Don't Sleep, There Are Snakes (32 page)

Without syntax, Pirahã’s grammar would lack verb phrases, noun phrases, embedded sentences, and so forth. In fact it does seem possible to interpret all Pirahã sentences as beads on a string, with no need for more complex structure of the type that phrase structures would predict. A sentence would be simply the list of words needed to complete the meaning of a verb, plus a minimum of modification, usually no more than one adjectivelike or adverblike modifier per sentence. Pirahã would lack syntax, in my rather extreme view, to guarantee that nonassertions do not appear within declarative sentences, in violation of the IEP. The IEP allows declarative clauses to contain only assertions. Therefore the IEP constrains the grammar of Pirahã.

Take the original relative clause sequence I overheard from Kóhoi: “Hey Paitá, bring back some nails. Dan bought those very nails. They are the same.” There are two assertions here,
Dan bought the nails
and
the nails are the same.
But in the English relative
the nails that Dan bought,
there is no assertion. So the immediacy of experience principle is violated.

If my reasoning here is on the right track, what other predictions does it make about Pirahã grammar? Exactly the right ones, it turns out.

It predicts that Pirahã will lack coordination, because this also involves the general property of recursion, something that has been eliminated from the grammar of Pirahã, as we have already discussed, in order to avoid having embedded nonassertions in declarative assertions. Coordinate structures are of course common in English and many other languages. Their recursion is shown in examples like
John and Bill came to town yesterday,
where the noun
John
and the noun
Bill
both occur in the longer noun phrase
John and Bill.
Coordination of verbs and sentences is also ruled out, so Pirahã has no sentences like
Bill ran and Sue watched
or
Sue ran and ate.

The IEP’s restriction against recursion also correctly predicts that Pirahã will lack disjunction, as in
Either Bob or Bill will come, I had some white meat, chicken or pork,
and so on. Pirahã lacks disjunction because it, like coordination, requires putting phrases inside of other phrases—recursion. The Pirahãs would say, for example, rather than “Either Bob or Bill will come,” something like “Bob will come. Bill will come. Hmm. I don’t know.”

These do not exhaust the predictable consequences of the lack of recursion in Pirahã. Other predictions are being tested now by a range of psychologists and anthropologists. This is interesting because the fact that there are testable predictions of the immediacy of experience principle shows that it is not simply a negative statement about what Pirahã lacks, but a positive statement about the nature of Pirahã grammar and how that grammar differs from many well-known grammars.

The statement is positive because Pirahã imposes and enforces a cultural value on its grammar. It is not, again, simply that Pirahã accidentally lacks recursion. It doesn’t want it; it doesn’t allow it because of a cultural principle.

In addition to the Pirahãs’ grammar, the IEP helps to account for the other gaps in the language we have already discussed, such as the absence of number and numerals, the absence of color words, the simplicity of the kinship system, and so on.

The prohibition against abstractions and generalizations of the immediacy of experience principle is a very narrow prohibition. It by no means entails that Pirahã culture prohibits abstract thought. It is also not a prohibition against all abstractions or generalizations in the language. For example, the Pirahãs have words for kinds or categories of things, as all languages do, and these words, usually nouns, are by definition a type of abstraction. How is this seeming contradiction tolerated in Pirahã?

Grammar had once seemed to me too complicated to derive from any general human cognitive properties. It appeared to cry out for a specialized component of the brain, or what some linguists call a language organ or instinct. But such an organ becomes implausible if we can show that it is not needed because there are other forces that can explain language as both ontogenetic and phylogenetic fact.

Like most linguists today, I once believed that culture and language were largely independent. But if I am correct that culture can exercise major effects on grammar, then the theory I committed most of my research career to—the theory that grammar is part of the human genome and that the variations in the grammars of the world’s languages are largely insignificant—is dead wrong. There does not have to be a specific genetic capacity for grammar; the biological basis of grammar could also be the basis of gourmet cooking, of mathematical reasoning, and of medical advances. In other words, it could just be human reasoning.

On the evolution of grammar, for example, many researchers have underscored the fact that our ancestors had to talk about things and events, about relative quantities, and about the contents of the minds of fellow members of their species, among other things. If you can’t talk about things and what happens to them (events) or what they are like (states), you can’t talk about anything. So all languages need verbs and nouns. But I have been convinced by the research of others, as well as my own, that if a language has these, then the basic skeleton of the grammar largely follows. The meanings of verbs require a certain number of nouns, and those nouns plus the verb make simple sentences, ordered in logically restricted ways. Other permutations of this foundational grammar follow from culture, contextual prominence, and modification of nouns and verbs. There are other components to grammar, but not all that many. Put like this, as I began to see things, there really doesn’t seem to be much need for grammar proper to be part of the human genome, as it were. Perhaps there is even much less need for grammar as an independent entity than we might have once thought.

Although a language might possess a strong cultural constraint, like the immediacy of experience principle in Pirahã, such constraints cannot override general forces and results of evolution nor the nature of what it means to communicate. Nouns and certain kinds of generalizations are part of our evolutionary heritage and cultural principles can not override these, even though they do show that culture and grammar are intimately connected.

On the other hand, research is ongoing. The issue of whether Pirahã does or does not have recursion is far from settled. But the evidence being gathered and interpreted by independent researchers is consistent with my conclusions.

One phenomenon that has attracted my attention since the beginning of my reflection on the possible connections between grammar and culture is that theories, regardless of how useful they are in many ways, can impede novel thinking. Our theories are like cultures. Just as there are gaps in Pirahã culture for counting, color words, and so on, some theories can have gaps where other theories might have robust explanatory mechanisms. In this sense both theories and cultures shape our minds’ ability to perceive the world, sometimes positively and sometimes not so positively, depending on the goals they set for themselves.

What are the implications for Pirahã grammar if it lacks recursion?First of all, a lack of recursion in its grammar would mean that its grammar is not infinite—there would be an upper limit to the number of sentences generated by it. But this would not mean that the
language
is finite, because recursion is found in Pirahã stories—parts of stories are built up, there are subplots, characters, events, and all sorts of relationships among all of these. This is interesting because it means that the role of grammar in the infinitude of a language is not so important—you can have a nonfinite language with a finite grammar, something Chomsky’s recent theory on the importance of recursion can neither accommodate nor elucidate. This would also entail that you could specify the upper size of a particular sentence in that language, though not the upper size of a discourse. That sounds bizarre for a language. Some linguists or cognitive scientists might even leap to the conclusion that an absence of a recursion could render a language deficient in some way. But this would not be correct.

Even if a language’s grammar is finite, this doesn’t mean that the grammar is not rich or interesting, however. Think of something like chess, which has also got a finite number of moves. This finitude of chess moves doesn’t have much practical effect, though. Chess is an enormously productive game that can and has been played for centuries. The fact that chess is finite tells us very little about its richness or its importance. The Pirahãs discourse is rich, artistic, and able to express anything that they want to say within their self-imposed parameters.

So if there were a finite grammar, this wouldn’t mean that the grammar was spoken by abnormal humans, nor would it mean that it was a poor source of communication. It wouldn’t even mean that the language with that grammar was finite. If there were such languages, however, where and under what conditions might we expect to find them?

If you build into the foundation of your theory the constraint that all grammars are nonfinite and that they must therefore be recursive, the absence of recursion will elude you. Your theory will have hindered you, just as my culture’s lack of experience with dangerous animals outside of zoos could render me easy caiman prey.

On the other hand, if our theory doesn’t require that recursion be a crucial component of language, where does recursion come from? No one can dispute that it is found in most human languages. Nor can anyone seriously doubt that it is found in all human thought. My view is that recursion comes from the brain’s general cognitive powers. It is part of how all humans think—even when it is not part of the structures of their languages. Humans have recursion because they are smarter than species without it, though recursion could be a cause or an effect of this greater intelligence—no one knows right now, regardless of what claims are found in the literature.

In fact, as we saw earlier, Herbert Simon claimed almost exactly this in “The Architecture of Complexity.” As I noted, in this article he argued that recursive structures are fundamental to information processing and that we use them not just in language, but also in economics and problem solving.

And recursion is crucial in almost all of the stories that we tell. Surprisingly, human discourses have never been the subject of Chomskyan research. But as we just saw, this is a massive oversight, since recursion can be found outside the grammar, tremendously reducing the importance of the grammar in understanding the nature of language and communication. Discourses are purposely ignored by Chomsky as social or at least nonlinguistic constructs. Yet when we examine the stories that the Pirahãs tell, we find recursion, not in individual sentences, but in the fact that ideas are built inside of other ideas—some parts of the story are subordinate to other parts of the story. Such recursion is not part of the syntax proper, but it is part of the way that they tell their stories.

We might propose, following Simon, that recursion is absolutely essential to the human brain, and that it derives from the fact that humans have larger brains or brains with more convolutions than other species. Ultimately, though, it’s not even clear that recursion is unique to humans. And it most certainly is not clear that recursion is part of grammar, rather than coming into languages because it is a useful, preexistent cognitive tool.

The crucial application of Simon’s proposal for studies of language is that hierarchical structures found in languages that have so long been the focus of Chomskyan research are “emergent” properties, rather than basic properties of language. That is, they appear in languages in response to the interaction of the brain’s ability to think recursively and problems or situations in the culture or society that are more efficiently communicated recursively.

If recursion is proposed to be, as Chomsky and many of his followers would have it, the core faculty of the human language capacity, and if recursion is absent in one or more languages, then the Chomskyan proposal is falsified. But if recursion is not the core faculty, then Pirahã suggests that the kind of theory of language we need is not one in which language is an instinct. Instead, we might be better served by looking at syntax, along with the other components of language, as one part of the solution to the problem of communication, that is, the need to communicate appropriately in a specific environment.

I doubt Pirahã is the only language that will challenge our thinking about recursion, human language, and the interaction of culture and grammar. If we look at other groups—groups in New Guinea, Australia, and Africa—we are likely to find similar cases of esoteric communication and societies of intimates that could lead to the lack of recursion. Esoteric communication could very well contribute to our explanation of the some of the more controversial aspects of Pirahã grammar.

The usefulness of the concept of esoteric communication for understanding Pirahã is demonstrated in part in current research by the psychologists Thomas Roeper of the University of Massachusetts and Bart Hollebrandse of the University of Groningen. This research suggests that recursion might be a device that is useful for packing sentences with more information in societies with a higher degree of exoteric communication where more complex information is the rule, such as modern industrialized societies. But in a society like the Pirahãs’, the esoteric nature of their communication renders recursion less useful, while the immediacy of experience principle is incompatible with it.

What we need to look for are groups that have been isolated, for various reasons, from larger cultures. The Pirahãs’ isolation is due to their very strong sense of superiority to and disdain for other cultures. Far from thinking of themselves as inferior because they lack anything found in other languages and cultures, they consider their way of life the best possible way of life. They’re not interested in assimilating other values. So we see little seepage from other cultures or languages into Pirahã. And these are the kinds of culture-language pairings we need to be studying.