Blog Rating

Selected Books by Edmund Blair Bolles

  • Galileo's Commandment: 2500 Years of Great Science Writing
  • The Ice Finders: How a Poet, a Professor, and a Politician Discovered the Ice Age
  • Einstein Defiant: Genius vs Genius in the Quantum Revolution

« Computation Is So Pre-Darwinian | Main | Adoption Among Apes »



You say, "This blog has argued from its beginning that apes are smart enough to use language". This is a statement that needs some heavy rethinking, I'm afraid. Aren't you collapsing the notions of "using words, some elements of language, some concepts in some ways, etc." and "using language"? The former is amply documented. The latter, I'm afraid, is simply not the case, as no apes are capable of using language (speaking, writing, arguing, narrating, explaining, etc.) beyond those very simple elements. Perhaps you need to distinguish the use of protolanguage from the use of language.
BLOGGER: Well, yes. They use a lexicon without syntax. But they can do what one- and two-year-old children do. And believe me, if some investigator came out of Uganda tomorrow with news that chimpanzees in the wild had been observed using words, the response would be very noisy. The basic question is: apes can use words as well as one-year-olds, but they don''t. Why not?


I find Daniel Hutto's work on the "Narrative Practice Hypothesis" highly fascinating so I'm pleased that the authors lend support to it. On the other hand I get the feeling that they are attacking a straw man.
Almost nobody argues that what arises around 4 years as a Theory of Mind is language-independent. In fact people like Jill de Villiers and Janet Astington have convincingly shown that language (including the mastery of certain syntactic constructions) may be very important in the development of a complex ToM including things like false belief understanding). The work on ToM and Nicaraguan Sign Language by Judy Kegl and Jennie Pyers is another case in point which illustrates that a complex ToM needs equally complex linguistic interaction to develop.

I think it is very important that they point to the fact that motivational factors have probably played and important part in the development of our linguistic and ToM capacities.
But something I don't really get (I don't have access to the paper): It is clear that chimpanzees in captivity and the wild point IMPERATIVELY i.e. using others as "social tools" to get s/th. This is something that clearly does not require complex forms of intentionality-understanding. But what about declarative and informative pointing just to share attention and help, this kind of pointing does not develop in any of the other apes even if they are brought up in a human environment and it arises before children acquire language (Tomasello 2005 - Why Don't Apes Point, and also nicely illustrated in his 2008 book). Even if we acknowledge the huge impact that social praxis and linguistic and nonlinguistic complex interaction have on cognitive development, there has to be something special and unique about human babies that enables them to develop the kind of "shared intentionality infrastructure" that other apes lack.

Dave Leavens

If I ask a language-trained chimpanzee, "Where is X?" (where X can be a person, an object, or some other entity) and the chimpanzees points to X, that is not an imperative point, but a point in response to an imperative. Virtually all language-trained apes do this (starting with Witmer, 1909). The notion that apes only request is a myth. Even wild apes use objects to elicit attention from others (the classic definition of a proto-declarative), as documented by Plooij (e.g., 1978). Some people seem to want to believe that apes only communicative imperatively, but that is contrary to the empirical record.

A second myth is that so-called "proto-declarative" gestures index a qualitatively different psychological process from imperative gestures. There is no direct evidence that this is the case and, indeed, in their original definitions, both "proto-imperatives" and "proto-declaratives" were explicitly and clearly defined as instrumental acts (Bates, Camaioni, & Volterra, 1975; and see, e.g., Leavens, Racine, & Hopkins, 2009, Leavens & Racine, 2009, Moore & Corkum, 1994). Proto-declarative pointing does not implicate a theory of mind in any species--to assert that it does constitutes, in my judgement, wishful thinking.


Thank you, Prof. Leavens, for pointing out the errors in my comment. My claim that the pointing behaviour of other apes is qualitatively different from that of human infants and not in the same way (proto-)declarative was absed on Tomasello's (2008) following argument: 1) in the few studies that have been done, even "language-trained" apes such as Kanzi pointed imperatively 96-98% of the time, and the other 2-4% had no clear function. 2) the kinds of answering gestures you mention are in his view not truly declarative, but "recognitory or classificatory, as the ape simply recognizes something and produces the associated sign in recognition“ (Tomasello 2008: 38)."
Tomasello I believe argues that infant pointing involves different psychological processes, which of course, can be doubted and I am also not aware of any evidence that would settle the question. For me, however, examples like these given by Tomasello support the idea of granting infant pointing a different psychological status than
""At age 13.5 months, while Mom is looking for a missing refrigerator magnet, L points to a basket of fruit where it is (hidden under the fruit). Gloss: Attend to the basket of fruit; it’s there.""
"“At age 13 months, J watches as Dad arranges the Christmas tree; when Grandpa enters the room J points to tree for him and vocalizes. Gloss: Attend to the Christmas tree; isn’t it great?”" (Tomasello 2008: 114f.). These of course can be interpreted differently and one definitely has to be careful how much underlying psycholigcal processes one wants to attribute to such behaviours. They definitely do not implicate a Theory of Mind but may implicate a unique form of social cognition that may serves as one of the building blocks of Theory of Mind and adult social cognition.

Tim Racine

It has proven surprisingly difficult to find unanimous or even unambiguous answers to the simple question, “Do primates share intentions with others when they gesture?” One reason for this is healthy scientific debate about whether a particular gesture or class of gestures truly requires the coordination of intentional behaviour between interlocutors. However, another reason, one that is in fact causing considerable nuisance in contemporary comparative and developmental gesture research, is non-scientific. The root problem is the lack of consideration for the meanings of terms that are in play in such research and the philosophical positions that are taken, explicitly or otherwise, by the researchers who interpret such terms in particular ways. Consider imperative or declarative gesturing, which both manifestly require means-end reasoning and previous experience in some particular gesturing context. The influential theory of Michael Tomasello and his colleagues who explicitly, but by no means uniquely, couch their explanation of these gestures within the Representational Theory of Mind tradition that largely still has a stranglehold in this research area, perhaps so much so that most researchers do not even recognize it as philosophical theory. To simplify, for example, Tomasello et al. stipulate that one type of pointing is caused by RTM-styled mechanisms, whereas the other is not. But if you take away the RTM, which is part and parcel of the cognitivist framework that Tomasello and colleagues unwittingly attempt to defend on empirical grounds, there is no difference in postulated abilities underlying these forms of gesture. In my view, the apparent difference is caused by an imposition of philosophy on science. However, the theory of Tomasello et al. casts a long shadow over the field and researchers now have to worry about whether agents who manifestly are engaging in coordinated intentional behaviour are in addition engaged in ‘shared intentionality’, a term of art that Tomasello’s research team has borrowed from a small set of literature in the philosophy of mind. This is another RTM-styled device that while possibly philosophically useful for certain things has no literal parallel in developing agents be they human or nonhuman primates. This research team is mixing up empirical, theoretical, and conceptual issues and the upshot of my advice for gesture researchers who are concerned with these issues will be to abide; there is ample reason to believe that the cognitive revolution that begun in 1956 is in its end game.

David A. Leavens

@ Michael,

Your cited examples:

"At age 13.5 months, while Mom is looking for a missing refrigerator magnet, L points to a basket of fruit where it is (hidden under the fruit). Gloss: Attend to the basket of fruit; it’s there."

“At age 13 months, J watches as Dad arranges the Christmas tree; when Grandpa enters the room J points to tree for him and vocalizes. Gloss: Attend to the Christmas tree; isn’t it great?”" (Tomasello 2008: 114f.).

Why are these NOT the same kinds of recognitory or classificatory gestures that Tomasello attributes to great apes? At bottom, what you have are babies pointing to something and no direct evidence, whatsoever, on their motivation for doing this. The alleged motivations are entirely fabricated by the observer (i.e., Tomasello's glosses are single examples of plausible interpretations from an extremely large set of mutually inconsistent plausible interpretations). Apes do these sorts of things, too (e.g., Leavens, Hopkins, & Thomas,2004; Premack & Woodruff, 1978; Vea & Sabater-Pi, 1998; Yerkes, 1943; among many others), and there is no empirical basis for attributing one set of entirely hypothetical psychological processes to non-human apes and another set of entirely hypothetical psychological processes to human children. This is because nobody actually measures these psychological processes--they are nothing more than guesses and wishful thinking.

In our article, Tim Racine and I (a) showed that so-called "proto-declaratives" (e.g., exhibition of self and the use of objects to capture the attention of another)are commonplace in great apes in the wild as well as in captivity and (b) criticised the currently fashionable notion that the only way one sentient organism can communicate with another is to speculate about the invisible causes of their behaviour. The double-standard of interpretation is that if apes do it, it must be because of relatively simple psychological processes, whereas if human infants do the exact same thing, it must be because they have species-unique capacities for sharing intentions. I recognise that some people really want human babies to have special cognitive capacities that are absent in all other animals, I even recognise that it is theoretically possible that human babies have some unique cognitive and/or motivational capacities, but as a scientist, I think these sorts of claims need to be interpreted in light of the full range of relevant empirical evidence, not a subset of theoretically congenial findings cherry-picked from the literature.


I have doubts about how necessary theory of mind is to language use. My eldest son is high-functioning autistic. As a toddler, he lacked any semblance of theory of mind. Other people were basically furniture decorating his environment. But he developed speech normally and on time, it just was not used for the "Look at that!" sort of interaction- he was completely happy to talk to himself, or to talk AT other people. He didn't attempt to direct my attention whatsoever until he was at least five years old. Perhaps had I not provided his wants without his asking, it might have occurred sooner, but I don't really know that. Theory of mind and sense of empathy came later, as people failed to behave in ways he anticipated, and he had to puzzle out why they wouldn't. So, by my observations, Theory of mind may be related to language use, but language is not predicated on its existence.
BLOGGER: Thanks for that comment. That's the first time I ever heard a theory of mind sound like a real theory.


@ Prof. Racine & Prof. Leavens
Thank you both for these clarifications. I am clearly not familiar enought with the relevant literature and the fact that I have mostly read about the research of Michael Tomasello and his colleagues at the Max-Planck-Institute for Evolutionary Anthropology probably has given me a too one-sided view of the issue. In addition, I'm a linguist in training and this discipline often comes with its own share of hidden assumptions.
You are certainly right in stressing that humans infants and other apes have vastly different rearing and learning histories and that this has to be taken into account before making any claims about human-specific cognitive mechanisms. And indeed, this is probably all the more justified given that enculturated chimpanzees seem to be much more attentive to the rationality of goal-directed actions (e.g. David Buttelman et al. 2007).
I completely agree with you that we have to be extremely careful to what extent we interpret infant behaviours in terms of a representational theory of mind and in terms of cognitively sophisticated processes. Pointing may very well be not a hallmark of uniquely human psychological processes and I really have to get hold of your paper and read more critical opinions on this topic.
Still I am deeply fascinated by the experiments done by people like Henrike Moll indicating that human infants seem to make specific assumptions that at first glance seem to be adaptations for a special kind of social interaction. At 14 Months they seem to be able to discriminate who they shared a specific experience or ‘perspective’ (of a very simple, perceptual-based kind) with and base their subsequent actions on this kind of “knowledge” or rather familiarity. These assumptions are probably to be explained in terms of specific kinds of non-cognitive processing features, and it may very well be that other apes would also make the same assumptions given similar learning histories and interaction patterns, but it is a highly fascinating research avenue nevertheless and might serve as stepping stone for tentatively assuming some kind of uniquely human processing tendencies (again, maybe of an extremely simple and non-sophisticated kind) and then this interpretation could be properly elaborated in light of all the available evidence.

The comments to this entry are closed.

Bookmark and Share

Your email address:

Powered by FeedBlitz

Visitor Data

Blog powered by Typepad