First, the sentences:

He was there for a good seven years.

An additional three people are required.

A mere four nations recognize that standard.

She collected an amazing and heretofore unprecedented forty million dollars.

What we have is “a” and then some adjective phrase, and then a quantified nominal. There are some interesting questions to be asked: first, what is the range of adjectives? It seems sort of limited: a grueling 100 miles, but ?an asphalt-paved 100 miles. All the examples given so far involve some sort of “evaluation” (shock, amazement, disappointment, unprecedentedness, etc.). Maybe someone nice will do a corpus study and report the findings (and of no one does it soon, I might just have to).

Next question: does the whole thing act as a singular or plural phrase, for the purposes of subject-verb agreement? The comments seem to show that, depending on the “context” (how the NP is construed semantically, let’s say – either as a divisible group of individuals or as a lump), you might get singular or plural agreement.

A good 100 people have/*has arrived.

A mere four nations recognize/*recognizes that standard.

A mere four nations is/are not enough

To be clear, it’s not just this funky “a” construction that has this effect: I think 100 people has the power to convince our boss that he’s acting inappropriately. That is to say, “100 people signing the petition” or something similar. The CGEL, unsurprisingly, has discussion of this (pages 501 onward, especially 504 on “measure phrases” as contexts for agreement overrides).

Perhaps, in fact, the idea of “measure phrase” is about right for the funky “a” constructions. That brings us to the next (and very related) question: what do these things mean, anyway? Claire (a commenter on Mark’s more recent post) has the intuition that in

a stunning 19.32 seconds

what we are dealing with is

a stunning record/time of 19.32 seconds

which has an “equative” of, which appears not just in quantified expressions but elsewhere (compare Seven of the 68 took less than 80% of the prescribed dosage over the study period for the reason of lack of response to DJT.). The idea that the “of” is “implicit” would need some further specification, but the idea would potentially be that adjectival modification of a particular type creates a sort of pseudo-measure-phrase out of the quantified nominal. This might then also explain any variability in subject-verb agreement (since there is both the quantified nominal as well as the singular semantics of the implicit measure phrease).

Radek (replying to Claire) gives some more evidence. One bit is something that I also mentioned in my initial comments, about semantic composition. He notes (and I agree) that in a mindboggling 281187 words, which is mindboggling is not the words, but rather “the dummy word ‘amount’.” This is basically what I think, though you don’t necessarily need a “dummy” (unpronounced?) word, but simply a semantic entity — the quantity of words — that can be modified. Whether that meaning has to be represented by some implicit word is a separate question.

Mark, too, I think, was grappling with this when was dealing with the original interesting question, namely what is the constituency structure?

However, I’ll admit that the constituent structure doesn’t feel like

[ [a modifier number] noun]

but rather feels like

[ [a modifier] [number noun] ]

I think this too involves semantic intuitions influencing syntactic ones (but not being a mind-reader, I could be totally off; there are also prosodic factors to consider, and just plain “well, that’s how it looks to me”). The latter structure would capture the idea that what is mindboggling, shocking, amazing, etc. is not the number per se but the quantified expression.

Going further afield, I have the vague feeling that he plays a solid third base and she sings a mean alto are in the same family of slightly-odd modification and determination. Also in this family might be the France of my youth and the always-controversial San Francisco.

[some format editing was done soon after posting...in case you were wondering]

I don’t want to get into the details of this particular case, but I do want to comment on something in Geoff’s final paragraph, which begins:

Here on the Linguistic Crimes desk we try to highlight the lighter side of language offenses: the zany character of victimless criminality that amount to no more than uttering strings of letters of syllables, the mad asterisking of words too awful to print, the giggleworthy character of loony attempts to suppress free speech.

Now I am sure Geoff has thought about the philosophy of language, and in particular the philosophy of what language “means,” far more than I could have if I had started when I was born and not stopped until last night, but there is something interesting in this particular statement. That is, on the one hand, we have the fact that “uttering of strings of letters and syllables” cannot reasonably constitute victim-ful criminality (being inherently meaningless, one presumes). On the other hand, we have the importance of the freedom of speech, which one presumes is an important thing to do because sometimes speech is meaningful or somehow experienced as meaningful.

But like I said, I’m no philosopher.

A side-note: check out the photo of Alan facing Ramona. On the shelf in the background there’s a crank-style pencil sharpener. Anyone else think there might have been a better place to install it?

The approved list of kanji, the Jinmeiyoo Kanji (‘Chinese characters for use in personal names’), consists of 983 kanji that do not appear in the standard 2000-odd standard kanji used in everyday writing, giving parents about 3000 characters to chose from. Excluded from the list are many characters that indicate culturally taboo or offensive concepts, like prostitution, cancer, and various emotional states (resentment, e.g.). Once, there were parents who attempt to give their child a name like ‘demon’ or something similar, and this name was rejected as a form of abuse of parental powers, due to the expected social difficulties that the child would be expected to experience (but, I haven’t heard anything about Japanese parents tying to put symbols in names, like in the Chinese story).

There is apparently also a list of approved Chinese characters for personal names in Korea (5151 characters). According to this article, offensive/etc. characters are included in the list, but cases like the ‘demon’ one haven’t come up. But, the article says that “there have been cases where a proposed name has been rejected due not simply to non-standard Chinese characters, but because the name was difficult to pronounce, or would cause serious harm to the child if s/he used it [as a primary name] in public life.” Also on the list of restrictions in Korea is that the name must be represented either completely in Chinese characters or completely in hangul, and also that in the latter case it cannot exceed five syllables (character-blocks).

These one seems like the least reasonable of the restrictions, as it seems the most aesthetic or technical. To be sure, a name written in both hanja and hangul would look weird, but probably because it just hasn’t been done yet. Hangul was designed to approximate the appearance of Chinese characters, so if good old King Sejong’s linguists did their job right, it should look just fine…no? As for the length restriction, I can’t imagine it comes up much at all, as most (all?) Korean names are three syllables long, but still: why continue to have the restriction?

Now approaching Millbrae station. Millbrae is your transfer point for BART, SFO, and the East Bay.

This would be made when heading either south or north. I found it a little odd that BART would not necessarily be a transfer point for, say, San Francisco. After all, as far as I can tell, there aren’t any other places where a person might actually want to chose between going to a BART station or a Caltrain station – they kind of service different parts of the city, at least if you’re walking or biking. So it seems like either they wanted to save a little time, or just didn’t want to reinforce the idea that you could travel to important locations in SF via BART (in fact, several more locations than you could get to on Caltrain).

Now, something else interesting happens on BART. Actually, it doesn’t “happen,” it’s printed on their system maps. In addition to BART lines, other rail systems like Amtrack and Caltrain are shown, with thinner lines and smaller station labels. interestingly, all the Caltrain stations south of Millbrae have station labels (Palo Alto, Redwood City, and so forth), but none of the stations north of Millbrae, where BART has service, have station labels. Nice one.

This reminded me of the plight that faces the field of linguistics, which so far has been mostly unable to get itself into the pre-college curriculum. Of course, it’s probably not exactly easy to convince people that knowing how to recognize grammatical functions and phonological alternations is important (and certainly not compared with knowing the locations of nations we’re currently occupying…or, say, the states that we live in). But in any case, I watched the 60-second ad available on the site, transcribed it, and then altered it as little as possible, to create the following. I recommend checking out the actual video first, though…

What’s wrong with this series of pictures? Half of young Americans can’t distinguish important morpho-semantic categories like tense and aspect. Twenty percent can’t even identify a passive clause. Without linguistics, our children aren’t ready for the world. Language is everywhere. It’s incredible sounds, rhythm, transformations, flavor. It’s economics, and politics. It’s change. Understanding connections between people and places is critical in the 21st century. That’s why we created thewonderoflanguage.org Go there now for your free parent and teacher action kits, and give our kids the power of linguistic awareness. Because kids who understand our languages today can succeed with them tomorrow.

There is then some back and forth between the two, and eventually everything is clarified. Then the caller, being exlanatory/self-deprecatory, remarks that he tends to “take things literally.”

Though the caller may indeed take many things literally that aren’t meant to be, perhaps in this case he should have said that he tends to screw up interpretations of series of adjectives. No doubt his friends avoid talking to him about their Turkish history teachers and fake leather wallets, hoping to avoid wasting precious minutes explaining what they’re actually talking about.

The question is, assuming that most people understand this fellow’s self-characterization, what the heck does “literal” mean? Now, in the context of the commercial it’s clear what is meant by “literal,” since the source of the guy’s misunderstanding obvious. I just wonder how many people, asked to give examples of misunderstandings due to taking things too literally, would provide cases of attachment ambiguity. (“Oh, the guy you were spying on had the telescope? I guess I just take things too literally…”). Perhaps a native interpreter can provide intuitions.

As a side note: the service rep, when she gets that ridiculous question, admits that the words “extra leg room” do “imply” the meaning that the guy initially had. That’s also interesting, since when I think of something implying another thing, both things can be true at once. But in the extra leg room case, one reading can’t really imply the other. My guess that this is related to, loosely speaking, “perspective.” That is, for the rep there was a the “core,” intended meaning, but there was an additional one that was slightly covert, i.e., “implied.” Somehow I’m reminded of the semantics of a (medicinal) “side-effect,” which of course can become the main effect if the drug is repurposed.

I then flipped through a few more articles, and found in the middle of one a blown-up quotation (what are those things called?) that mentioned mirror neurons. Now, the mirror neuron is a relatively significant finding in some corners of cognitive linguistics, so I took a closer look. I found that one of the researchers mentioned (and indeed, the one mentioned at the very beginning of the article) was Lisa Aziz-Zaheh, who I met when she spent a year at ICSI. See, it’s really only a few steps until mainstream linguistics gets the front page treatment from Time!

Then, later this evening, I found that Stephen Colbert had interviewed Pinker last night. I watched the interview (which was nice, with some good lines from Colbert; I enjoyed the geek/rich geek analogy), and was amused that the first question from Colbert was about Pinker’s hair, which is admittedly quite noticeable. Noticeable enough that, in fact, it’s pretty easy to find somewhere on the web evidence of yours truly also commenting on the psychologist’s hair (though at a time when it was a bit shorter than it is now).

As a side note, what exactly is the semantics of frequently asked question? I know it appears in places where even the author admits that the questions answered are not frequently, if ever, asked, indicating that the literal meaning is still alive (otherwise why make disclaimers). But it could also just mean “questions that, if people felt like asking questions, would be common questions”. But if you look at the pamphlet titles, not all of them are questions, like Linguistics and National Security and Bilingualism. But I suppose that’s just semantics quibbling. [hmm...a new topic for a pamphlet? What is Semantics, really?]

One also might wonder who is frequently asked these questions. I’m sure linguists are (I’ve been asked six of them…outside of a classroom), including the people who wrote and commissioned the pamphlets. But how many has the LSA, as an institution, been asked? Of course, the website does not state that the LSA has been asked the questions per se, though I suppose the Why Join is reasonable.

But Mapuche tribal leaders have accused the U.S. company of violating their cultural and collective heritage by translating the software into Mapuzugun without their permission. They even sent a letter to Microsoft founder Bill Gates accusing his company of “intellectual piracy.” “We feel like Microsoft and the Chilean Education Ministry have overlooked us by deciding to set up a committee (to study the issue) without our consent, our participation and without the slightest consultation,” said Aucan Huilcaman, one of the Mapuche leaders behind the legal action. “This is not the right road to go down.”

Predictably, most of the slashdot crowd was dead set against the Mapuche position. Given that in that part of the web, the slogan “information wants to be free” is something of a mantra, the predominant view was that trying to regulate the usage of a (human) language was the most radical type of intellectual property stupidity. Some discussants on slashdot also mentioned (reasonably, I think) the fact that the Mapuche leadership hasn’t acted (publicly) against some other online dictionaries (though this may be because they were consulted in their creation). They are also working with Chilean leadership to help revitalize and increase use of the language — though among non-speakers of the Mapuche community, not of the larger national community. This also caused confusion among the slashdot crowd — wouldn’t a localized version of popular software do nothing but help the revitalization process? In the end, it’s unclear from the news article exactly what the tribe leaders want, from Microsoft in particular or from anyone in general, who wishes to do something related to their language. It has, of course, been studied by both South American and international scholars, and I don’t know what sort of agreement those scholars had with the people who they consulted with.

The question of who “owns” a language, as I have learned in my field methods class, a very important issue for linguists and other (ethnographical/anthrolopological/etc) fieldworkers who work with politically disenfranchised people. There are many stories of groups of native people who are distrustful of university researchers who, it is feared, want to go into the community, extract knowledge about the culture and language, and then retreat to the university to publish cultural information in obscure places (like, say, academic journals) in obscure ways (like, say, using the jargon of theoretical linguistics). For a community attempting to keep their language alive, it might seem foolish and/or insulting to invest time and effort so that that an outsider can learn their language/culture, only to give nothing back to the community while profiting from that knowledge. It may also seem like a slight against them that much or any of their cultural knowledge is now fodder for academic discussion. And for this reason, many communities (I have heard mostly about those in California, but it may well be true for many other American groups) work out very (sometimes legally) explicit deals with those linguists who they work with. And those linguists are often expected in turn to help the community in working with their language, if such help is desired (such as publishing accessible grammars and dictionaries, or training native linguists or teachers, or making recordings and texts publicly or tribe-internally available).

It’s hard for me to take a clear position on the Microsoft-Mapuche conflict, since the facts are about as clear as mud. Though in general I’m a fan of free information, that’s normally in the context of everyone involved having the same ideas about what can possibly count as “information,” and what sorts of things are potentially classifiable as “private.”