Don't Sleep, There Are Snakes (34 page)

I thought that this was all going pretty well. No problem with bringing Pirahãs to the city after all. I wondered what I had been concerned about. True, I did find curious the Pirahãs’ insistence on walking single file through town, just as they did through the jungle.

As we strolled down the city’s sidewalks, Xipoógi walked behind me, with Xahoábisi behind him. I slowed down to let them catch up. They slowed down too. I slowed down more. Ditto. I stopped. They stopped. They simply would not walk beside me, not even when I asked them to. This makes sense on a narrow jungle path. There isn’t room, unless you double the back-wearying work of clearing a path in the jungle by making a path wide enough for two to walk side by side. And it would be unsafe to do this anyway. People walking abreast make a bigger target for predators and offer each other less protection against snakes and other dangers. In the city, though, walking abreast, while spatially inefficient, allows the walkers to converse more easily and to be perceived as a group. I smiled about our walking arrangement and waited at the stoplight for the signal to walk across the street, Pôrto Velho’s busiest.

I said to Xipoógi and Xahoábisi as I led the way, “Follow me. We’re going to that store there,” pointing toward a grocery store across the street.

Three-quarters of the way across the street, I looked back and Xipoógi and Xahoábisi were frozen in fear, looking at the cars that stood perpendicular to them on the road, waiting at the stoplight, revving their motors. I started back toward them but the light changed. The cars started moving forward and honking vigorously at the two Pirahã men, now quite visibly frightened. They were clearly on the verge of panicking and running through the traffic, unable to predict the movements of the cars, so different from any wild animals they had ever encountered. I got to them and took their hands, leading them back to my side of the street. We made it to the sidewalk.

“Those things scare us,” they exclaimed, still not over their tension.

“They scare me too,” I agreed.

“They are worse than jaguars,” Xipoógi concluded.

T
he debate raging around Pirahã, again, is whether it forces us to rethink major theories about language or culture. Chomsky, founder of the most famous and influential of modern theories of language, says that languages with the properties I have described for Pirahã do not exist—that Pirahã is pretty much like any other language. But to understand why his own theory leads him to this state of denial, we need to know more about that theory.

Chomsky’s view is that he is trying to discover the “true theory of universal grammar,” where the latter is proposed to be a language-specific component of our biological endowment. It isn’t clear what Chomsky means by the “true theory,” but I assume that he means the one that completely matches reality (it is hard to know what most scientists and philosophers mean when they use the word
true,
so this problem is hardly unique to Chomsky). This is worth thinking about more carefully. At one level universal grammar seems to be almost a necessary concept—after all, neither plants, rocks, nor dogs talk—only humans do. We all agree there is something about human biology that underlies language. In this sense, Chomsky is trivially right. But the real question is how specific this endowment is to language (as opposed to, say, the proposal that our capacity for language just follows from general cognitive properties) and how much whatever this biological endowment is determines the final form of the grammar of any specific human language. And a related question, one that might initially seem tangential, is how we as scientists come by the knowledge to test our hypotheses.

There are two typical research locales in science—the laboratory and the field. The so-called hard sciences, such as physics and chemistry, as well as most of the social sciences, are done in climate-controlled, comfortable rooms, furnished with the equipment the researcher needs for his or her work. As pursued in wealthy countries like the United States, Germany, England, or France, science is done by a privileged few for society as a whole. At least on paper, its sponsors expect results to benefit the larger society in which the science is supported and takes place. Young scientists work under the security umbrella of an established leader in their field. In linguistics, Chomsky is a Daniel Boone figure, and the majority of linguists are settlers on his land.

Linguistics has changed over the decades. It was at one time more like the “field sciences,” those branches of inquiry such as geology, anthropology, and biology, wherein learning entails leaving the laboratory for the rough world of fieldwork. Of course, many linguists continue to do field research on languages around the world.

But the explosive growth of linguistics after the advent of Chomsky in the 1950s has altered the ethos of the discipline in profound ways. Chomsky’s attraction for many linguists, including me, is the elegance of his theory, not field research. The lemmas and axioms first given in his breakthrough work,
The Logical Structure of Linguistic Theory,
written while he was in his twenties, and the subsequent books such as
Syntactic Structures, Aspects of the Theory of Syntax, Lectures on Government and Binding,
and
The Minimalist Program,
convinced generations of linguists that Chomsky’s theory was likely to lead to significant results. Like many others, I read all of these books cover to cover. I have taught graduate courses on most of them.

The culture of Chomskyan linguistics also spread because his department at MIT attracted some of the best students in the world. This new linguistics culture brought enormous changes in linguistics methodology as well as goals, another defining feature of the Chomsky group. Prior to Chomsky, to be an American linguist almost obligatorily entailed one or two years of living among a minority language community and writing a grammar of their language. This was nearly a rite of passage in North American linguistics. But since Chomsky himself did no field research and apparently had learned more interesting things about language than any fieldworker, many students and incoming professors working under the influence of Chomsky’s assumptions understandably believed that the best way to do research might be to work deductively rather than inductively—from the institution rather than from the village, starting with an elegant theory and predetermining where facts best fit.

Here is my understanding of these ideas. An inductive approach to the study of language would be to allow each language to “speak for itself.” We might do this by cataloging the observations of the language made by the field researcher and then working out a narrative of just what the elements of that language are (its words, phrases, sentences, texts, or however the field researcher cares to label them—whatever he or she finds most useful in discussing this particular language) and then how these elements fit together (such as, how do speakers of this language form sentences or paragraphs or whatever units they do form, and how do they use these to construct conversations, stories, and other forms of sociolinguistic interaction).

A deductive approach, on the other hand, begins with theories—prelabeled boxes—and fits aspects of language into them. New boxes can be made, but this is frowned upon. Much of the debate in deductive theories is about the aspects and borders of the boxes. The cultural values that have gotten a grip on linguistics partially as a result of Chomsky’s deductive approach to language study should not be ignored. These include at least the following: field research is unnecessary to be a good linguist; it can be as important to study one’s native language as it is to study previously unstudied languages in the field; grammar is a formal system independent of culture.

In the twenty-first century our knowledge of the form and meaning of parts of language is claimed by some to far exceed prior knowledge. This follows from the concept of scientific progress and the notion that we build “precept upon precept” upon the knowledge of our predecessors, in what Mortimer Adler, in his introduction to the
Great Books of the Western World,
refers to as the “great conversation” of life.

But there is a competing concept that many scientists believe in simultaneously: the notion of the scientific revolution. As developed in the work of the philosopher Thomas Kuhn, this is the idea that scientific theories can paint themselves into a corner and that scientists will then be trapped unless someone blasts a hole in the building and proclaims freedom to do science outside the confines of the previous approach. This blasting takes place as recalcitrant facts begin to accumulate, facts that a particular theory can only handle with a lot of patchwork and tearing and straining—what Kuhn calls “auxiliary hypotheses.” Pirahã presents numerous recalcitrant facts, as do languages like it—and I have no doubt that more will be discovered. These facts require a good hole in the wall leading to a new theoretical edifice. This is what Pirahã says to me about the prevailing theory.

Our attempts to study other human beings can be as culturally influenced as my attempt to get the Pirahãs to walk side by side with me in the city. And culture not only affects the scientist-observer, but the subject under study as well. To understand theories about human languages requires us to consider the influence of culture on theory construction as well as the role of culture in shaping the object under study.

This is a controversial point. In one well-known book, Steven Pinker’s
The Language Instinct,
little importance is attributed to culture in shaping human grammar. True, Pinker allows that culture bears significant responsibility for the things we talk about (so Americans of a certain age may compare Marlon Brando and Elvis Presley for relative sexiness or star power, or the influence of Google on research in modern American society; the Pirahãs, on the other hand, are more likely to talk about encounters with jungle spirits or the best way to catch bass).And cultures also determine vocabularies. In Scotland we encounter the word
haggis.
The ingredients of haggis are (usually) sheep’s “pluck”(heart, liver, and lungs), minced with onion, oatmeal, spices, suet, and salt, mixed with stock, and traditionally boiled in the animal’s stomach for approximately three hours. I like it. But it isn’t for everyone. And it is only a traditional dish in Scotland. We are not surprised that the Scots have a word for this traditional part of their culture.

Another example is the Brazilian word
jeito
(ZHAY-tu), which literally means “to lie” or “to rest” and refers to a Brazilian concept of themselves as having a special knack or skill for solving problems. It is common to hear Brazilians say, for example,
“Nos brasileiros somos muito jeitosos”
(We Brazilians are good at
jeitos
). This skillfulness and the talking about it among Brazilians is a cultural value. It is neatly expressed as a single word by the members of Brazilian culture—the culture in which talking about the concept is so important. It is therefore another example of culture and language working hand in hand.

And, of course, the Pirahãs have words like
kaoáíbógí
(fast mouth) for a kind of spirit unique to them.

But there is no role attributed to or even allowed for culture in shaping the grammar proper in most linguistic theories. This is why it is important to study languages like Pirahã, in which culture appears to shape grammar in ways that few theorists imagine possible.

We can begin to appreciate Pirahã’s relevance to our understanding of the nature of human language by considering one of Chomsky’s main concerns, the explanation of similarities among languages.

When we look at the world’s languages, we see many similarities; so many and so recurring in fact are these similarities that we know that they cannot be merely coincidental. We are obliged as scientists in the tradition of Western culture to offer an explanation for these.

Chomsky has urged us to place the explanatory locus for these similarities in genetics, and this is a reasonable place to look for explanations. After all, it is our common genome that unites
Homo sapiens
into a single species and produces other similarities among us, including many of our needs, desires, common experiences, and emotions.

So Pygmies and Dutchmen may look very different, but their similarities far outweigh their differences, because they, like all humans, come from the same genetic base. Without an understanding of evolution and genetic explanations, the nature of our species would elude us. So it is worth thinking about some of the similarities across languages that genetics might explain.

First, it could explain why languages all have similar parts of speech (verbs, nouns, prepositions, conjunctions, etc.). It may turn out that not all languages have each of the entire set of possible parts of speech, but what any one language does have, so far as we have seen, is like the categories of other languages.

Or it might explain why languages have similar psycholinguistic constraints on the processing of sentences (which is why in any language a perfectly grammatical sentence with the structure of
Oysters oysters eat eat oysters
can be hard to understand). The problem with this example is that it has “center-embedding”—a clause (the relative clause
oysters eat
) stuck in the middle of another clause (the main clause
Oysters eat oysters
). We can make the example easier to understand by inserting a marker to show us where the embedded clause begins, as in the much-easier-on-the-ears
Oysters that oysters eat eat oysters.

And languages share similar constraints on meaning. For example, there is no verb in any language we know of that requires more than three nouns to complete its meaning (some theories say no more than four nouns). A language can allow verbs to appear with no nouns, or just nouns that do not refer to anything, a sort of placeholder noun. An example is the
it
of
It rains.
Or there can be verbs that appear with only one noun, as in
John runs;
or with two nouns, as in
Bill kissed Mary;
or even with three nouns, as in
Peter put the book on the shelf.
But not more.
John gave Bill the book Susan
is not possible. To say something with four or more nouns we need multiple verbs, multiple sentences, or prepositions, as in
John gave the book to Bill for Susan.
^*