Of Minds and Language (61 page)

Friederici and co-workers (2006a) entitle their paper “The brain differentiates human and non-human grammars,” and they show that violations of FSG rules activate an area of the brain called the frontal operculum. In contrast, when subjects detect violations of the rules of a recursive grammar, this also activates Broca's area. Friederici et al. (2006a: 2460) argue that

Results indicate a functional differentiation between two cytoarchitectonically and phylogenetically different brain areas in the left frontal cortex. The evaluation of transitional dependencies in sequences generated by an FSG, a type of grammar that
was shown to be learnable by non-human primates, activated a phylogenetically older cortex, the frontal operculum. In contrast, the computation of hierarchical dependencies in sequences generated according to a PSG, the type of grammar characterizing human language, additionally recruits a phylogenetically younger cortex, namely Broca's area (BA 44/45).

The area of the brain that deals with recursive grammars is phylogenetically newer than the part of the brain that deals with FSG, indicating that this might indeed be something that is qualitatively new, and specific to humans and language.

Before finishing, I would like to say something about “what we would like to know” about language. My wish list is long, but time is short, so I will choose only one wish. We have come a long way in the understanding of basic, universal aspects of language, which seemed the impossible challenge in the 1950s, when it was very much in question whether universal properties of languages even existed. However, we still need to understand much more about language variation. In his book
The Atoms of Language
, Mark Baker (2001) provides a very readable and accessible account of the principles and parameters model developed in the early eighties (Chomsky 1981).
⁵
This model assumes that language variation is systematic and results from the interaction of a finite number of binary parameters – aspects of grammar that must be specified according to the input. The model has been very successful in the discovery of systematic aspects of language variation, and it is largely due to this success that we can now ask certain questions about variation. I agree with the minimalist perspective that we can no longer entertain the view of a rich and highly elaborate UG, as envisaged in the principles and parameters model. Something makes language malleable, and I think we still do not understand this well enough. Progress in unraveling the mysteries of the complex phenomenon of language entails progress in unraveling the mysteries of our own nature, and I do hope that many dark mysteries of today will be shared wisdom tomorrow.

H
IGGINBOTHAM
: A very brief remark about the history that you gave and the quote from O'Donnell in particular. There is nothing in that quote that couldn't have been written by William of Sherwood in the fourteenth century or the medieval magicians. There is nothing of substance in it about the reuse of familiar elements and so on that wasn't known to the Stoics. So the interesting
historical question, I think, is how come that knowledge was lost in behavioral science.

L
AKA
: I agree with you in the sense that you could go as far back as Panini and find recursion, and there was an Icelandic monk who discovered phonemes. So if you go back, there are many people who have looked at language and have hit upon these properties.

H
IGGINBOTHAM
: I am sorry but I meant something stronger than that. It was actually common sense. I mean, Panini was a relatively isolated figure, but this was common sense among the relevant scholars – the medievals who were interested in language.

C
HOMSKY
: That's right; it was common sense up until the twentieth century. And now it is not common sense among philosophers. In fact Quine rejected it and so did the whole Quinean tradition. Behavioral science rejected it completely, and in fact if you look at what is called the advance of science, it is a little bit like learning phonology. You cut things out, but the trouble is that very often, what is cut out is the right things. I mean, if you look at the (seventeenth-century) Port Royal Grammar, it had recursion, it was explicit. It actually came from Galileo, who noticed that it had phrase structure, something like phrase structure grammar, it had something like transformational grammar, it had intension and extension almost exactly. (It was using it for explanation, it had the concept of explanation – they were trying to explain some funny descriptive fact in French which is called the rule of Vaugelas, which people spent a century on and they gave an explanation in terms of extension and intension.)
⁶
All of this was sort of there. Also some very important things I mentioned before and maybe I will talk about later, about meaning, which go all the way back to Aristotle and were almost totally forgotten. And it is something about the way science developed in the nineteenth and twentieth centuries that just lost lots of things. I mean, things that were pretty clear through the seventeenth and eighteenth centuries. Actually I gave an example in my talk. The observation about minds being properties of brains was standard after Newton (Locke, Hume, Priestly, Darwin). Today it is what Francis Crick called an “astonishing” hypothesis.
⁷
But I mean, what else could it be? It was understood right after Newton that it has got to be true. Why is it astonishing? The other quote I gave from the Churchlands – a “bold hypothesis.” No, not at all. The trivialities have often been forgotten, and I think if you look at the history,
you can see why. I mean, new discoveries came along, they made it look as if what had previously been described kind of intuitively and informally was baggage and we could get rid of it. The trouble is that the baggage they were getting rid of included everything that was important.

L
AKA
: I certainly agree. When scholars look at language without prejudice, they hit upon these things, because they are there. If they are not objective however, they miss important insights. Bloomfield is an example you like to quote…

C
HOMSKY
: Yes, Bloomfield is extremely interesting, and a strikingly good example of this. He was completely schizophrenic. For those of you who don't know, he was kind of like the patron saint of modern structural linguistics in the United States and most other places. I was a student; he was God. But actually, there were two Bloomfields. There was one who was the scholar, and the guy who used common sense and thought about what language has to be. He was writing grammars in the style of Panini, like his
Menomini Morphophonemics
, which was economy in grammar.
⁸
On the other hand, there was the Bloomfield who was part of the Vienna Circle, which he mostly misunderstood, if you look closely. He was heavily involved in logical positivism. If you read his book
Language
(“the Bible”), it is logical positivism.
⁹
You know, anything about rule ordering is total mysticism, everything that I am doing in my other life is nonsense, etc. When we were students in the late forties, we never knew the other Bloomfield. And in fact, a little like Mark Baker's point, he published that work in Czechoslovakia, he didn't publish it in the US. I don't know what was going on in his head, but somehow this wasn't real science, because it was common sense, the whole Paninian tradition, which he knew as a good Indic scholar, and it had results. I mean, it was so extreme.

Just to give a personal example: when I was an undergraduate, I was kind of doing stuff on my own. I did it that way because what other way could you write a grammar? And not a single person on the faculty, all of whom were scholars and knew Bloomfield, not a single one told me that Bloomfield had done the same thing six years earlier. I learned about it fifteen years later when, I think, Morris Halle discovered Bloomfield's
Menomini Morphophonemics
.
¹⁰
But it was out of their range. Some of it was really remarkable, because one of them was an Indic scholar himself and knew all this stuff. But he was such a schizophrenic that he couldn't bring it together. And the same is true in philosophy and in psychology, and Jim is absolutely right, all this stuff is all
there, it somehow just got pruned away and had to get rediscovered, step by step.

U
RIAGEREKA
: I was just going to add a footnote. In the fourteenth century, you have the exact same situation with people like Thomas of Erfurt and Radulphus Brito doing what looked like serious generative grammar, and then the philosophers going for their jugulars saying you have to study thought directly. The results of that? Well, we're still looking.

G
ELMAN
: Just a comment. Lila, myself and Jerry Fodor ran a graduate seminar where we assigned important papers, which included Chomsky's critique of Skinner. Jerry said exactly what you just said, in a somewhat different context. He said, “Now that I've reread it, I wish Noam had pointed out that this was just a bump in history.”

Núria Sebastián-Gallés

This talk is going to deal with variation in languages, a subject that we have heard mentioned quite often at this conference. As we know, the problem of why there are so many different languages on Earth has been solved. Genesis 11 gives us the answer with the story of the Tower of Babel – the proliferation of languages was a punishment from God. So the issue that I want to talk about here is not how all these languages came into being, but about another type of variation: why it is that when we try to learn a second language, some people are very good at it, while the rest of us are not.

Modern life seems to require that we learn different languages, but this is something new. In the old days, human beings, by definition, only needed to know one language, except for example when soldiers from different kingdoms marched off together to war, or when a wise king, such as Alfonso X, King of Spain, gathered in his court scientists and intellectuals from different cultures (Jews, Muslims, and Christians) to work together on the issues of the day. To do so, they needed to speak all in the same language, most likely Latin. Today, however, we live in situations where many, many people from different nations interact and therefore learning new languages has become imperative.

Not all of us are successful at it, however, so certain questions arise again and again when addressing the subject of second language (L2) learning. Is it more important to learn a language early or to have a lot of exposure? What is the main determinant explaining why some people are better than others at learning an L2? Is there a critical period for acquiring an L2? This latter question, which is of obvious theoretical importance, turns out to be quite controversial.

One way of describing the ability of non-native speakers is to insist on the fact that there is no evidence that anyone has ever mastered an L2 to the same degree as a native in all different domains. While this claim may be true, we can look at it from a different perspective, since the same statement seems to suggest that there may always be something that can be learned at a native level in an L2 (Birdsong 2004; Marinova-Todd 2003) – a case of the glass being either half-empty or half-full. At any rate, it is quite clear that not all aspects of second languages are equally easy to learn. Vocabulary is relatively easy, for instance, but we all know people who, despite living in a foreign country and having had years of exposure and opportunities to learn the language, still have very strong accents and a tendency to make particular mistakes. Conversely, we also know of people who move to another country and very rapidly are able to speak like natives – to the envy of most of the rest of us. The question then is, why are some people so poor at it and others so good?

One popular explanation is the importance of
age of acquisition
. Clearly, learning a language early in life increases the likelihood of doing well in that language. A second classical explanation is
amount of exposure
. Age of exposure will not ensure good learning if amount and quality of exposure are insufficient. A third explanation often given is
motivational factors
: motivated learners acquire better new skills. And then we come to the tricky question of talent. We know that some people have an “aptitude” for language, but what exactly is “aptitude” or “talent”?

Today, neuro-imaging techniques are beginning to provide new insights into this question, and this is what I would like to focus on for the remainder of this talk. I am going to present the results of different types of brain imaging studies that have tested L2 learners in a wide variety of situations, and explain some of the brain areas that have been found to be different between “good” learners and “poor” learners. We are going to examine two different types of evidence. The first type will present data from structural studies. In these studies differences in brain structure between different populations are analyzed. In particular, the brains of good and poor perceivers are compared using different techniques. The second type is activation or functional studies, examining which brain areas are activated while doing a particular task. All of these studies are very recent and more data is needed, but they nevertheless point in a direction that is very suggestive, albeit premature.

Mechelli and coworkers (2004) addressed the issue of whether differences in brain structure could be found as a function of age of acquisition and as a function of final attainment (proficiency). For this study, the authors chose individuals whose L1 was Italian and who learned an L2 (English) between the ages of 2 and 34 years. The way they assessed competence in the L2 was
through a battery of standardized neuropsychological tests. Participants were tested in their reading, writing, speech comprehension, and production skills (the typical neuropsychological tests) and a global L2-proficiency score was computed. Using voxel-based morphometric analyses, Mechelli et al. were able to observe that the more proficient L2 learners had more grey-matter density in the left inferior parietal cortex than poor learners. It was also observed that the density of grey matter in that particular area was also a function of age of acquisition in the L2. Late learners had less grey matter density compared with early learners.