Origins of Human Communication (Jean Nicod Lectures) [Paperback]

Michael Tomasello is perhaps the consummate product of contemporary sociobiology. A tenet of sociobiology is that we can understand the behavior of a species by comparing and contrasting with closely related species, and with species that have found similar means for solving their social problems. Tomasello offers us deep insights into human communication and learning by comparing and contrasting our behavior with that of our nearest evolutionary relatives, the great apes.

Tomasello is not only a creative and incisive scientist, but also a learned intellectual, who is at ease bringing philosophical issues to bear on complex questions in behavioral science. In this book, we not only find out about human communication, but also are rewarded with an appreciation of the philosophy of the great Ludwig Wittgenstein. Wittgenstein is a difficult philosopher because he writes in quasi-aphorisms and follows something like the Socratic Method in asking questions rather than answering them. A good example is the quotation from the Philosophical Investigations with which Tomasello's book starts: "Point to a piece of paper. And now point to its shape---now to its color---now to its number... How did you do it?" Wittgenstein's point is that gestures as a form of communication are primary and part of our essence as humans. They are not translations of verbal linguistic structures into visual form. As Wittgenstein says in his unpublished notes (The Big Typescript, 2005[1933]), "What we call meaning must be connected with the primitive language of gestures."

Tomasello takes Wittgenstein literally: we share with chimpanzees and other apes the capacity to communicate by gesture, so it is likely that this was a capacity possessed by our most recent common ancestor. "My central claim," Tomasello writes (p. 2), "is that to understand how humans communicate...using a language...we must first understand how humans communicate...using natural gestures. Indeed, my evolutionary hypothesis will be that the first truly human forms of communication were pointing and pantomiming." Tomasello does not prove this thesis (if it can be proved at all), but rather uses the differences between gesture in apes and humans to develop a story of why humans catapulted so far beyond the apes in the use of communicative tools.

Tomasello's hypothesis from a careful study in contrasts between the role of gesturing in human and non-human primate societies is that "there must be some fairly specific connections between the fundamentally cooperative structure of human communication...and the especially cooperative structure of human, as opposed to other primate, social interaction and culture in general." (p. xi). Humans communicate because they want to help one another, he asserts, and a highly flexible system of communication is more helpful than a series of pants and grunts.

Tomasello gives a number of compelling examples contrasting human and other primate communication. Here is one: "When a whimpering chimpanzee child is searching for her mother," he writes, "it is almost certain that all of the other chimpanzees in the immediate area know this. But some if nearby female knows where the mother is, she will not tell the searching child, even though she is perfectly capable of extending her arm in a kind of pointing gesture" (p .5). Human communication, he argues, is a fundamentally cooperative enterprise, dependent upon deep commonalties in the consciousness of humans, including a common conceptual ground and cooperative communicative motives.

This human commonality is what Tomasello calls "shared intentionality": "The proposal is thus that human cooperative communication...is one instance...of a uniquely human cooperative activity relying on shared intentionality." At this point Tomasello relies on Gilbert and Searle, who are fine philosophers but whose theory of "collective intentionality" I think is completely without merit. Neither the conceptual arguments nor the empirical examples provided by these philosophers (and other of this school of thought) are compelling, and I believe that a combination of game theory, gene-culture coevolutionary theory, and the psycho-social theory of norms is a better starting point to understand how individual intentional agents succeed in coordinating their activities so as to produce human cooperation.

The problem with the theory of collective intentionality is that it depicts cooperation as a process in which all participants have no motives except for that of accomplishing the "team goal," whereas in fact this is rarely even remotely the case. The triumph of human cooperation is that people manage to coordinate their activities even though they generally have highly heterogeneous motives. For Tomasello, humans developed language because they want to help one another. However, it is not clear what fitness benefit comes from helping others, and the notion that we developed huge brains, complex voice boxes and its associated aural production physiology because we "want to help each other" is implausible. Indeed, among the more prominent predilections of humans is to lie, cheat, and attempt to free-ride on the prosocial behavior of others.

I find it more plausible to posit that human language flourished when humans became sufficiently adept at punishing social miscreants (gossip, shunning, beating, ostracizing) that it became plausible that communication would be truthful, on balance, and the detection and punishing of untruthful utterances could occur with high probability. Once the veracity of communication was ensured, it became possible to coordinate much more complex activities (e.g., warfare, conflict adjudication, hunting strategy, complex verbal agreements of intention), and individuals with the best command of language were afforded special privileges, including more and higher quality offspring, thus justifying the costs of developing the communication physiology.

I am not saying that Tomasello is wrong in stressing the humans have an inordinately highly developed propensity to help each other for purely altruistic reasons. They clearly do, and the predisposition towards prosocial behavior is one of the preconditions of human cooperation. However, the notion that language developed because people like to help each other and have a collective, or shared, intentionality is not a plausible basis for a theory of human communication. Human society is just more complicated than that.

This is one of the few books I've read that changed the way I think. The notions of "cooperative attention", "common ground" and "shared intent" are profound and leave me thinking and wondering. Let's see if I can describe one, common ground. As I see it, in a conversation participants share a framework, a "common ground". For example, the common ground may be the task "preparation for dinner". This framework permits many shortcuts. For example, common ground makes pointing an effective and informative means of communication. In the context of preparation for dinner, pointing to the salt may mean "please bring that to me so I can cook". Tomasselo argues that the 3 essential features derive from human social groups being inherently cooperative. An example would be a hunting party, where a single plan could be established and differential roles assigned. Tomasselo argues that ape societies are not cooperative, and disputes claims of cooperation from primate behavior research.

One of my interests is "Theory of Mind". Tomasselo argues (counter to his thinking 15 years ago) that apes do have a theory of mind. This means, roughly, that apes recognize that conspecifics (or humans) see the world differently than they do and are separate agents with separate goals and desires. The notion of "shared intent" rests on theory of mind. Tomasselo contends that apes do not have shared intent, due to the lack of natural cooperation in their social structure. Our presumed ancestors possessed both theory of mind and shared intent, necessary precursors for language development.

One of the most surprising Tomasselo ideas is that human language developed not from primate vocalization but from primate gesture and pantomime. He successfully demonstrates that primate vocalization is rigid, while pointing and pantomime have many key features of proto-language. He shows that pointing and pantomime serve important proto-language roles in 12 month old children. He is very convincing. The difficulty, for me, is how in our ancestors language made the jump from gestural to vocal. This seems a large jump.

I have two small quibbles with the book. One is that I would like to have heard a more direct account of the relation of "shared intent" and "theory of mind". Tomasselo is an expert on theory of mind -- he publishes reviews, etc, -- but the topic is not explicitly introduced. Second, as an outsider I would like to have a more explicit history of these ideas. For example, the contributions of Wittgenstein and Grise. While these writers are frequently cited with reverence, the history of these remarkable ideas is not presented.

This is a really good book. It is the best among the current evolution of language books. And it is an easy reading excursion into the attentional and gestural dynamics of great apes and children. Although intended for an academic audience, it still retains the feel of its origins in a series of lectures. For those who don't know Michael Tomasello, he is a quasi institution in his own right now in his academic niche at the cognitive, cultural, and linguistic intersection of the developmental abilities of apes and children. From his position at the Max Planck Institute in Leipzig, his publishing and research outputs are prodigious. In comparison to his earlier books this one is cleaner and covers more ground. Like other authors writing on the origin of language, the key issue is the origin and effects of cooperation. His new twist is to relax the altruistic nub of cooperation into observed cases of collaboration where participants happen to be focused on the same goal at the same time. In some ways the book restates classic Tomasello arguments, but they appeared clearer, more succinct, and better researched. The arguments on common ground backed up by the work of Herbert Clark (Using Language) made the analysis of cooperative situations make much more sense. His long chapter on syntax in an evolutionary context was a treat and well argued. Language change, common conceptual ground as a wider form of joint attention, and the the possible requirements for a shift to helpful expectations in forming the Gricean communicative intention were welcome new emphases.

The evolution and origins of language appears to be one of those flash fires in the intellectual landscape in this decade. It's an exciting time to watch the sinews of communication be carved out of pragmatics, the baselines of the abilities of chimpanzees and bonobos, and the hints in the 6 million year evolutionary story of hominins. Compared to other evolution of language authors such as Hurford and Burling, I notice more similarities than differences, but I found that Tomasello fills in more gaps in the story. Just the anecdotes of research alone make the book fascinating. Noteworthy besides the insights attributed to Herbert Clark were those of William Croft on language change. I've grown to feel lucky that we have the institution of Tomasello, but I was still surprised at how this book delivered.