Human language is a conventionalized system of body movements that can represent and convey complex propositional messages .
The grammar of a language expresses the relationship between all possible language movements (utterances) and messages in that language.
The level-element model of grammar includes levels of phonology, morphology (lexicon), syntax, semantics, and pragmatics including reference.
Human language works by recombining elements at the various levels: morphemes into words; words into phrases and clauses.
This combinatorial or recursive aspect of language underlies our ability to create and understand novel utterances--including complex sentences with >1 clause. See Limber (1977) on the creative/projective aspect of human language.
The meaning of an utterance is built up out of the meaning of its morphemes in their syntactic (phrase) context as well as context of use. Linguists talk about the sense of an expression as well as the referent of an expression. For example the expressions "instructor in 512", "writer of these notes" and "John Limber" all refer in this context to me (John Limber) but all three expressions have a different sense based on our conventional meanings for the morphemes in those expressions. Note that "John Limber" does not have a sense in the same way that "instructor in 512" does. Proper names may only have a referent.
All human languages follow these same patterns; linguists think of all 5000 or so languages as dialects of a universal human language that evolved anywhere from 20,000 to 2,000,000 years ago. I would guess between 50,000 to 200,000 years ago for the first human language comparable to existing ones today. As discussed in class, however, I expect that hominids (habilis, erectus) had some form of referential language --unlike current ape communication systems -- based on the existence of tools dating from 2 million years ago. At this point, this is shear speculation based on the impoverished "teaching" methods observed in contemporary chimps' use of tools. One recent paper reports chimp tools dating back several thousand years. Comparison with presently used chimp tools reveals no improvements over this time. Contrast that with changes in human culture and technology over this period. Is this due to language?
Specializations for language include large brain, specific brain "circuits" into the generally left neocortex, and a specialized vocal tract. All of this must be somehow encoded in human genes but not in our close relatives. Note the difficulty Vicki had just making a simple voluntary articulation. However human language can also be expressed using manual signs; the same brain components seem involved. My guess that at least one critical component is a fine-grained control of movements required for both speech and signed languages.
The second year, roughly 20 to 36 months, is marked by extraordinary changes in children's speech. Length of utterances goes from a bit over one to around 4, with increasing variability. Vocabulary accelerates increasing by 4 to 9 words per day. Complex syntax--compound sentences, complement clauses, and relative clauses appear. Abstract verbs and modals like think, know, pretend, perhaps, might are used. Absolutely nothing like this occurs in any other animal --not even home raised chimps.
We may see some video on the Williams syndrome children. These children, with IQs testing around 50, show good syntactic capabilities and have good social use of language. This suggests that much of language acquisition is instinctive for humans, not a matter of exceptional primate intelligence. The case studies of Creole language formation also support this perspective.
Much of this research was done by Cheney, D., and Seyfarth, R. (1990). How Monkeys See the World: Inside the Mind of Another Species . Chicago: University of Chicago Press.
Temerlin, M. K. (1975). Lucy : growing up human : a chimpanzee daughter in a psychotherapist's family. Palo Alto, CA: Science and Behavior Books. (My guess as to the identity of "Lucy.")
Vicki was another chimp raised by psychologists (the Hayes') in their home. Like Gua, Vicki was clever and learned a lot of human activities except language. See her struggle to control her vocal tract, finally resorting to her hands to manipulate her lips.
Does this remind you at all of "Bonzo?"
Note how the younger, more paedomorphic animals interact directly with their human investigators while the older animals "work" behind bars.
Also note how after most performances, the animals are given a spoonful of jam. Imagine how well children would acquire language if this were necessary in human language acquisition. (This relates to my remarks earlier on lack of imitation and observational learning in non-human primates. See Tomasello, 1992)
Premack's conclusion seems correct. Apes have no "grammatical competence"--they have no sense of grammatical relationships expressed in symbols as to who did what to whom. (Relate to the aphasia victim below.) Premack has published a number of recent papers on human language and its evolution, stressing the significance of "recursion."
We have seen and discussed these issue before. Philip Lieberman, who has attempted reconstruction of vocal tracts from fossils, gives his ideas on language origins. There's nothing new here except his important emphasis on rate of human speech production and corresponding perception. We are much faster at transmitting vocal signals than other primates--in addition to transmitting much, much more information, i.e. more diverse unpredictable signals (messages, sentences). Timing of movement, is of course, crucial.
There remains much debate about the language capabilities of Neanderthals.
(See Limber, 1982 for a discussion of language evolution.)
(see Limber, 1982 )
Similar parts of the brain are involved in speech and signing.
There are unique sign languages in various cultures; they typically have complex grammatical structure comparable to spoken languages. (Its my impression that the "signing" of trained apes is as much like fluent human signing as the speech of 18 month-old humans is like the speech of fluent normal adults--and maybe worse since the fine motor control necessary for many signs seems beyond even the most dexterous non-human primate.)
Especially note the interview with the woman. She was given a standard picture interpretation task and was unable to use language to assign grammatical relationships (She didn't know who was said to eat who.. This is just what Premack said about apes!!!) Listen to her speech; she is clear but uses simple sentences, slowly articulated.
There are only about two "words" for snow in Eskimo; there are many combinatorial expressions about types of snow involving suffixes on those stems. This is how the MYTH of "hundreds" of words for snow began. See Martin, L. (1986). "Eskimo Words for Snow": A case study in the genesis and decay of an anthropological example. American Anthropologist, 88, 418-422.
Benjamin Whorf's ideas, while intriguing, are probably not as valid as once thought. See Hunt, E., & Agnoli, F. (1991). The Whorfian hypothesis: A cognitive psychology perspective. Psychological Review, 98, 377-389. for a recent review. Even more recently, the reports of the obscure Piraha, a group of several hundred Amazon natives have rekindled the Whorf issues.
Colapinto, J. (2007). "The interpreter: Has a remote Amazonian tribe upended our understanding of language?" The New Yorker 83(8): 119-137.
At the moment though, Malotki's conclusions about the relation between language and thought seem sensible.
