Ask the linguist

Miscellaneous and Personal Stuff I Must Share
1

I’ve seen a number of really, really cool language threads in the last few weeks, so as a new member, I thought it’d be neat to volunteer my pseudo-expertise to this forum. I know we have a few linguists floating around other than myself, so please feel free to chime in!

Anyway, down to it: I’m a graduate student in linguistics, and I specialize in language acquisition and historical linguistics. The former is, like it sounds, concerned with investigating how people acquire their first language, as well as how adults acquire new languages. It’s a wonderful thing to be interested in, but since we need a couple hundred years of advances in medical technology before we can really look at the biological mechanisms my methods tend to be much more roundabout, and I swear I occasionally catch the stench of garlic and burning witches in my laboratory.

Anyway… does anybody have questions about language, linguistics, or the like? A lot of linguistic theory is up for grabs, so nothing I tell you will be “correct”… I personally don’t know what a word, a sentence, or a language is at this point… but I’d love to see what questions people have, and I’d be even happier if people who disagree with me feel free to shoot a paw up and argue their views.

2

Allow me to be the first to throw myself on the obvious question:

Are you cunning?

[SUB]sorry[/SUB]

3

I refuse to answer that question on the grounds that I resemble it. :wink:

It’s funny though, that joke would’ve actually worked just as well when the word was invented. We’re pretty sure that “cunning” comes from “cunnen,” which came from the Proto-German “kunnan,” which means “To know,” and which probably ultimately came from the Proto-Indo-European root “gno”, which is also the base that the English “know” came from. The joke would still work, however, because “cunnen” actually meant EITHER “to know, have power, be able to” OR “to have carnal knowledge”.

So I’m going to say that I’m a cunnen linguist, and let you ponder which meaning I’m insinuating. :wink:

4

What’s the consensus (if there is one) on whether all language derive from one original language or if more than one language arose at different times during human evolution? Given what we know about human evolution, it would hard to imagine that the latter were the case.

5

I just asked my son, who is very interested in language, this question:

Why do some word take an article in American English but not in UK English? For example, in the USA, you go “to the hospital” but in the UK you go “to hospital”.

So why is it that? College, not “the college”. “The university” not “university”. What makes the difference?

6

Well I’ve never studied evolution, so I’m not really qualified to argue the latter case. The former, however, is colloquially called the Proto-World hypothesis. Some good linguists have worked on it, but in general historical linguists tend to be dimissive of proto-world as a family. There are a few reasons for it, some more convincing than others.

One big problem is that the main tool we use to figure out what languages are related, and what their mutual ancestors (“proto-languages”) looked like, which is called comparative reconstruction, can only go back about 16,000 years at best, and around 9,000 years at worst.

Essentially (and you can find lots of great info in Lyle Campbell’s Historical Linguistics: an Introduction), the comparative method involves assembling lists of cognates between two or more languages that we suspect are related. Once we have a large amount of data (you can get a good theory with about 50 items, but to be convincing you need a lot more) you begin looking for correspondences.

A correspondence is a phoneme (sound) that is the same in more than one language. For example, let’s say we have four imaginary languages A, B, C, and D, and their words for “dog” are “Kma,” “Gma,” “Lma,” and “Desaba”, I would say that we have a “m” correspondence between languages A, B, and C, since all three have the same phoneme in the same spot. We would use this to propose that if these languages are related, their proto-language had a *m sound. (we use an asterisk to indicate a proto-form, in this case).

Now we look at the initial sound. Language A has “kma” and language B has “gma”. The “g” sound is actually articulated in the exact manner as the “k” sound, they’re both examples of velar stops. (they’re called stops because the sound is made by forming a closure in the mouth to stop outward airflow, waiting for the pressure to build up, then releasing the pressure to create an audible noise. The popping sound of a champagne cork works on the same principal. They’re called velar because the stop is actually created at the velum, which is back behind the palatte on the top of the mouth). They’re both velar stops, and the big difference between them is in voicing, that is, whether or not your vocal folds are vibrating when you release the stop closure. (you can play with voicing on your own: gently place two fingers on your throat, and alternate between saying “ssssss” and “zzzzz”. Notice the vibration in “zzz”? That’s your vocal folds vibrating.)

So far, we’ve assembled three “features” of the initial sounds in kma and gma: both sounds are velar, both sounds are stops, but one sound is devoiced and one sound is voiced. Where do we go from here? Well, if we look at the next sound, the “m”, we’re going to notice that it’s voiced. (Don’t believe me? You have a working three-dimensional model on top of your shoulders: try making it yourself and decide.) A very, very common type of sound change is for voiceless sounds to become voiced when they occur adjacent to a voiced sound, and so I’m going to propose that the first sound was originally “k” in both languages, and that language B developed a “sounds adjacent to voiced sounds assimilate the voicing” rule. (please note that there are a LOT of alternate things that could have happened, and it’s impossible to say for sure unless we have ten or twenty examples of k + m/g + m cognates in both languages.)

As for the vowel, this one is actually easy: there’s an /a/ in languages A, B, and C, so I’m going to go out on a limb and propose that it was a vowel in the proto language.

So now we’ve reconstructed three potential phonemes of our proposed proto-ABCD language: *k, *a, and *m. (Remember that asterisks indicate a proto-form!) The m and a correspondences have been seen in 3/4 languages, and the k/g alternation has been seen in 2 of them. Based on this evidence, I would argue that A, B, and C all had a common progenitor. Language D, on the other hand, is totally different… but we don’t want to discard it out of hand. Rather, we need to look for more cognates to make sure that it doesn’t have any significant amount of sound changes in common. The reason for this is because a lot of things could explain the lack of a cognate here: maybe D borrowed the word for “dog” from a unrelated language that was spoken in the same area, or maybe A, B, and C borrowed a word for "dog’, and D actually has the original proto-word. It’s tough to say from this data, so off we go to find more stuff!

That’s the comparative method in a nutshell. But please don’t ever, ever do it like I just did: you need a huge amount of cognates before you can start saying ANYTHING significant about observable relationships.

And that, as a matter of fact, is why most historical linguists refuse to entertain any current theories about a Proto-World language: language has existed for much long than 9,000 years, and because of this even if every language did descend from a single original language, it’s impossible to tell: languages have been travelling, changing, and inbreeding so much at this point, that it’s generally considered to be impossible for a researcher to find a statistically significant number of correlations, since by this point a lot of correlations will have arisen by chance.

Did that do a fair job at answering your question?

7

This is really, really cool! I haven’t actually heard of this alternation before, so my first step is going to be to ask one of our professors who specializes in linguistic syntax (sentence structure), who happens to be English.

My first guess, without any research to back me up, is that this is some kind of collocation, but since that sounds like too much of a cop-out I’m going to have to ask you to let me get back to you.

8

I’ve noticed something interesting in East Asian people when they speak English. Their languages, IIRC do not have tense marking so what I’ve noticed is that when speaking English, if they intend to signifiy past tense, they will omit the marker, but only if the word has one of the standard past tense markers. For example “I sent it yesterday” becomes “I send it yesterday” (because “send” ends in “d”). I observe this phenomenon in speakers that would otherwise use the past tense correctly. Most interestingly, I see it occur when the verb in question ends in the irregular ending /-en/, for example with “to happen”. “I didn’t know it had happen”, for instance. It seems the entire set of English past tense markers gets internalized in the second language learner, even rare, irregular markings.

9

Is there any fundamental difference in brain activity between speakers of synthetic languages vs. analytic? Does the speaker of an analytic language learning a synthetic language later in life (or vice-versa) experience any change in brain pattern/cognitive function?

10

Hmm. Well a lot of this is going to depend on the specific language we’re talking about, but based on your observations I’m guessing that the speakers you heard were native speakers of one of the Sinitic (Chinese) languages, which tend to be isolating languages and are infamous for not marking tense information in the verb. With that in mind, my best guess is that the mistakes you hear are due to bad teaching: it’s extremely hard to adapt to the English tense system if you’re a native speaker of an isolating language, and what most students end up doing is just memorizing agreement rules by rote: “if you’re talking about the past, verbs in this group get “-ed” added to them, while verbs in this other group get “-en” added to them”.

Since you don’t see this type of agreement in many Sinitic languages, it’s going to be one of the things speakers will tend to easily leave out.

(I don’t actually speak any of the Chinese languages, so if we happen to have any fluent speakers or ESL students I’d love to hear your perspective)

11

That’s an awesome question. There is, but unfortunately I haven’t seen any evidence of a direct correlation between languate typology and neural activity. There isn’t one “archetype” of brain architecture when it comes to language, for one thing, since individual variations are extremely common. For instance, while most middle schoolers learn at some point in their career that language is processed in the left hemisphere of the brain, it’s actually the case that only about 90% of north americans tend to manifest increased activity in the left hemisphere during language-related tasks. Most of the remaining 10% actually uses the right hemisphere, and a smaller percentage of that group seems to use both hemispheres simultaneously.

The problem is that we don’t have a precise way of telling which regions do what, when it comes to syntax. Speech production is significantly easier because a lot of it is simpler (we know that Broca’s area and Wernicke’s area always occur unilaterally, for instance, or that the right hemisphere of the primary auditory cortex analyzes prosodic characteristics, whereas the left hemisphere tends to identify phonemes).

When it comes to syntax, we can’t be nearly as precise as we need to be. The lab I work in is actually lucky enough to have a great MRI, but we can only get about five centimetres-squared of accurate resolution, and that isn’t nearly enough to discriminate between the various regions that light up during speech.

Rather, a lot of what we do is based on logic. We know based on observation and experimentation, for instance, that three and four year olds have Complementizer Phrases based on the various types of movement they allow, and based on errors in determiner (words like ‘the’ and ‘a’) usage we’re pretty sure that kids have trouble marking for definiteness and specificity. (in particular, we find that native English speakers tend to overuse ‘the’ and underuse ‘a’ until they get into the 5-8 year old range. And please yell at me for citing ages, because if there’s one thing acquisitionists need to understand, it’s that age is a very poor indicator of lingual competency.)

If you’re interested in a more experimental approach, I would recommend looking into lexical and pragmatic priming. It’s really hard to do with kids, since it requires a lot of patience, but adults are a great source of data and there are some really neat, well-proven methods for documenting how priming occurs.

12

Similarly, Brits say a patient is “in a critical condition,” and Americans say a patient is “in critical condition.” This is very difficult to explain to someone, like a Thai, whose language contains no articles.

13

There was actually a pretty cool study written around determiner errors a few years ago. They gathered a whole slew of native Russian and Japanese speakers, who had little exposure to english, and taught them english over a year while monitoring their errors.

What was interesting is that even though neither language’s determiner usage is anything close to what English does, the Russian speakers didn’t make nearly as many mistakes. I think Russian students messed up usage of the/a/an in an average of 30% of the sentences they produced during the year, while Japanese speakers were way up around 60%.

14

Two questions:

  1. Assuming I am unavailable :wink: who or what should be the ultimate arbiter of “correct” usage for the English language? If it’s the polloi, why should the educated succumb to a pooling of ignorance? If it’s the educated, could not one argue that they have their own ultimate usage roots with the masses? If it’s the editorial board of a standard-usage book, in what sense does their consensus not derive from either or both of the first two? If it’s none of the preceding, should we spend any time scolding the polloi for butchering the King’s–um, I mean, the President’s–English? (Or should I say, scolding the President for butchering the polloi’s English?)

  2. (An easy one…) Is there any language better than English?

More if you get these two right… :slight_smile:

15
  1. Nobody. As a descriptive scientist I’m going to be opinionated and claim that it’s stupid to think that we can tell people how to speak, and insane to judge intelligence or class based on that speech. One of the serious problems with the way grammar is taught today, at least in American schools, is that it’s heavily prescriptivist: they tell students what’s right, what’s wrong, and try to justify this with extremely bad logic rooted in using Latin grammar as a model for English.

Now, that isn’t to say that I don’t like English classes: I think that English speakers in general need lots and lots of practice in order to learn how to write in a clear, concise, understandable manner. But telling someone how they should talk is just silly. A person is going to have acquired practically all of their intuition before they ever set foot in school, and the best an educator is going to be able to do is eliminate some regional slang.

In English, anyway. There ARE languages that have extremely complex socially driven rules, and in many cases these aren’t learned until later. Japanese and Tibetan are two classic examples, since each has an extremely complex system of politeness that most people don’t learn until they’re 12-15 years old. As such, you can usually pick out the native speakers who grew up in another culture, because they’re the ones who have perfect grammar and absolutely no idea how to use honorifics.

  1. Depends on your personal opinion, since I can’t think of any definition for “better”. Personally, I have a serious beef with the English writing system, but other than that I think it boils down to one’s own opinion.
16

I’d go with my first guess of It’s Semantics. The two varieties of English have different semantic rules regarding the noun in question, ‘hospital’. For the US variety, the word requires (usually) a determiner.

I received my AB in Linguistics on June 17, 2004.

17

  1. Why was I caused to learn diachronic and synchronic linguistics using Anttila when none of us spoke Finno-Ugric?

  2. If the colorless green nouns slept less furiously, would fruit flies like a banana?

18

In GB English you can say “in the hospital” or “at the hospital” unless you are hospitalised - then you are “in hospital”. Is this different in the US?

19

Yes. At or in **the **hospital.

20

I’m slightly out of my depth here, since I’m a phonologist, but if semantics is involved I think that it’s only involved insofar as the lack of DP is motivated by some kind of weird collocational restriction.

My reasoning is that determinerless PPs (prepositional phrases) occur quite frequently in both north american and British english: on ice, by hammer, at school, in gaol, on video. The point of giving these examples is to note that their occurence is often predicated by one particular meaning: if we’re ice-skating I, as a Canadian, have to say “We’re on THE ice,” but if the mafia have killed us, I’m allowed to say “We’re on ice”.

This is essentially what a collocational restriction is, and since some north american determinerless PPs (at school) are suspiciously similiar to British “at hospital,” I think we’re seeing British english gain new contexts that allow for determiner dropping.