Η παρακάτω έρευνα υπόσχεται ότι επιβεβαιώνει τη θεωρία του Chomsky περί «grammar in our head». Για να ισχύει αυτό όμως δεν θα πρεπε να αποδεικνύει τον απριορικό χαρακτήρα αυτού του internal grammar μηχανισμού; Γιατί πχ τα αποτελέσματα της έρευνας δεν συμβαδίζουν με ένα interactivistic model ?
In the 1960s , linguist Noam Chomsky proposed a revolutionary idea: We are all born with an innate knowledge of grammar that serves as the basis for all language acquisition . In other words, for humans, language is a basic instinct. The theory, however, has long been met with widespread criticism — until now. A new study presents compelling evidence to suggest Chomsky may have been right all along.
The ability to walk upright for long periods of time is distinctly human; it sets us apart from our closest genetic cousins, the great apes. However, walking is both innate and learned, and while every human child is born with the underlying mechanisms needed to do so, the skill will never manifest without proper guidance and examples, Slate reported.
In this respect, Chomsky taught that language is much like walking. Although humans learn by example, he proposed that we are all born with a fundamental understanding of the underlying mechanisms of language. Chomsky’s original work, called universal grammar, is the reason why humans can recognize grammatically correct yet nonsensical phrases, such as “colorless green ideas sleep furiously.” Past research has shown our ability to distinguish words from nonwords , even without an understanding of the language, is a skill that even non-verbal babies possess. Researchers have long failed to prove this same instinctual knowledge also exists for grammar.
The most commonly accepted viewpoint on language acquisition suggests humans learn language by observing and memorizing grammatical cues. This theory posits that our understanding of language is built solely on experience, not an internal language processing feature. However, researchers from New York University recently used new technology to prove Chomsky’s theory may have been factual all along (not unlike these other scientists whose ideas were ahead of their time ).
For the study, the team recruited volunteers to listen to word phrases spoken in both English and Mandarin Chinese, including predictable sentences like “New York never sleeps,” grammatically correct yet less predictable sentences like “Pink toys hurt girls,” and word lists like “eggs, jelly, pink, awake,” a press release reported. These sentences were specifically designed so all obvious indications of grammar, such as voice intonation cues, were missing. This ensured the only indication of grammar would come from the subject’s own minds, not the sentences themselves.
As subjects listened on, researchers measured their brain activity using two tools: magnetoencephalography and electrocorticography. The first measures tiny magnetic fields created by brain activity and the second measures brain activity in patients undergoing brain surgery.
Results revealed brain activity changed depending on whether the volunteers had listened to a sentence, a phrase, or a word list. This showed that the subjects were able to process the grammar minus the obvious learned cues. «Because we went to great lengths to design experimental conditions that control for statistical or sound cue contributions to processing, our findings show that we must use the grammar in our head,» explained researcher David Poeppel in the release.
According to Poeppel, our brains lock onto every word to comprehend phrases and sentences. He said, “The dynamics reveal that we undergo a grammar-based construction in the processing of language.”
In an email to Medical Daily, Poeppel explained that although it is difficult, if not impossible, to prove theories, the data ascertained in his research supports crucial aspects of Chomsky’s theory, namely that listeners build abstract, hierarchical constituent structures of linguistic information.
“I’d say, on balance (comparative language research, language acquisition research, these kinds of brain data) the empirical research favors the Chomskyan view, as unpopular as it is,” wrote Poeppel.
Poeppel also recognized the controversy in his finding, seeing as the preferred view is that grammar is achieved by using acoustic cues such as intonation, and statistical cues, like word transition.
“However, we demonstrate that linguistic structure building happens in absence of those cues – so grammar based structure building must exist,” wrote Poeppel. “That is, in brief, the controversy.”
Source: Ding N, Melloni L, Zhang H, Tian X, Poeppel D. Cortical tracking of hierarchical linguistic structures in connected speech. Nature Neuroscience . 2015.
Update : This article has been updated to include quotes from researcher Dr. David Poeppel.