Mind Matters Natural and Artificial Intelligence News and Analysis
two young baboons
the young baboons are sitting on a log talking to each other
© Susan Flashman/stock.adobe.com

Can Animals “Reason”? My Challenge to Jeffrey Shallit

He believes that animals can engage in abstract thinking. What abstractions do they reason about?

Dr. Jeffrey Shallit is an atheist mathematician who holds to the odd belief that animals, like humans, are capable of reason. It would seem that a highly intelligent man who makes his living by doing mathematics would understand that animals don’t, and can’t, do mathematics. But Dr. Shallit remains confused on this point, as he makes clear in his response to my recent post on that inability of animals to think abstractly or to reason (“An atheist argues against reason”).

I observed that reason is defined traditionally in a very straightforward manner as the capacity for abstract thought. Shallit comments,

Whenever Egnor talks about something being “accepted” or “simple and straightforward”, you can be pretty sure that the opposite is the case. Anyone who wants to check Egnor’s claim can just go to the Oxford English Dictionary and type in “reason”. There are three senses for the word, two as a noun and one as a verb. The uses as a noun include 17 different subdefinitions and another 15 or so different usages in phrases. The uses as a verb include 8 different subdefinitions. The word “abstract” appears nowhere in any of these subdefinitions (it does appear in two citations, but not in the sense Egnor refers to). So Egnor is wrong twice: the “accepted definition” of the word is neither simple nor straightforward, and the meaning Egnor claims is not an “accepted” one. Jeffrey Shallit, “Yet More Egnorance” at Recursivity

For unclear reasons, Shallit thinks that recourse to dictionaries in metaphysical discussions is decisive. I remind him that dictionaries provide information about the popular use of words and are demonstrate nothing in a philosophical discussion. Otherwise, philosophers would only need to look up words and then declare issues resolved.

The widely-held philosophical understanding of reason is straightforward. Reason, which is a subset of abstract thought, entails the ability to think without particular objects in mind. That is, abstract thought is the ability to contemplate universals, rather than particulars. Shallit seems unfamiliar with this issue, which entails 3000 years of philosophical discussion and literature.

His dictionary must be an abridged edition.

He continues:

Of course, if you understand the theory of evolution, you realize his claim is likely to be utter nonsense. Abstract thinking is not a black-white thing; it’s a range of capabilities that, even among people, we see a huge variation in. Any capability with huge variation is subject to selection, and so it can evolve. Since people are descended from earlier ape-like creatures, it is quite believable that non-human animals would also display the ability for abstract thought, in varying degrees. And they do! Ethologists, who actually study this kind of thing, disagree with Egnor. (Also see baboons and crows, to name just a couple more examples.)

I do understand Darwin’s theory of evolution, which is why I think it’s junk science.

That, of course, has nothing to do with the nature of reason, so it’s unclear why Shallit brings it up. Reason is an immaterial power of thought—that is, it is the ability to contemplate abstract concepts and relations without recourse to thought about particular things. Aristotle and countless philosophers who followed him have pointed out that an ability to contemplate abstract objects of thought that are not linked to particulars is an immaterial power of the mind.

Immaterial powers can’t “evolve,” because Darwinian evolution, for what it’s worth, can’t act on anything that is not material. “Evolution of the power to reason” is an oxymoron. Shallit should look up “oxymoron” in his dictionary.

More from Shallit:

This is vintage Egnor—a flat assertion, made with no evidence, and contradicting what we know about (for example) machine learning. Machines can abstract from specific cases to more general concepts; that is exactly what is done routinely in machine learning. (To cite just one example, see here.)

Machine learning is a metaphor. Machines don’t learn. They don’t think at all. Machines are mechanical objects. We learn, using machines. Over time, we make machines that work better. But the machines don’t really learn anything.

Note to Dr. Shallit: A “slippery slope” isn’t really a physical incline. Animals don’t fall from the sky when it’s “raining cats and dogs.” Elvis didn’t mean that his girl was a canine when he sang ‘you ain’t nothin’ but a hound dog’. Watches don’t “tell time.” Machines don’t learn. Shallit should also look “metaphor” up in his dictionary.

Souls don’t exist; there’s no evidence for them. There’s no evidence for “immaterial powers.” Egnor’s claim is disputed by many, and it’s a plain lie to say it’s “obvious to all”.

Oddly, Shallit assumes that the existence of the soul is a matter for empirical science to investigate, akin to the existence of chloroplasts or asteroids. The soul in metaphysics is defined as the form—the intelligible principle—of the body. There is a profound metaphysical system— hylomorphism—in which the soul is a central concept. Shallit should read up on it, for example:

A human being is by nature a rational animal. That is to say, a human being is something which by its nature exercises both the animal powers of nutrition, growth, reproduction, sensation, appetite, and locomotion, and the intellectual and volitional powers possessed by angels. Hence it exercises powers of both a material and an immaterial sort.

Edward Feser, “What is a soul?” at Edward Feser (blog)

Shallit is denying the obvious. That animals can think only about concrete things, and not about abstract ones, is obvious. All of our experience with animals tells us this. It’s even more obvious when we consider just what it means to think abstractly.

All thought—concrete or abstract—must be directed to something. We think about stuff. When we (or animals) think about a concrete object, we think about the object. When we think about abstractions, we (by definition) aren’t thinking about a concrete thing. We are thinking about an abstract concept. And, since all thought is directed to something, what is thought about abstraction directed to? It is directed, of course, to the abstraction, expressed in language.

Language is essential to abstract thought. How can we think about an abstract concept if we have no word to name it? We can only think about “mercy” if we have a word for it. Try it: think about mercy, without thinking of either a concrete thing or of the word “mercy.” You can’t. Language is what makes abstract thought possible. Only humans have language, and only humans think in abstractions. Both are true, and they are intrinsically related. Language makes human thought—abstract thought and reasoning—possible.

This is obvious, when you think carefully about it. Why would an intelligent man like Dr. Shallit (he’s a professor of mathematics) not understand something so obvious?

So I have a challenge for Dr. Shallit, who claims that animals have the ability to reason and to think abstractly, without thinking about particular things. When an animal is “reasoning” about an abstract concept—number theory say, or financial markets—just what is in the animal’s mind, if not a thought of a physical object or a word? What do these animals who can “reason” reason about?

Also by Michael Egnor University fires philosophy prof, hires chimpanzee to teach, research Michael Egnor: A light-hearted look at what would happen if we really thought that unreason is better than reason.


An Atheist Argues Against Reason. And thinks it is the reasonable thing to do

Michael Egnor

Senior Fellow, Center for Natural & Artificial Intelligence
Michael R. Egnor, MD, is a Professor of Neurosurgery and Pediatrics at State University of New York, Stony Brook, has served as the Director of Pediatric Neurosurgery, and award-winning brain surgeon. He was named one of New York’s best doctors by the New York Magazine in 2005. He received his medical education at Columbia University College of Physicians and Surgeons and completed his residency at Jackson Memorial Hospital. His research on hydrocephalus has been published in journals including Journal of Neurosurgery, Pediatrics, and Cerebrospinal Fluid Research. He is on the Scientific Advisory Board of the Hydrocephalus Association in the United States and has lectured extensively throughout the United States and Europe.

Can Animals “Reason”? My Challenge to Jeffrey Shallit