Just the Facts

Photo: Well, at least the whites of their eyes are still white. (An important point on this blog, see e.g., Savanna Eyes)

Last week’s New Yorker carried an article by John Colapinto (“The Interpreter,” photo summary here) that challenges every view of human nature that says something absolute about language.

• You think we can all be reached by God’s word as given to us in the gospels? The Pirahã (pee-da-HAN) language may not be given to generalities and is therefore incapable of expressing timeless values. In fact, they have no mythology of their own to be converted from.
• How about recursion? Are you with Chomsky in believing that all languages can embed phrases to integrate thoughts? The Pirahã are reportedly unable to say, “I saw a dog down by the river get bitten by a snake.” They would have to say: “I saw the dog.” “The dog was at the beach.” “A snake bit the dog.”
• So then, what about joint attention? This blog has championed the notion that speech is a form of shared attention, but, says the New Yorker, the Pirahã don’t point, which is the primary action for sharing attention.
• Phonology? The Pirahã language doesn’t depend on syllables.

Confronted by such a bizarre set of claims, one’s natural reaction is to say, this cannot be. After all, who knows what the truth about the Pirahã might be. They are a small tribe (300 people) living beside one of those remote little tributaries of the Amazon. Very few people from outside the Amazon have ever been in their country, and apparently nobody until the present generation of outsiders has ever claimed to understand their language. The New Yorker article describes several outsiders in Pirahã country—most notably the author and Tecumseh Fitch (a leading scholar studying the evolution of language) but they are dependent on a linguist named Dan Everett for an explanation of what is happening in front of their own eyes.

Everett has been studying the Pirahã for decades and published a long article about them in Cultural Anthropology back in 2005 (article here). In his introduction, Everett writes:

This paper … suggests that Pirahã culture severely constrains Pirahã grammar in several ways, producing an array of otherwise inexplicable “gaps” in Pirahã morphosyntax. These constraints lead to the startling conclusion that Hockett’s design features of the human language, even more widely accepted among linguists than Chomsky’s proposed universal grammar, must be revised. With respect to Chomsky’s proposal, the conclusion is severe—some of the components of so-called core grammar are subject to cultural constraints, something that is predicted not to occur by the universal grammar model. … these apparently disjointed facts about the Pirahã language … ultimately derive from a single cultural constraint in Pirahã, name, the restriction of communication to the immediate experience of the interlocutor. [p. 622; italics his]

This constraint would seem to limit communication to joint attention at its most fundamental, pointing to things in the immediate environment without drawing attention to any internal or ceremonial features of an experience. However, the claim sounds a tad naïve since culture is so critical to what we do experience.

The New Yorker article did say that the Pirahã don't point, so I was relieved to see in the actual article that they do:

… they tend not to point with individual fingers, at least when talking to me. Commonly, if they use any part of their arms for pointing, they tend to extend a flat hand turned sideways or an open palm facing up or down. More often, they point as is common around the world, with their lower lip or jaw or a motion of the head. [p. 624]

Whew! Yet I expect speakers to draw attention to more than the empirical world available to computers. Just a couple of weeks ago, this blog discussed three types of sentences whose evolution reflects a break from the here and now (see: The Rise of Mystery). The sentences were:

The wind blew out the fire. This statement of immediate observation appears to be possible among the Pirahã, even if the statement is a report and not a comment on an event that has just happened.

My childhood has gone with the wind. “My childhood” is a generalization of experiences and thus seems to be outside the range of things the Pirahã can talk about.

The Holy Spirit came unto me like a great wind and my soul was transformed. Not only do we get two experiences in one sentence (a no-no for Pirahã speakers), but these refer to a world outside immediate experience. And yet Everett does report that the Pirahã believe in a spirit world “that they [claim to] have directly experienced” (p. 622). The fact that they can speak of a spirit world makes the constraint of “immediate experience” a bit more vague than it originally seemed. There are many people for whom the sentence about the Holy Spirit would be said to reflect immediate experience, as anybody who has ever read William James or Marilynne Robinson will know.

The idea that there is a people who never speak at all three of these levels seems incredible, especially if they do speak of a spirit world. Everett does mention marriage in Pirahã society (p. 632) and I consider a system that ceremonially alters the possibility of sexual relations to be one of the ways in which culture imposes itself on concrete reality. I did not see anything about “coming of age” in Everett’s piece, but such ceremonies are another way culture changes concrete identities. I'm going to have to get confirming reports before I believe that the Pirahã never tangle cultural abstractions with their immediate experience.

Another problem in dealing with Everett’s work is that he is trying to explain “gaps,” things unobserved. It leaves him open to the suspicion that he has missed something. Observations demand an explanation; lack of observation may demand an explanation or merely the rejoinder, “Look harder.” The focus on gaps has another weakness, like the legendary job applicants who tells us all the things they refuse to do, we are left wondering about what they do do. The New Yorker article referred to late night chatter in the village, but we are not informed what they were talking about. Can it really be that the Pirahã were up all night talking about immediate experience?

But the very discomfort caused by Everett, who after all has spent a great deal of time and effort among the Pirahã, should warn us not to dismiss the claims too cavalierly. When travelers return from the wilderness with strange tales, few of us can be sure whether they are liars, fools, or prophets. Often they are all three lumped together (who among us is not?) and it takes a lot of sorting to determine a report’s worth. Apart from the case of outright frauds (and there is no evidence to lump Everett in this class, and good reason to exclude him) these reconsiderations are worthwhile because of the way they force us to return to basics and think about them clearly. In Everett’s case, he is forcing a reconsideration of the relation between culture and syntax. He is going after the primal Chomskyan doctrine that there is a universal grammar, no matter what one’s cultural situation. Joining Everett in the commentary published alongside his article is Chomsky’s most articulate foe, Michael Tomasello. He offers the alternative view:

Because they [the Pirahã] talk about different things, different things get grammaticalized. [p. 640]

There will be much pointless wailing and some valuable observation before the question is resolved.

Comments

[TLTB]

There's a lot that can be said here, and your skepticism is warranted. Reading Everett's claims about the Piraha makes me think back to old discussions of "non-configurational" languages that supposedly had no word-order restrictions. Deeper investigation disproved that myth.

But if Everett's claims are right, its still preliminary to take the Piraha as evidence against the nativist claims about language. The Piraha are an isolated group with very limited outside interaction. Its possible that their linguistic aberration lines up with a genetic one as well. If Everett's claims can be confirmed, the next step is to do some DNA analysis and check out the FOXP2 and other language-related genes looking for mutations.

When I spoke with Fitch back in February, he said they don't have any conclusions yet, but what he's mostly interested in is the fact that the Piraha don't have any numbers. What he wants to know is, do they merely lack words for numbers or do they actually lack the concept of numbers/counting. Are they like non-human primates (which we know can only distinguish groups of one, two, three, and more than three) or can they count higher? This bears on the linguistic stuff as well since Fitch, Hauser, and Chomsky have supposed that the recursive property of language that gives us infinitive sentence length is the same as the recursive property of numbers which gives us infinite counting ability.