[bitwize]

The idea that the ability to grok recursive structures is due to some part of human physiology that is unique to humans is pretty much taken as a given in Chomskyan linguistics, almost to the point of religious belief -- it's Chomsky's organ kundabuffer. Ask a Chomskyist to identify this organ or structure and they will handwave: "Well, we don't know exactly what or where it is but we know it's there, humans have it and animals don't so nyeh. We've prebunked any promising animal language study you can produce."

[marcosdumay]

> is pretty much taken as a given in Chomskyan linguistics

That he believes in it is irrelevant. His linguistics do not require the feature to be absent in animals, only that humans have it.

Chomsky is famous for having some extremely insightful ideas that are important in several areas; and also for holding a lot of extremely stupid ones that are ridiculed by several areas. We don't throw the first set away because of the second.

[astrange]

Chomsky’s reputation isn’t helped by the fact that he has a public email address and can and will answer more or less anything anyone asks him.

It’s nice he has a lot of energy for a 93 year old but I’m not sure WSJ writers really need to ask him for comment on everything. You kind of already know what he’s going to say.

[naniwaduni]

Being right more often than a stopped clock doesn't make you a good timepiece either, though.

[timeon]

I do not want to be that guy but is not Chomskyan linguistics an animal language study since it concerns humans? It seems strange to me that one would think that ie. birds are more closely related to other mammals then other mammals are to humans.

[technothrasher]

Obviously here, animal means non-human animal. If, as Chompsky proposes, humans have evolved a unique structure for language processing that no other animal has, then it is irrelevant how closely or distantly related any other animal is to us. Relation would only be important if Chomsky is correct about the structure but wrong about it evolving in humans rather than in the pre-human evolutionary line.

[inglor_cz]

We may well have a special structure for language processing, but Chomsky's claim that it is unique may be wrong.

Convergent evolution is a thing in nature, some structures evolved independently multiple times. Humans and octopuses have very similar eyes, which emerged from different structures.

I wouldn't be surprised if corvids developed human-like intelligence traits independently.

[blippage]

> In the new study, two crows were trained to create embedded sequences ... The idea that the ability to grok recursive structures

I have this vision that in a year's time we'll walk into Twitter's HQ to find it being staffed by crows pecking at keyboards and spitting out Typescript.

[eurasiantiger]

Is this why we are still studying birdsong when birds are obviously not singing, but talking to each other in phonemes and words?

[beebeepka]

Singing? Sparrows have actual verbal fights all the time. I have seen and heard actual scandals many times. Even their body language speaks volumes