Revisiting Fitch and Hauser's observation that tamarin monkeys can learn combinations based on finite-state grammar
Shigeru Miyagawa
November 2021

Chomsky’s hierarchy (Chomsky 1959) defined formal grammars starting with finite-state grammar (FSG) as the simplest, followed by phrase-structure grammar (PSG), and so on. Although Chomsky’s original intention was to use this for human language, in particular, to show that FSG is insufficient for modeling language, a large body of work recently has utilized this hierarchy for nonhuman primate and bird systems of communication. In a groundbreaking work, Fitch and Hauser (2004) showed that cotton-top tamarins can master combinations based on FSG, but not PSG, while humans have no problem with both. While accepting their conclusion about humans, I question the assumption that the stimuli that the tamarins were able to master are FSG. In nature, monkeys are never exposed to systems that can be modeled by FSG; the alarm calls are predominantly isolated verbal units, not combinatorial. Old World monkeys do have an ability to combine calls, but the combination is limited to two items. This is not FSG in any meaningful sense. I suggest that the Chomsky hierarchy, which is productively applied to human language, does not apply to nonhuman primate calls, which are severely limited in their combinatorial possibilities.
Format: [ pdf ]
Reference: lingbuzz/006336
(please use that when you cite this article)
Published in: Frontiers in Psychology
keywords: chomsky hierarchy, broca, frontal operculum, syntax
Downloaded:304 times


[ edit this article | back to article list ]