Evolution of revolution
Pacitti talks to Noam Chomsky
Your first classic work Syntactic
Structures transformed the study of linguistics back in 1957.
Prior to that, traditional grammars had presupposed the
intelligence of the reader in their explanations. But as soon as
the attempt was made to account for this intelligence, a number of
serious problems arose. Almost two and a half decades of
subsequent inquiry tried to solve these problems by extracting
rules to account for one’s innate knowledge of grammar. All of
this came to a head in 1981 with your “principles and
parameters” approach based on lectures you gave in Pisa. And
this rather cut the Gordian knot and paved the way for your
current and equally revolutionary “minimalist” programme. Is
Yes. The principles and parameters approach said: Look, there are
no rules – earlier theory was wrong. There are no constructions.
There’s no passive construction in Italian or verb phrase in
Hungarian: these are taxonomic artefacts. There are no rules for
forming relative clauses: that’s not the case. But what we
actually have is quite general principles which are fixed and part
of the genetic endowment. We have certain options – or
parameters – which have to be set, so that the child comes into
the world having incorporated principles and knowing that these
are the choices and when he has to set the choices. So, do I speak
a language where the verb precedes the noun or where the verb
follows the noun? Well, those kinds of choices you can make simply
on the basis of elementary data. If the choices are fixed, the
language is in place. It’s as if the genetic instructions
determine a fixed network of principles and a certain few switches
that aren’t set yet, and the child has to set the switches on
the basis of simple data. Once the switches are set, the whole
system operates. And the lexicon works the same way. Roughly in
this framework, more has been learned about language in the last
20 years than in the preceding two thousand. There’s been an
enormous amount of research in depth and across extremely varied
languages to try to show the fine system of this sort.
framework also allows you to think about some new questions. For
the first time in history you have what you might call a genuine
theory of language, something that can deal with both the problem
of accurate description and the problem of acquisition (known
technically as descriptive and explanatory adequacy respectively).
You can deal with both without contradiction and you can now ask
lots of questions that you couldn’t contemplate before. One
question has to do with optimal design. No matter what you’re
discussing, you’re going to have the worst, most complicated and
impossible badly designed system there is. You always want to
construct the best theory of it. But there’s another question:
How well designed is the object itself? How well does the object
itself satisfy the minimal conditions that it must satisfy to
function? That’s a different question. You might be able to
raise that question about language. So, to what extent is the
language organ optimally designed to serve minimal functions?
That’s a new question.
we can give an answer. The language faculty has to interact with
other components of the person and we know what some of those
components are. We also have thought systems, conceptual systems,
and the expressions of the language faculty have to be accessible
to those, otherwise we can’t express out thoughts. We can call
these ‘interfaces’ between the language faculty and other
systems. To a non-trivial extent we can specify the properties of
the interfaces and we can then ask a concrete question. The
minimal condition the language condition must meet
is what we can call the ‘interface condition’. The information
it presents must be accessible to the external system. The
question is: Is that also a maximal condition? That is, is the
language faculty optimally constructed to satisfy that minimal
condition? When you pursue this question, you’re pursuing the
few years ago it seemed hopelessly crazy, but now there’s
already been enough work to indicate to a rather surprising extent
that it may be correct that the language faculty is
an optimal solution to the minimal conditions. It’s as if an
engineer inserted a language faculty into a brain that didn’t
have one and did it in an optimal way so that it would be
accessible to the other systems. When you try to spell this out,
then we get back to the first challenge: the accuracy of the
complex descriptive devices that are used in successful accounts
of language, because they’re not optimal, so you have to try to
show that they don’t exist and that if you eliminate them,
you’ll get as good results or maybe better results. And that is
what the work over the past few years has been – to try to take
point by point the devices that are used in linguistic description
and explanation and to show that they’re unnecessary,
superfluous or incorrect, and that we can get better results by
eliminating them. That’s a very hard research programme, but I
think it’s had some interesting consequences and a lot of
indication that it’s true.
EFL teachers may be wondering why it is that the results of your
research programmes in linguistics haven’t filtered through to
language teaching as much as one might have expected. Somehow the
considerable insights achieved in theoretical linguistics still
seem far removed from the concrete business of teaching foreigners
NC: How much have the results of the very interesting studies in the cognitive sciences, for example object perception, penetrated? How much do they extend? Not much. How much does the study of physiology enter into teaching people how to swim? I mean, it’s probably a good idea for a swimming coach to know something about the body, but not much.
What we do is so far beyond what we understand that the
understanding is useful but it’s rarely going to be helpful in
the applied world. I mean, take even the hard sciences. Until the
mid-19th century physics and chemistry had almost
nothing to say to engineers. And these are subjects that have a
rich, deep history of achievement. It takes a long time for them
to begin to get to the point where they teach people things.
Then it will take some time before your linguistic theories filter
down to EFL teaching?
I mean, what we know intuitively is far beyond what we can
There are some language purists at Florence’s authoritative
Accademia della Crusca who fear that linguistic degeneration has
set in owing to the infiltration into Italian of anglicisms such
as “fast food”, “trust”, “trend”, “spot”,
“spray”, etc. The fear has even been vented that within a
hundred years English will quite literally have eliminated Italian
altogether. How does this strike you?
Well, it’s rather curious for Italian scholars to say this. Just
look at what’s happening in Italy. A hundred years ago, fifty
years ago, Italian wasn’t the language. Everybody’s
grandmother spoke some other language, but what happened to all
those languages? They were eliminated by Italian, which was the
language of Dante. And that has essentially wiped out a whole
complex range of languages in Italy. Just in our lifetimes there
has been a huge destruction of languages right here in Italy.
Well, I don’t think it’s a good thing. They’re dying because
they’ve been taken over by Italian, which became a national
language. Italian has just wiped out these languages from the
peninsula. The disappearance of languages and cultures is like the
destruction of bio-diversity, except more important to us because
it is human diversity. It’s cultural diversity and is being
destroyed, and that’s a bad thing.
I believe your parents emigrated from Eastern Europe to the United
States and was wondering where exactly they came from.
My father came from the Ukraine and my mother came from what’s
I myself am third generation Italian and have come to feel that I
still have some marked Italian characteristics. Don’t you think,
setting aside questions of genetic preprogramming or philosophical
determinism, that it’s natural to feel this? Do you ever feel
this yourself with regard to your own Russian background?
You’re not Jewish, I take it?
No, I’m not.
It’s quite different. Jews are a tribe, and the place where they
are historically doesn’t matter very much, like my parents, who
were immigrants, who would never talk about Eastern Europe. They
just didn’t want to talk about it. They wanted to forget Russia
They didn’t want to know about it, they didn’t want me to know
about it. Later, when I was older, I got interested and I would
ask my father, but it was just not part of our childhood. Their
native language was Yiddish, but they never let us hear a word of
Yiddish. My wife was the same, same background, and neither she
nor I know a word of Yiddish except what we learnt through our
grandparents. Until very recently Jews were outsiders. There’s a
famous Jewish joke that asks why there are so many Jewish
violinists and not many Jewish pianists. Well the answer is: You
can pick up your violin and run.
Yes, that certainly makes the point.
It kind of captures the thing. Look, when I got to Harvard in the
early 1950s, it was very anti-Semitic. It wasn’t until the late
1950s that these barriers fell. It’s a very recent phenomenon.
And the same applies to Britain and also to continental Europe.
Jews were respected, but they weren’t really part of the elite
culture until very recently.
I think it requires some effort for non-Jews to understand all of
It’s just kind of part of your background, you know. That I
could become a Russian is for me simply quite foreign and
Note: This interview took place at the Certosa di Pontignano on November 9 1999 and was first published in full by JUST Response on May 20 2002. A shortened version appeared in the January 2000 issue of EL Gazette.