCH 5, 1821

… the United States… will always have it in their power to adopt such measures … as their honor and interest may require.

the right claimed by the United States for their citizens to take and cure fish …

The great interests which the United States have in the Pacific…

The situation of the United States in regard to their resources, the extent of their revenue…

The United States now enjoy the complete and uninterrupted sovereignty over the whole territory…

President Jackson's Proclamation Regarding Nullification, December 10, 1832

… the laws of the United States, its Constitution…

To say that… is to say that the United States are not a nation.

Treaty of Peace, Friendship, Navigation and Commerce Between the United States and Venezuela; May 31, 1836

… whatever… can be… lawfully imported into the United States in their own vessels…

Morocco - Treaty of Peace; September 16, 1836

If any of the citizens of the United States, or any persons under their protection…

Treaty with Hanover of Commerce and Navigation; June 10, 1846

… whatever… can be… lawfully imported into the United States in their own vessels…

… the growth, produce, and manufacture of the United States, and of their fisheries

The United States agree to extend…

Treaty of Guadalupe Hidalgo; February 2, 1848

… those invasions which the United States have solemnly obliged themselves to restrain.

The United States engage, moreover, to assume and pay to the claimants…

The United States do furthermore discharge the Mexican Republic…

The United States, exonerating Mexico from… the claims of their citizens… undertake to make satisfaction for the same…

Gadsden Purchase Treaty : December 30, 1853

The United States, by its agents, shall have the right…

… the effects of the United States government and its citizens…

… the United States may extend its protection as it shall judge wise to it when it may feel sanctioned…

Treaty concerning the Cession of the Russian Possessions in North America …to the United States of America : June 20, 1867

In consideration of the cession aforesaid, the United States agree to pay…

Convention Concerning the Rights, Privileges, and Immunities of Consuls: December 5, 1868

Henry Shelton Sanford, a citizen of the United States, their Minister Resident near His Majesty the King of the Belgians

Trade-Mark Convention Between the United States and Austria-Hungary; November 25, 1871

The President of the United States of America, John Jay, their Envoy Extraordinary and Minister Plenipotentiary from the United States of America to His Imperial and Royal Apostolic Majesty…

Extradition Convention Between the United States and Argentina; September 26, 1896.

The President of the United States of America, William I. Buchanan, their Envoy Extraordinary and Minister Plenipotentiary, to the Argentine Republic…

First Inaugural Address of William McKinley, MARCH 4, 1897

The United States has progressed with marvelous rapidity…

Second Inaugural Address of William McKinley, MARCH 4, 1901

… the … policy of the United States in its relation to Cuba.

Inaugural Address of William Howard Taft, MARCH 4, 1909

… the mainland of the United States and in its dependencies.

… the United States can maintain her interests intact…

The policy of the United States in the Spanish war and since has given it a position of influence…

Inaugural Address of Herbert Hoover, MARCH 4, 1929

The United States fully accepts the profound truth…

The United States seeks by these reservations…

[Above is a guest post by Rich Rostrom.]

Grace Wu sent me a photograph taken at Taipei Storyland, shown at the right (click on the image for a larger version).

The characters running down the right side of the picture read as follows:

WO3 YAO4 SHUO1 GUO2YU3, BU4 SHUO1 FANG1YAN2
"I want to speak the national language, not the topolects."

In other words, "Let's speak Mandarin, not Taiwanese, Hakka, Cantonese, etc."

This injunction to speak Mandarin at the expense of the regional Sinitic languages ties in with numerous Language Log posts, such as Arnold Zwicky's "How many ethnic groups?" and my own on the "Mutual Intelligibility of Sinitic Languages."

Moreover, it also comports with the Zeitgeist reflected in several recent articles in the South China Morning Post.  The first (October 6, 2009, p. A6), entitled "Cantonese almost became the official language," is by He Huifeng:

Putonghua is the official language on the mainland, but if history had played out differently the vast majority could have been speaking Cantonese.

In 1912, shortly after the fall of the Qing dynasty, the founding fathers of the republic met to decide which language should be spoken in the new China.

Mandarin - now known as Putonghua [the common language] - was then a northern dialect spoken by the hated Manchurian officials. While it had served as China's lingua franca for centuries, many perceived it as an "impure form" of Chinese.

Many of the revolutionary leaders, including Sun Yat-sen, were from Guangdong - which has long been China's land of new ideas. A great debate started between the delegates and eventually led to a formal vote. Cantonese lost out by a small margin to Putonghua and the rest is history.

While historians today still argue about the authenticity of the story, it is something Guangdong people love to tell. Many Cantonese speakers feel proud of their native language, saying it has more in common with ancient classical Chinese than Putonghua - which is a mix of northern dialects heavily influenced by Manchurian and Mongolian.

Linguists agree to some extent. "Cantonese is closer to classical Chinese in its pronunciation and some grammar," Jiang Wenxian , a Chinese language scholar, said. "Using Cantonese to read classical poetry is a real pleasure," he said. "Many ancient poems don't rhyme when you read them in Putonghua, but they do in Cantonese.

"Cantonese retains a flavour of archaic and ancient Chinese. Nowadays few people understand classical Chinese, so Cantonese should be protected as a type of language fossil helping us study ancient Chinese culture."

Cantonese is spoken by about 70 million people in Guangdong, Hong Kong, Macau and communities abroad.

In the 17th and 18th centuries, Guangdong was the only Chinese province allowed to trade directly with foreigners. Many Westerners at the time learned Cantonese. Up till very recently, there were more Cantonese speakers in overseas Chinese communities than Putonghua speakers. In Canada, for instance, Cantonese is the third most commonly spoken language after English and French.

On October 11, Chloe Lai published a long article entitled "Linguistic heritage in peril:  A group of friends are hoping to revive Cantonese in Guangzhou," from which I here give brief excerpts:

"Speak Putonghua, write standardised characters, use civilised language, be a civilised person." The words are printed on a red banner hanging in the main entrance of a primary school in Guangzhou, a city that once set the standard for the Cantonese-speaking community.

"It is a common practice; many schools are doing the same," said Yao Cheuk, an artists' agent in the city. "They are doing this because it is national policy to promote Putonghua. From time to time, there is news that kids got punished for speaking Cantonese in schools. It is outrageous. They are eliminating Cantonese."

Angry about the official bias, Yao went on to explain the superiority of Cantonese, which he described as a more mature language with a richer linguistic history than Putonghua. He cited soccer player David Beckham's name to illustrate his arguments. Cantonese translates his family name using two characters, while Putonghua uses four.

"You know why?" asked Yao. "Because Cantonese is an ancient language that has a rich phonetic system, it takes only one character in Cantonese to mimic the English sound 'ham', whereas it takes Putonghua two Chinese characters."

He pointed out that Putonghua has only 23 vowel sounds, while Cantonese has 59, leaving Putonghua relying heavily on the context for meaning.

Yao's friend, surnamed Pang, stressed they were not anti-Putonghua. "Language is for people to communicate. I speak Putonghua whenever there are people whose native tongue is not Cantonese," the college student said. "Kids will do the same when they need to communicate with their friends. Why force us to abandon our native language?"

Both insist on using Cantonese pronunciations to spell their names in English. Pang and Yao are among a group of Guangzhou natives who fear for the future of Cantonese in the capital of Guangdong. Their worries are not without basis. For example, more than 80 per cent of cabbies do not speak Cantonese and often drivers will suggest that Cantonese speakers use Putonghua for directions….

Pang, Yao and their friends believe Putonghua speakers in Guangzhou already outnumber Cantonese speakers, because of the influx of migrants from other parts of China and the national policy of promoting Putonghua. The trend, they say, will continue, leading eventually to the extinction of Cantonese in Guangzhou….

Cantonese is regarded as a modern variation of the ancient Han language, said Roxana Fung, an assistant professor at Polytechnic University's department of Chinese and bilingual studies. The Cantonese system - pronunciations, vocabulary and usage - is very similar to the official language used during the Tang dynasty (618-907)….

Professor Fung does not want to see Cantonese eliminated.

"Dialects are language fossils, they keep many characteristics of the ancient language. Through dialects, we can understand many ancient scripts," she said.

All is certainly not yet lost, since among certain sectors of youth, Cantonese is experiencing a resurgence.  In another SCMP article by He Huifeng, entitled "Trendy Shenzhen teenagers spearhead Cantonese revival" (updated on October 6, 2009), we read:

A new craze is sweeping through the ranks of Shenzhen's teenagers. Whether it is in school, at the shopping mall or the KTV club, there's only one way to prove you are a real "Shenzhener" - by speaking Cantonese. In the past couple of years, there has been growing concern that regional dialects are being lost to the relentless tide of Putonghua. But in Shenzhen, many immigrants are swimming against the current. Li Zhen is a 16-year-old high school student who was born in Wuhan and moved to Shenzhen at the age of 10. She insists on talking to her friends in Cantonese. "My parents do not speak Cantonese and we speak Putonghua or Wuhan dialect at home," Li said. "But in school, we only speak Putonghua in class. All my friends are Cantonese speakers. Cantonese is the fashionable language among Shenzhen teenagers." Li's friend, Wang Zijing, said speaking Cantonese made them feel more international. "Being bilingual, we feel we have more in common with international cities such as Shanghai, Hong Kong or New York than with people from the hinterland who can usually only speak Putonghua," she said….

There is a two-pronged attack on the local language - internal migration on the one hand, and central government policies of a "common language for a unified country and harmonious society" on the other. In the 1980s, the universal adoption of Putonghua was enshrined in the constitution and in all schools from kindergartens up. In the 1990s, local dialects were even banned in many provincial and state-controlled television stations….

But things have taken an interesting turn in the past decade. As the second generation of migrants grows up, they are embracing Cantonese culture and language. "We feel no different from Cantonese natives," Li said."We speak Cantonese with no accent. We watch Hong Kong television dramas. We enjoy Cantonese cuisine such as herbal tea and fish balls. We sing old Cantonese songs at KTV. But actually, we are Shenzheners, or new Cantonese."

…In Guangdong, while the official policy of promoting Putonghua over local dialects remains unchanged, officials are increasingly putting emphasis on developing "Lingnan [Cantonese] culture". The reasons behind this go beyond pride in the indigenous culture. Guangdong leaders - many of them migrants from other provinces - are starting to realise the role of culture in social and economic development. There have been concerns that the Pearl River Delta is falling behind the Yangtze Delta in attracting skilled workers and top talents, all because of a perceived weaker cultural environment. In response, Guangdong invests billions of yuan in cultural development each year. The result is a wave of cultural propaganda showing off ancient Cantonese culture.

But to return to the anti-topolect slogan in Taipei Storyland with which I began, the National Language Movement (GUO2YU3 YUN4DONG4 國語運動) has a history going back even before the beginning of the Republican Period in 1912.  When the Guomindang (Kuomintang / KMT) was defeated by the Communists on the mainland and retreated to Taiwan, they brought Mandarin with them and promoted it rigorously.  In the 50s and later, it was illegal to speak Taiwanese in schools, universities, and other public places.  During this period, one could even be put in jail for compiling Taiwanese language teaching materials (I know someone who was incarcerated for having done so).

Under President Chen Shui-bian, Taiwanese, Hakka, and even the aboriginal (Austronesian) languages experienced a strong revival.  Now, however, Chen languishes in prison under a life sentence, the KMT is back in power, and Mandarin is being promoted vigorously.  Hence the slogan pasted on the window frame in Taipei Storyland.

In a paper entitled "How to Forget Your Mother Tongue and Remember Your National Language" that I published on the Web,  I attempt to put the dialectical dance of the "dialects" with Mandarin in historical and linguistic context.

[Hat tip to June Teufel Dreyer.]

Stephen Colbert on Olympia Snowe (Colbert Report, Oct. 14):

We are now one step closer to a nightmare future where everyone has health insurance. And I will tell you who I blame: Maine Senator Olympia Snowe, the only Republican who voted in favor of the bill. And folks, I am angrier than an Eskimo… because I have 300 words for Snowe, and I can't say one of them on TV.

(Hat tip: Greg Howard.) The title for this post is lifted from the Twitter feed of Michael Covarrubias (aka Wishydig):

Susan Collins, R-Maine has hinted at a 'yes' vote. and only linguabloggers have "snowe-clone" in their repertoire of bad puns.

When I tried to read Dilbert this morning, comics.com showed me this instead:

I reckon that the sequence was:

FIrst, discussions of evolution focused on the concept of a "missing link" in the fossil record. According to the OED, this actually began with with a remark about purely geological history:

1851 C. LYELL Elem. Geol. xvii. 220 A break in the chain implying no doubt many missing links in the series of geological monuments which we may some day be able to supply. 1862Caledonian Mercury 11 Jan. 7/6 Until the existence of some animal was discovered which should supply the missing link between man and the gorilla, there was a great gap even in Mr Darwin's theory of the origin of species.

This phrase quickly took on various figurative uses:

1862 G. DU MAURIER Let. Oct. in Young G. du Maurier (1951) 178, I..said that if he would take the trouble to make a post mortem on the Irish roughs I intend to kill next Sunday in the Park, he might convince himself that the ‘missing link’ had been found.  1863 G. O. TREVELYAN Competition Wallah (1864) v. 113 The performances of these thin-legged, miserable, rice-fed ‘missing links’ are perfectly inexplicable according to our notions of muscular development. 1904 ‘O. HENRY’ Cabbages & Kings 22 The faces of missing links. 1990 in J. E. Lighter Hist. Dict. Amer. Slang (1997) II. 561/2 No date with the missing link tonight?

(Ben Zimmer discusses the history of the expression in much more detail here.)

Independently , there developed a series of illustrations of evolutionary progress, showing a procession of successively "more evolved" creatures  becoming taller and more erect. (See here for some discussion, with a couple of examples of the many visual puns that have been overlaid on it, and a reference to Stephen Jay Gould's Iconography of an Expectation, which gives a dozen or so more.)

Recently, Dave Whamond created another such visual pun, bringing in an imaginary photographer (in an enlightenment-era wig and lace collar? who is he supposed to be?), and adding the joke that someone named "Link" should have been in the procession, but is "missing". This is a verbal pun on "missing link", and a visual pun on the stereotyped evolutionary-progress image.

FInally, comics.com added another layer of pun, in which "link" is not the evolutionary "missing link", nor the guy named Link missing from the photographer's posed picture, but the "404 Not Found" missing hyper-link to the Dilbert strip that I was hoping to read.

Off hand, I can't think of any similarly dense meta-puns, but no doubt LL readers will be able to fill the gap.

There are many ways NOT to learn Chinese characters, but one that I just found out about today is probably the worst, even worse than T. K. Ann's Cracking the Chinese Puzzles.  It was written by Alison Matthews ("a statistician who has worked in the oil, aviation, tourism, medical and software industries") and Laurence Matthews (author of books that claim to help you find Chinese characters fast) and is called Learning Chinese Characters:  A revolutionary new way to learn and remember the 800 most basic Chinese characters.

You can find the Matthews' miraculous tome on Google books here.

If you start leafing through the book, as I did, you will find on any given page hilarious explanations such as the following:

"tree 木 + several 几 = machine 机":  It took several trees to provide enough wood to make the parts for the huge machine.           

This is accompanied by a picture of a wooden contraption behind which are four trunks of trees that have been felled and beyond that three trees that are still standing.  Whereas the Matthewses gushingly enjoin us to "see how the 'several trees' have indeed been felled to make the large 'machine' that is taking shape," this is actually a pictophonetic (or semantosyllabic) character in which MU4 is the semantic indicator and JI1 (not JI3 ["several") is the phonophore.

"wrap  勹 + a drop 丶 = ladle 勺":  When he had wrapped it up he put a drop of perfume on the package even though there was only a ladle inside.        

This tortuous explanation is accompanied by a picture of two hands in front of a belly; the right one is holding a perfume dropper out of which has come a drop of perfume that is wiggling in mid-air, while the left hand is holding the handle of a ladle that is wrapped in cloth or paper and tied with a string.  The ladle appears to be resting on a flat surface, or possibly partially submerged underwater.  In fact, this character goes back to the period of the oracle bones (earliest stage of the writing system, circa 1200 BC), at which time it depicted a ladle with a drop of liquid in it.  At the time of its creation, the character had absolutely nothing to do with "wrapping"; the idea of "wrapping" is an artifact of a later stage of evolution when the ladle was transformed into what became Kangxi radical 20 勹.

These are just the first two characters that I happened to turn to as I perused the book.  Looking further, I find that there are even more outlandish explanations for many other characters.  If one tried to use this method to learn 8 characters, it might work, but if one attempted to learn all of the 800 characters in the book this way, it would be a horribly frustrating experience.  If one did not go insane in the process, at the very least one would have lost hundreds, if not thousands, of hours in vain hopes of mastering the strokes, sounds, and meanings of so many characters in this absurd fashion.  And, if one should ever be so foolhardy as to try to employ the method of the Matthewses to learn 8,000 characters, one would certainly go stark, raving mad.

I pity anyone into whose hands this book falls and who actually tries to learn the characters by using it.  The only thing it will do for you is turn your brain into mush.

Hat tip to Jonathan Smith

In the celebrated libel case brought by the British Chiropractic Association against Simon Singh, Singh has won a round in court. Or rather, he's won the right to appeal a previous loss in court.

Last May, Sir David Eady ruled that a passage in Singh's article "Beware the spinal trap" was a statement of fact rather than a comment, and also that when Singh wrote that  the BCA "happily promotes bogus treatments", he accused the BCA of conscious dishonesty, not just of, well, promoting bogus treatments, i.e. treatments that don't work.

Addressing Mr Justice Eady's previous judgement in the case, Lord Justice Laws said Eady had arguably risked swinging the balance of rights too far in favour of the right to reputation and against the right to free expression. Lord Justice Laws said Eady's judgement, centred on Singh's use of the word "bogus" in an article published by the Guardian newspaper, could be seen as "legally erroneous".

So now Singh gets to appeal Eady's rulings.

In the latest New Yorker (October 12), Tad Friend takes us into the chilling wonderworld of entertainment-business reporting, in a Letter from California, "Call Me: Why Hollywood fears Nikki Finke" (Finke runs the website Deadline Hollywood Daily). Apparently real life in the entertainment business in Hollywood goes beyond the parodies in movies and television shows.

A linguistic point about the business (p. 99):

Hollywood's leaders work with the understanding that facts are not fixed pillars but trial balloons that you inflate with the gas of vehement assertion. The truth is always negotiable.

There then follows a convoluted story (one of many in Friend's piece), about who said "We are no longer doing movies with women in the lead" and in what circumstances.

I'm reminded of the way facts and truth are treated in some other domains, politics and advertising in particular.

And "scientific management", treated by the excellent Jill Lepore in a piece ("Not So Fast") later in the issue, on the history of Taylorism.

Some of my Language Log colleagues are too modest for their own good, neglecting to mention here relevant things they've published or blogged on in other places.

A little while ago, I learned that Dan Jurafsky has a cool Language of Food blog (here), an outgrowth of a Stanford Introductory Seminar he's taught a few times. I found out about his blog only because, knowing his interest in the topic, I sent him a link to a recent posting on my own blog about nouns denoting food or drink being usable, metonymically, to refer to events ("After pizza, we watched a movie"), and he told me about his LoF blog.

It's a rich domain: vocabulary (in many languages) for food and cooking, including both everyday vocabulary and (semi-)technical vocabulary, cultural forms in these areas (when do we eat? what do we eat then? who does this? how much variation is there? how do we prepare this food? who prepares it? what cultural values do we assign to these foods?), and much else. Check it out.

Scott Timberg, "Maurice Sendak rewrote the rules with 'Wild Things' " (Los Angeles Times, October 11):

In "Wild Things," a single sentence can take pages to unfold, its meaning changing slightly with each image. And this book with numerous wordless pages ends with a half-sentence and no accompanying image. Sendak works similarly to the directors of the French New Wave, who used jump cuts and other techniques to dislocate their editing. (link)

Apparently this half-sentence has a dislocating effect. But what is this dislocating half-sentence? This, (1):

and it was still hot.

(it refers to Max's supper, still waiting for him on his return from his adventures among the wild things).

Suppose this last page had been this, (2):

And it was still hot.

(with a period at the end of the immediately preceding text). Then we'd have a sentence with an initial coordinator. Such sentences are fiercely reviled in some circles, on the grounds that they are not complete sentences but only sentence fragments. But no reputable writer on usage shares this prejudice. Here's Mark Liberman on No Initial Coordinators:

There is nothing in the grammar of the English language to support a prescription against starting a sentence with and or but — nothing in the norms of speaking and nothing in the usage of the best writers over the entire history of the literary language. Like all languages, English is full of mechanisms to promote coherence by linking a sentence with its discourse context, and on any sensible evaluation, this is a Good Thing. Whoever invented the rule against sentence-intitial and and but, with its a preposterous justification in terms of an alleged defect in sentential "completeness", must have had a tin ear and a dull mind. Nevertheless, this stupid made-up rule has infected the culture so thoroughly that 60% of the AHD's (sensible and well-educated) usage panel accepts it to some degree.

And here I am, following up on Mark:

Mark notes that the AHD note for and rejects NIC out of hand, and he provides a smorgasbord of cites (and statistics) from reputable authors.  Similarly MWDEU.  Paul Brians, collector of common errors in English, labels sentence-initial coordinators a "non-error".  Bryan Garner denies, all over the place, that NIC has any validity.  Even the curmudgeonly Robert Hartwell Fiske tells his readers that there's absolutely nothing wrong with sentence-initial coordinators.  A point of usage and style on which Liberman and I and the AHD and the MWDEU stand together with Brians and Garner and Fiske (and dozens of other advice writers) is, truly, not a disputed point.  NIC is crap.

But still NIC lives on in the popular mind. Presumably Timberg was treating (1) as (2) and criticizing it as a NIC violation (while, perhaps slyly, committing a NIC violation with "And this book with numerous wordless pages ends with a half-sentence …"). (Thanks to Phil Resnik for pointing me to the L.A. Times piece and noting Timberg's NIC violation.)

However, Sendak didn't write (2). He wrote (1), which isn't punctuated as a separate sentence. Here's what Sendak wrote as the final sentence of Wild Things, with line divisions as in the original; the sentence is spread across four pages, with some wordless pages intervening:

The wild things roared their terrible roars and gnashed their terrible teeth
and rolled their terrible eyes and showed their terrible claws
but Max stepped into his private boat and waved good-bye

and sailed back over a year
and in and out of weeks
and through a day

and into the night of his very own room
where he found his supper waiting for him

and it was still hot.

(You'll see that Sendak isn't fond of commas.)

This sentence is a coordination of three clauses: the first in lines 1 and 2, the second (introduced by the coordinator but) in lines 3 through 8, the third (introduced by the coordinator and) in line 9, which is our old friend (1). So, yes, (1) isn't a full sentence, just one clause of a sentence, but there's nothing grammatically wrong with it.

Maybe Timberg expects every page in a children's book to stand on its own as a text, and it looks like most (but not all) children's books are arranged that way, but it's not the way Sendak (with his few words and many images) works. In fact, not one of the four pages above stands on its own as a text (granted, the first is missing only a period).

Back in June of 2002, one of William Safire's On Language columns began this way:

'The South Carolina primary between Mr. Bush and Mr. McCain in 2000," wrote Eleanor Randolph, the New York Times editorialist, referring to Representative Lindsey Graham's current campaign for the Senate, "left Republicans in his state bitter and divided. That said, both President Bush and Senator McCain have already campaigned for his election to the Senate."

In olden times, those two sentences would have been written as one, with the first clause subordinated: "Although the South Carolina primary . . . left Republicans . . . divided, both Bush and McCain . . . campaigned for his election. . . . " Or they could have remained as two sentences, with the second beginning however instead of with the voguism that said.

Immediately after introducing "[t]his absolutive participial construction now spreading like wildfire through our discourse", Safire quotes a linguist:

We turn now to Prof. John Lawler in the linguistics department of the University of Michigan: "That said is an abbreviated form of the absolutive participial phrase '(With) that (having been) said."' (You were wondering where I got that "absolutive participial" from? You think I make up this stuff? If I had taken Latin, I'd be able to explain the closely related ablative absolute.) Lawler goes on to the essential meaning of that said: "It announces a change of subject, often despite whatever was just said."

By calling that said a "voguism", and noting that things would have different in "olden times", Safire flirts with linguistic peevishness, as he often did. Many of his readers no doubt preferred to interpret him that way, just as LL commenters often write as if the question under discussion were whether a particular usage annoys them. And in this case, Safire was put onto the scent of that said by emails like "The latest abomination is the substitute for however in its many forms — having said that, that having been said, et cetera. That said, I guess I'll just have to get with it."

But after some alliterative harrumphing ("the vocabulary of vacillation"), he describes that said as an inverted-order rhetorical equivalent of "to be sure" ("X. To be sure, Y" == "Y. That said, X"), and closes with the opinion that

To be sure, that said has its good side.

Why am I bringing up this seven-year-old column? Because one of the many interesting talks that I heard at AACL 2009 was Laurel Brinton's 'The development of 'that said'".

Brinton cites the results of a COCA search showing that that said has indeed increased in popularity over the past couple of decades, apparently by moving from (some registers of) the spoken language into writing:

She also gives the results of a search of the Time Magazine archive (also available on Mark Davies excellent web site at BYU), which suggests that a fairly rapid increase in popularity beginning in the 1990s:

And she presents evidence to contradict a temporal implication of Safire's quotation of John Lawler: 'That said is an abbreviated form of the absolutive participial phrase '(With) that (having been) said."'

I suspect that John probably meant this as a description of a cognitive process, or perhaps simply as a way to clarify the grammatical structure involved; but as an account of the historical development of this phrase, Brinton argues that it is  false.  Before giving her own new evidence, she cites earlier authorities who give a different account:

Curme 1931 (A Grammar of the English Language. Vol. II: Syntax, 153): absolute constructions without copula are original, more modern trend is to take tense (having) and voice (having been), especially when active

Jespersen 1946 (Modern Grammar on Historical Principles. Part V: Syntax, Fourth Volume, 46, 55): that said is the original construction; the "more clumsy construction" that being said "begins to appear in the 16th century; that having been said dates from the 18th century.

Visser 1972 (An Historical Syntax of the English Language. Part Two: Syntactical Units with one Verb, 1259ff.): generally, absolute constructions with the simple past participle are original, forms with having/being/having been are introduced in the modern period and are now often preferred.

Before pursuing the history, Brinton gives a table suggesting that American and British usage may now follow somewhat different patterns:

COCA BNC [punc] {that, this} said, 773 134 [conj] {that, this} said, 67 5 [prep] {that, this} said, 52 3 all (of) {that, this} said, 59 4 {that, this} being said, 172 16 {that, this} having been said, 36 4 Total passive forms 1159 166 Frequency per million
(passive forms) 2.89 1.90 having said {that, this} 698 247 Frequency per million
(active forms) 1.73 2.82

(But note that the BNC material was collected in the early 1990s, whereas COCA contains material spread equally over the period 1990-2009 — since the pattern is presumably changing on both sides of the Atlantic, an accurate account of differences will require looking across comparable sources across comparable time periods..)

She also observes that CGEL (p. 1350) analyzes that said as an "absolute construction within the class of supplements, or "elements which occupy a position in linear sequence without being integrated into the syntactic structure of the sentence", where "absolutes" express adverbial notions such as cause, condition, time, concession, manner, and attendant circumstances. She also cites the description of that said by Quirk et al. (1985) as a "contrastive-concessive disjunct".

She notes Garner's observation (in Modern American Usage, 2003) that having said that “is a frequent source of DANGLERS” when not anchored to a speaker in the main clause; he advises deleting the “casualism” as it “doesn’t say much anyway”. However, Quirk, Greenbaum & Leech's 1985 grammar (p. 623n) claims that having said that “has become so stereotyped that it can violate” the “subject-attachment rule”; and Curme (1931:158-159) also argued that in this case, the construction is no longer connected with a subject at all.

She cites the OED's gloss, added as a 2007 online supplement to the 3rd edition's entry for say ("In phrases introducing a concessive clause. having said that (also that said, that being said): even so; nevertheless") with citations only back to 1908 for that being said and 1923 for that said:

1908 Manitoba Morning Free Press (Electronic text) 1 Aug., The story of Sir James Douglas might have been told in smaller compass… That being said, James Douglas certainly deserved a place among the makers of Canada. 1923 Times 14 Aug. 5/2 The change does not appear to be popular… That said, there is little to criticize in the performance last night. 1975 A. V. GRIMSTONE in K. Sekida Zen Training 21, I believe it would be possible..to mount a convincing refutation of the argument… However, having said that, I would add that I do not believe it is really necessary to defend the practice of Zen in that way. 1986 C. SNYDER Strategic Def. Deb. 222 We have little choice; today's technology provides no alternative. That being said, we will press for radical reductions in the number and power of strategic and intermediate-range nuclear arms.

And she observes that Safire missed the chance to give the traditional name of "procatalepsis" to the rhetorical figure that he describes as follows:

Writers of opinion articles know how to use what we call a to-be-sure graph. After making an argument, some of us feel the urge to show that we are not simpletons — that we know the counterarguments, have taken them into due consideration, but still maintain our positions. In goes the to-be-sure graph, disarming our opponents with its "Yes, we know all that folderol on the other side" […]

To be sure is a rhetorical device to set up and counter the opposition after your initial point has been made. You often present the other side's position as a straw man easy to knock down and then repeat your opening argument with great force at the end. This to-be-sure trick is described by grammarians as "concessive" — that is, I'll give you this; it costs me nothing and makes me appear reasonable.

However (to use an old construction), that said is a device that works in the other direction. The point to be negated is made first: "My opponent is a great guy, a real patriot and a quick study." (End of concessive construction.) "That said, he doesn't know what he is talking about."

Compare the relevant section of Henry Peachum's 1593 Garden of Eloquence:

Procatalepsis is a forme of speech by which the Orator perceiving aforehand what might be objected against him, and hurt him, doth confute it before it be spoken, or thus: when the Orator putteth forth the same objection against himselfe, which he doth thinke his adversarie would, and then refelleth it by a reason, whereby he doth providently prevent him. Cicero: as if some Judge or commiissioner might say unto me, thou mightest have contended with a lighter action, thou mightest have come to thy right by a more easie and profitable way: wherefore either change thine action, or resist me not as Judge: or if he do prescribe after what sort I ought to sue for my right, to which objection he maketh this answere. Notwithstanding he seemeth either more fearfull then is reason a Judge should be: or else he dareth not judge that which is committed to him. Likewise against Verres, Cicero saith, that he knoweth some men will marvell, seeing so many yeares he defended many, and hurt none, he doth now come to accuse Verres, then he doth shew them that this accusation against Verres is a defence of their fellowes.

Returning to the history of that said, Brinton notes that despite the antiquity of procatalepsis, this particular way of expressing it seems to be of fairly recent origin. The earliest examples in the OED are from 1923 (that said) and 1908 (that being said), with a 1975 citation for having said that, and no examples of the other forms.

She found no examples in the following corpora of Early and Late Modern English:

Helsinki Corpus of English Texts, Modern English Section (1500–1710) 551,000 words
Corpus of English Dialogues (1560–1760) 1.2 million words
Lampeter Corpus of Early Modern English Tracts (1640–1710) 1.1 million words
Corpus of Late 18C Prose (1761–90) 300,000 words

She checked these additional corpora:

CLEMT Corpus of Late Modern English Texts (1710–1920) 10 million words
CEN Corpus of English Novels (1881–1922) 25 million words
UofV University of Virginia Electronic Text Center, Modern English Collection (1500– present)
EEPF Early English Prose Fiction (1500-1700) 200 works
18thCF Eighteenth Century Fiction (1700-1780) 96 works
Early Canadiana Online (18th c. - 1920) 3 million pages
ED English Drama (1280-1915) 3,900 plays
TIME Corpus (1923 – 2000)

She was able to find a number of earlier uses, e.g. "And these wordes sayd, she streyght her on length and rested a whyle", from 1387-88. But these are the literal usage that she calls "temporal/sequential", not the "contrastive/concessive" use.  The earliest clearly concessive examples she was able to find are from the late 19th to early 20th century:

This being said, Mr. Hamilton and his colleagues of the executive council of the Territories will pardon me if I do not receive … the assurance given by them … (1894 Sessional Papers of the Dominion of Canada, Vol. 17, p. 57; ECO)

That being said we come to the question as to whether the work is worth the money (1900 Official Report of the House of Commons of the Dominion of Canada: Fifth Session, Eighth Parliament, p. 9663; ECO)

Thus contra Lawler, Brinton argues that "there is no evidence of the construction being a reduced form of (with) this/that having been said" — she found no non-recent examples of this pattern at all. The time-sequence uses of this said date from the late 16th c., with the earliest examples of temporal/sequential that said from a bit later:

This said, impatience chokes her pleading tongue (1592-93 Shakespeare, Venus and Adonis; UofV)

She is a hansome picture, And that said, all is spoken (1636 Massinger, The Great Duke of Florence; ED)

She found a few possible (but ambiguous) early examples of concessive that/this said:

They all are made my Lord, and some giue out, That ‘tis a blow giuen to religion, To weaken it, in ruining of him, That said, he neuer wisht more glorious title, Then to be call’d the scrouge of Hugenots (1608 Chapman, Charles Duke of Byron; ED)

Then, daughter, graunt me one request, To shew thou louest me as thy sisters doe, Accept a husband, whom my selfe will woo. This sayd, she cannot well deny my sute (1605 Anon., King Leir; ED)

Then there's a gap of three centuries or so, before clear examples of that said begin to appear:

The change does not appear to be popular … That said, there is little to criticize in the performance last night (1923 Times 14 Aug 5/2; OED)

So summarizing her findings in a table:

this said that said this being said that being said having said this having said that Sequential late 16th c mid 17th c early 16th c — early 17th c mid 18th c Concessive ?17th c 17th c?/

20th c late 19th c early 20th c mid 18th c 20th c

I observed in the question period after Brinton's talk that the Latin phrase his dictis ("these [things having been] said") was reasonably common — for example, it occurs six times in the Vulgate. Thus Luke 19:25

Et dixerunt ei: Domine, habet decem mnas. 26 Dico autem vobis, quia omni habenti dabitur, et abundabit: ab eo autem qui non habet, et quod habet auferetur ab eo. 27 Verumtamen inimicos meos illos, qui noluerunt me regnare super se, adducite huc: et interficite ante me. 28 Et his dictis, præcedebat ascendens Jerosolymam.

The Douay-Rheims translation:

25 And they said to him: Lord, he hath ten pounds. 26 But I say to you that to every one that hath shall be given, and he shall abound: and from him that hath not, even that which he hath shall be taken from him. 27 But as for those my enemies, who would not have me reign over them, bring them hither and kill them before me. 28 And having said these things, he went before, going up to Jerusalem.

His dictis also occurs several times in Virgil, e.g. Aeneid IV 54

His dictis impenso animum flammauit amore
spemque dedit dubiae menti soluitque pudorem.
principio delubra adeunt pacemque per aras
exquirunt;

So saying, she stirred a passion-burning breast
to love more madly still; her words infused
a doubting mind with hope, and bade the blush
of shame begone. (Theodore C. Williams translation)

Thus educated 16th- or 17th-century authors would probably be familiar with his dictis — but all the classical examples that I've been able to find are strictly temporal/sequential.

Brinton's conclusions:

The concessive meaning develops from, and has replaced, the temporal/sequential meaning.
The longer forms develop from the shorter forms.
That/this said and having said that/this have separate histories.
For the most part, the that forms have replaced the this forms.
That said shows a marked increase in frequency beginning in about 1990.
Semantic change in this construction conforms to well known paths of change.

With respect to the last point, she cites Traugott and Dasher’s “correlated paths of directionality” (Regularity in Semantic Change, 2002):

truth-conditional > non-truth-conditional content > content/procedural > procedural scope within proposition > scope over proposition > scope over discourse non-subjective > subjective > intersubjective

But still, it surprises me that this process didn't take place much earlier.  For example, there are plenty of Latin phrases that followed the path to "non-truth-conditional", "procedural", "scope over discourse", "intersubjective" uses.  So why didn't his dictis make this journey?

And was the 1990s that said vogue a random event? Or has there been something especially procataleptic about the past couple of decades?

Counting languages isn't an easy task; in particular, it's hard to say whether two varieties are related languages or dialects of a single language. Making these decisions on linguistic grounds is difficult enough, but political, cultural, and social considerations often intervene, to compound the difficulty. The latest Ethnologue (16th ed., 2009) advertises itself as "an encyclopedic reference work cataloging all of the world's 6,909 known living languages", but the introduction lays out the problems in identifying and counting languages and acknowledges that the methods used in reaching this very exact number are not the only possible ones and that these methods involve judgment calls at several points.

Same thing in counting ethnic groups. But sometimes an authority just stipulates a figure, as in this report on school textbooks in China ("The fragility of truth", The Economist, October 10, p. 45):

The authorities have said that 56 columns erected on Tiananmen Square for the celebrations [the recent 60th anniversary celebrations] will stay put. They represent China's officially recognized 56 "ethnic groups". It is a number that few Chinese schoolchildren would dare to challenge. Yet when the communists came to power there were found to be more than 400. Officials eventually settled on the much lower figure and suppressed further debate.

(That's 55 official minority nationalities, plus Han Chinese.)

I don't see any official pronouncement on the number of languages in China. What the Ethnologue says is:

The number of individual languages listed for China is 293. Of those, 292 are living languages and 1 has no known speakers. (link)

Language and ethnicity are related in complex ways, but 291 languages (discarding the 2 special cases) seems to me to be an awful lot for only 56 ethnic groups.

I've heard many interesting papers here at AACL 2009. Here's one of them: Bridget Jankowski, from the University of Toronto, "Grammatical and register variation and change: A multi-corpora perspective on the English genitive".  She was kind enough to send me a copy of her slides, from which I've taken (most of) the graphs below.

In order to study the history of choices like "Ontario's government" (s-genitive) vs. "the government of Ontario" (of-genitive), she created two small historical corpora, sampling Maclean's magazine and the Hansard transcripts of debates of the Ontario Provincial Legislature at three time points: 1906, 1956, and 2006. She picked three authors or three speakers from each source at each time point. All of the speakers and authors were men aged 30-60 at the time of the sample.

Her first result is a replication of the observation that the s-genitive has been gaining ground:

Compare, for example, this figure from Hinrichs and Szmrecsanyi, "Recent changes in the function and frequency of standard English genitive constructions: a multivariate analysis of tagged corpora", English Language and Linguistics 11(3): 437–474, 2007:

Jankowski then broke the trends down further by coding the possessors as

1. Human: a student’s schoolwork, Mrs. Hale’s reaction
2. Organizations (animate “collectivities of humans which display some degree of groupidentity”): the local school board’s ruling; the federal government’s plan
3. Places: Canada’s foreign language press, Ontario’s roads, the streets of Rome, the raw edge of the world, the people of this American continent
4. Inanimate objects, activities, units of time, states

This made it clear that the increase in use of s-genitives has been especially strong in the case of organizations, and even stronger in the case of places:

Her category 4 ("Inanimate objects, activities, units of time, states") was realized overall with of-genitive 96% in Maclean’s and 99% in Hansard. So her results are generally consistent with Otto Jespersen's observation in A Modern English Grammar on Historical Principles: Part VII (1949) that

In poetry and in higher literary style, the genitive of lifeless things is used in many cases where of would be used in ordinary speech. […] During the last few years the genitive of lifeless things has been gaining ground, (especially among journalists)…

but only if "lifeless things" is taken to include organizations and places, and not "inanimate objects, activities, units of time, states".

She also compared her results to data from a corpus of conversational speech collected recently in Toronto, using speaker age to create two "apparent time" collections comparable to the 1956 and 2006 samples. This suggests that in the spoken language, human possessors have almost always gotten the s-genitive, consistently across time, while inanimate possessors (in this graph including her categories 3 and 4) have consistently gotten the of-genitive:

On this analysis, the increase in s-genitives for human possessors in Maclean's magazine makes the journalistic prose more and more like the spoken language; but the parallel increase in s-genitives for inaminate possessors makes the journalistic prose less and less speech-like.

Her presentation also considered the effect of the length of the possessor (a shorter possessor is more likely to take an s-genitive) and the possessum ("shorter possessum will be more likely to take an of-genitive and so appear first in the construction, while a longer possessum is more likely to take the s-genitive"), as well as other relevant features such as "lexical density":

and "thematicity":

and did a multivariate analysis of all the various factors taken together.

You'll have to read her (I trust forthcoming) paper to learn how it all comes out — I have a 6:40 a.m. plane to catch — but I hope that this much is enough to convince you that there's a rich and interesting pattern of variation to be untangled here. It certainly convinced me.

And it also increased my general feeling that the time is right for the application of automatic or semi-automatic methods of analysis (here in assigning her four categories of possessors, in determining the lengths of the possessor and possessum constituents, in counting local phrases co-referential with the possessor, etc.) to the study of syntactic variation across time, genre, register and so on. Because she had to annotate everything by hand, Jankowski's sample was fairly small — 50K words of Maclean's, and 100K words of Hansards. With automatic or semi-automatic annotation, she could look at larger collections with denser time samples of more sources, and easily add other features, like various word and phrase frequencies, grammatical role and phrasal position of the whole genitive construction, etc.