Motivation and Speech

New Caledonian crows are smart enough to use tools in the wild. Like chimpanzees, they strip leaves and probe with the remains to catch insects.

Download PDF version of post.

Summer time is goof-around time, but I want to respond to a comment made this week by drawing attention to an essay published in the current PNAS by Alex Kacelnik, “Tools for thought or thoughts for tools?” (extract here). It’s a meditation on tool use inspired by a paper (abstract here) reporting on crows and tools. The paper describes a housebred rook whose conspecifics do not normally use tools in the wild, but this rook did spontaneously create tools including shaping wire to form hooks, use different tools in a sequence, and other actions that suggest a sharp intelligence. New Caledonian crows are known to use tools habitually, but not these rooks.

The finding is of interest to this blog because the rise of tool use is so important to the story of the human lineage. The rise of tools, language, and intelligence are commonly taken as working together. Kacelnik’s paper examines another element, motivation.

Kacelnik considers three explanations for the unexpected ability of rooks to use tools:

1. The common ancestor of the rooks and New Caledonian crows was a tool user, but since then the rook lineage has lost the motivation to use tools but has kept the intelligence.
2. The common ancestor had the intelligence to use tools, and then the New Caledonian crow evolved the motivation.
3. Both species have the intelligence to use tools but the rooks do not normally use them because they lack the environmental pressures to motivate them.

This group of explanations intrigues me because we see the same problem with apes and humans. Humans are like the New Caledonian crows; we use tools naturally. The rooks are like the apes; they are smart enough to use tools, but don’t do it as regularly as us. Experiments have also show that apes are intelligent enough to use language at least at the pidgin level, but they never do it in the wild.

Working on this blog has sold me completely on the notion that the central evolutionary events leading to speech were the ones that created a social interdependence that motivates us to share perceptions.. That's why I was pleased to find that Kacelnik’s essay cites Michael Tomasello point that it is the social structure of humans that uses collaboration and shared goals, not intelligence, that makes us use language. We have, says Kacelnk, “a unique motivation to share psychological states with others.”

So one comment posted this week that overlooks the peculiar human psychology did irk me. It came in a lively defense of Derk Bickerton's book, and mainly I was glad to see the comments. Despite my grave doubts about his latest book, he’s a hero of mine and has been for over 25 years. But one of the comments (here) included the note, “If the FOXP2 gene is responsible for language then where is my talking finch, gorilla, and ferret? Would perhaps language somehow not provide an advantage to those animals as it does to us humans?” I’m skipping over the fact that nobody says FOXP2 is “responsible for” language to address the more serious issue of the talking finch, gorilla, and ferret. Language seems so useful to us that we can suppose any species would benefit from it, but that overlooks the motivation issue. Gorillas, ferrets, and even finches lack social structures that permit collaboration and social sharing, so they have no motive to speak, no matter how well their brains and genes might be able to handle the task.

Comments

[raymondw]

There's nothing particularly mysterious about the rook's tool-using ability, despite the fact that (as far as observers know) it doesn't use tools in the wild. Although I don't have access to Bird and Emery's article, judging from the abstract, I would say that the rooks were trained via operant conditioning to use objects (rocks and twigs) to get the reward of a worm. This procedure is very much like what was done in the ape language experiments. The use hand gestures--after all the hand is a tool--or lexigrams to achieve a resulting reward is the same kind of behavior as the rooks. What Bird and Emery have demonstrated, as has been demonstrated thousands and thousands of times is the power of operant conditioning as a learning tool in most species of life.
----------------
BLOGGER: I'm not sure what in the abstract leads to the suspicion that operant conditioning was used, unless one argues that everything done by anyone at any time is the result of operant conditioning. That was the doctrine Chomsky exploded 50+ years ago.

[raymondw]

I'm sorry, I guess I read something into the abstract wasn't there. So I retract the phrase "judging from the abstract." But I continue to maintain the rest of what I said. Furthermore, operant conditioning is NOT a doctrine. It is a perfectly tried and true experimental procedure the results in altering the behavior of an organism. Also it is highly arguable that Chomsky exploded anything. It has never been experimentally demonstrated that operant conditioning is not the basis for language learning. All Chomsky did is argue verbally in a way that convinced many people that operant conditioning couldn't be the basis for language learning. Let's clear about this. This isn't an issue of competing ideologies but of empirical fact.
-----------------------
BLOGGER: Yes, operant conditioning is real but the suggestion that it explains language learning is not based on experiment. Chomsky did show quite clearly that Skinner's attempt to explain language in terms of operant conditioning was insufficient.

[raymondw]

We could argue this issue all day long using generalities about what Chomsky showed or didn't show. If you would like I can refer to a whole body of literature by behavior analysts that demonstrate otherwise on many linguistic issues. But I don't think this blog is the place to do it.
-------------------
BLOGGER: Why not give us three of the most important references? Especially, point to experiments that rebut Chomsky's "poverty of the stimulus" complaint. In recent years this blog has reported on criticism of that issue, coming from the opposite direction—awareness and understanding—but of course those points have nothing to do with operant conditioning.

[raymondw]

I think that we have veered far from my original comment on your piece on tool-using by rooks and how it might relate to language-using. Bird & Emery's abstract suggest that rooks are capable of insight into solving problems, like how to get a worm out of a plastic box by dropping a stone into the opening of the box, which causes the bottom of the box to open, thereby making the worm available for eating. I have read many animal studies on similar experiments and most of them involve using operant conditioning protocols. Thus, I suspected that the so-called insight for solving problems was just the result of similar kind of conditioning of the rooks. I will try to get a copy of the Bird and Emery article to find out just what kind of experimental protocols they used in studying rook behavior. When I do I'll get back to you. If you happen to have a copy of the article, I'd appreciate your sending it to me. If tool-using turns out to be a product of operant conditioning, then might it just possibly be that language-using is also a product of operant conditioning? Of course, whether it can completely explain language behavior is another issue.

As for rebutting Chomsky's "poverty of stimulus" argument, there have been plenty of those. particularly by Geoffrey Pullum, but others too. Should I just send the references as an add-on comment to your "Motivation and Speech" article?
--------------------------
BLOGGER: Just stick the refs in a comment where all visitors to the blog can see them.

[jkor]

I very much like this essay, and take issue with only one of your somewhat tangential statements:
"Experiments have also show that apes are intelligent enough to use language at least at the pidgin level, but they never do it in the wild"

Well, they may use language in the wild, and we just haven't figured that out yet. I do not see why this is not a reasonable possibility. I believe that at least one bonobo researcher has stated that this is likely.
--------------------------
BLOGGER: I was a college student when I attended the press event where Louis Leakey announced that Jane Goodall had found chimpanzees were making tools. That was big, but nothing compared to what would follow the discovery that chimpanzees or bonobos use a pidgin. Personally, I think it more likely that elephants have language. They have the society to support it.

[jkor]

Wow that must have been quite the event!

[raymondw]

The four comments that I made earlier have been challenged by the blogger. In the first and second comment I suggested that tool-using capability of the rooks was a result of operant conditioning, whereupon the blogger made the remark that operant (or instrumental) conditioning was a doctrine “exploded” by Chomsky, supposedly in his review of Skinner’s Verbal Behavior. First of all, operant conditioning isn’t a doctrine, but an experimental learning procedure, as well as a natural way (but not the only way) that most animals learn, including human beings. If you check almost any text on animal behavior, like Cognition, Evolution, and Behavior by Sara J. Shettleworth, you will find a discussion of this kind of learning. Secondly, Chomsky never denied the existence of operant conditioning. He just denied that it had any important role to play in language acquisition. His arguments did not go unchallenged by linguists, psychologists, and philosophers who were more empirically oriented, although they were largely accepted by many linguists and cognitive-oriented psychologists. Here are a few of these challenges to Chomsky’s review:

Andresen, J. T. (1990). Skinner and Chomsky thirty years later, Historiographia Linguistica. 17:145-166.

MacCorquodale, K. (1969). B. F. Skinner’s Verbal Behavior: A Retrospective Appreciation, Journal of the Experimental Analysis of Behavior. 12:831-841.

________________ (1970). On Chomsky’s Review of Skinner’s Verbal Behavior, Journal of the Experimental Analysis of Behavior. 13:83-99.

Palmer, David C. (2000). Chomsky’s Nativism: A Critical Review, The Analysis of Verbal Behavior. 17:51-56, reprinted from P. N. Chase & L. J. Parrott (Eds). (1986). Psychological Aspects of Language (pp. 44-60).

______________(2000). Chomsky’s Nativism Reconsidered, The Analysis of Verbal Behavior. 17:57-74.

_____________. (2006). On Chomsky’s Appraisal of Skinner’s Verbal Behavior: A Half Century of Misunderstanding. The Behavior Analyst, 29(2): 253-267.

Chomsky’s review attacked the basic concepts of operant conditioning and then tried to show that the nature of the syntax of human languages was much too complex to be mastered by its principles, since he considered them to be too simplistic. The above articles that address his review demonstrate clearly how muddled and misleading was his take on behavioral concepts. In the review there was no mention of the poverty of the stimulus argument (POSA). You won’t even find the word “poverty” used in his review. The POSA, or what Chomsky has also called Plato’s Problem does not seem to have appeared until the late 1970’s or early 1980’s. It is a ploy used by nativists to attack any empirical approach to understanding language acquisition, Skinnerian or otherwise. As Dan Slobin (see below) has remarked, it is really an “argument from the poverty of imagination.” Essentially, he derided the idea that if Chomsky can’t imagine a reasonable empirical account of language acquisition, then neither can anyone else.

Cowie, Fiona. (1999). What's Within? Nativism Reconsidered. Oxford University Press

Pullum, G. K. (1996). Learnability, Hyperlearning, and the Poverty of the Stimulus. Available online at http://users.ecs.soton.ac.uk/harnad/Papers/Py104/pullum.learn.html.

Pullum, G. K., & Scholz, B. C. (2002). Empirical assessment of stimulus poverty arguments, Linguistic Review, 19(1): 9–50. Available online at http://idiom.ucsd.edu/~boyd/pullum&scholz2002.pdf.

Sampson, Geoffrey. (2002). Exploring the Richness of the Stimulus, Linguistic Review, 19(1): 73-104. Available online at http://www.ucd.ie/artspgs/research/sampson.pdf

Scholz, B. C. & Pullum, G. K. (2005). Irrational Nativist Exuberance. Available online at http://people.ucsc.edu/~pullum/scholz/Exuberance.pdf.

Slobin, Dan I. (1988). Confessions of a Wayward Chomskyan, Papers and Reports on child Language Development, 27:131-136. Available online at http://eric.ed.gov/ERICDocs/data/ericdocs2sql/content_storage_01/0000019b/80/1e/3b/f9.pdf.

Besides these overarching arguments against the POSA, there are many empirical studies that have demonstrated that environmental and contextual factors act as independent variables in the acquisition of language. They are too numerous to list here.

Finally, I would strongly recommend to anyone interested in the Skinner vs. Chomsky controversy to prepare themselves for following the back and forth arguments by reading some basic works on Chomsky’s views of language and its acquisition, either by Chomsky himself or his followers, and some basic works on Skinner’s views. Since I am not a Chomskyan, I can’t recommend anything in particular. But in regard to Skinner’s views, I recommend his introductory book Science and Human Behavior and a recent standard textbook on behavior analysis, like W. David Pierce and Carl D. Cheney’s Behavior Analysis and Learning, 4th ed. Then thoroughly read Skinner’s Verbal Behavior. This is the only way I know of really equipping yourself to evaluate Chomsky’s criticisms of Skinner.

[raymondw]

I was wondering why the comment I tried twice to post hasn't been posted yet.
---------------------
BLOGGER: I'm sorry and alarmed to learn there is some sort of problem. I will see if I can learn anything. ... Ok. Did I solve the problem?

[isakerem]

I had some difficulty in understanding the discussion. Firstly, the issue here is about a crow's using tool and nothing else. Why do we draw conclusions about language use? And also, why did this article lead to a discussion about Chomskyan account for language acquisition? I do believe that we have the blueprint of language as we are born and that principles and parameters are there to be fixed. And I believe that no better explanation has been purposed so far including behavioral accounts like imitation, stimulus-response. If the issue is to draw a correletaion between tool use and intelligence on one hand and intelligence and language on the other, this is not the way I am inclined to think about language acquisition. Think about Williams syndrome where the patients have an IQ of about 50 but are fond of unusual words. As cited by Pinker, one of eleven-year-old child with Williams syndrome was observed to pour a glass of milk and say "I'll have to evacuate it." And they can also use perfectly grammatical sentences.

[isakerem]

and also, if I'm not wrong intelligence has to do with creating tools rather than using them.

[raymondw]

Dear "isakerem",

If you follow the comments I made earlier and the blogger's responses to them, I think it will be clear the episodic events that led to the comments that followed about Chomsky vs. Skinner. You are quite right in asking what does the tool-using and tool-shaping abilities of crows, ravens, and rooks have to do with language use. I believe that the blogger wanted to point out first the parallel between crows, which exhibit tool-using behaviors in the wild, and rooks, which do not but can be trained to do so, and humans and apes. Humans use language naturally, while apes do not, but show the capability of doing so, at least in a rudimentary way. Secondly, I think that blogger wanted to make the point that these abilities, whether potential or behaviorally manifest are perhaps the outcome of our social nature. On the other hand, species of birds and apes seem to lack the "collaboration and social sharing" that we human beings engage in, which in turn places limits on their linguistic behavior.

[topaze]

Thank You raymondw for your trouble providing the great links to the skinner and chomskey information. That is what I was looking for when I stumbled on this fine blog. I will check the archives and am excited about learning more.