Will count towards the SAS - Linguistics Major and Minor
Although many animals have the potential to communicate with others of their species, no animal communication system remotely approaches the expressiveness of human language. Humans can speak of times past and potential events, preserving history and technology through oral tradition and planning future acts in detail as a group. This obviously provides human beings with an enormous adaptive advantage over any similar species that can neither benefit from past collective experience, or coordinate the actions of groups with great precision in response to changing circumstances.
The theory of natural selection tells us that individuals with traits that permit them to adapt to their environment better than their competitors are more likely to survive and produce offspring. Those individuals will transmit their traits to their offspring (by means of their genes, a concept not available in Darwin’s day). Thus primates with the human language faculty (HLF) would thus likely reproduce faster than those without human language, other things being equal, because such individuals would belong to communities more successful at exploiting a wide range of environments and circumstances. There is little, if any, debate about this.
This leaves us with a mystery, however. Evolutionary theory, and its central mechanism of natural selection, is an interesting theory precisely because it explains, on the basis of simple and consistent principles, how complex biological structures could arise from small advantages accumulating in populations adapting to environments over long periods of time. But human grammar, the means by which we store our knowledge of language, appears to be a highly complex structure that includes features that do not appear to offer selective advantage.
When we consider what linguists actually know about the grammars present day humans can learn, however, the structure of a natural selection argument becomes even more muddy. For example, a natural selection argument only insures that primate+HLF will outcompete primate-HLF; it does not entail that the internal structure of HLF is itself determined by natural selection. It does not necessarily insure that a hypothetical HLF, HLF#1, which has all the other natural language properties except one linguistic construction K, would fail to compete successfully against HLF#2, which is an HLF that permits K like those found in the world today. It would have to be argued that primates with HLF#1without K would fail to mate successfully, and so their traits would gradually lose out to the primates with HLF#2. This form of argument about why HLF has the properties it does seems quite dubious. On the basis of what we know about linguistic constructions, it does not seem plausible that human languages have verb agreement because verb agreement has an evolutionary advantage, especially since there are many languages that lack verb agreement altogether, such as Chinese, which is spoken as a first language by more of the world’s people than any other language. So verb-agreement, found in most human languages, cannot have developed because people who have verb-agreement in their language have a better chance of producing healthy and fertile offspring. There are even well-known cases of Specific Language Impairment which involve a hereditary inability to process or correctly produce agreement relations in natural language speech (although the details of the effect are debated). The fact that this impairment runs in families, including families where the parent has SLI, show that this linguistic glitch is not being weeded out by selectional pressure, even when it involves an innate genetically heritable component that leads to sub-normative communication.
If the human language capacity is inborn (a big ‘if’ for some) and if it is complex in its internal structure (also a big ‘if’ for some), then it poses an interesting problem for evolutionary theory: How can this form of biologically based mental complexity be a product of evolution without the incremental pressure exerted by natural selection? This is an inquiry that cannot be solved without an interdisciplinary approach, and, fortunately, researchers in a wide variety of fields are currently interested in these questions. To see what is at stake from a linguistic point of view, you will need to get a crash course in linguistics – about a four to five class unit. Then, using linguistics as a point of departure, we will explore a number of neighboring fields and disciplines from which researchers have staked out positions as to how the languages capacity emerged and has been shaped by evolution. From the point of view of what linguists know, we will then evaluate whether or not these proposals establish what they claim to show. Throughout, we will try to develop a perspective on what such studies must establish to truly address the emergence the capacity that linguists know humans to have.
Interesting answers to this question are just beginning to emerge, because evolutionary biology has only recently (though with gaining momentum) taken into account what is actually known about the structure of HLF (in class we will discuss why this is so). In addition, much more sophisticated, fine-grained accounts of other human capacities and the capacities of other primates have led to a reassessment of what it exactly is that humans have - and other species do not. The connection between the latter factors and the design of the language faculty are currently quite speculative, but some interesting hypotheses have emerged that we will begin to explore.
The work of the course will begin with a thorough review of evolutionary reasoning about how complexity arises by natural selection and why the application of a natural selection account of certain aspects of human intelligence may not have the right properties to be successful. As we explore the issues that seem problematic for natural selection, we will be led into discussion of the evidence that has been brought to bear on the central question from a variety of other disciplines. For example, students will need to know some basics of brain and vocal tract anatomy, function and pathology and the paleontological record of primate origins. Students will need to know something of the comparative cognition of non-human species, including their communicative practices and abilities. Genetic explanations of behavioral traits will be considered, and measured against social constructions of knowledge. Just how much we explore in any one area will be determined in part by the issues students choose as research topics.
KEN SAFIR is a professor in the department of linguistics at Rutgers University, a department he helped to found in 1989, seven years after receiving his Ph.D. from MIT. He is a linguistic theorist and syntactician with interests in the syntax-semantics interface and the nature of linguistic of anaphora in particular. He has also served as editor of the Journal of Comparative Germanic Linguistics, of which he was also one of the founding editors. He has studied definiteness effects, the null subject parameter, crossover effects, small clauses, parasitic gaps, the structure of nominals and relative clauses and many other phenomena, but for the last 15 years much of his work has been devoted to the locality and interpretation of anaphoric relations and the connection between these relations and the morphology of anaphors, as evidenced by many of his recent publications, including The Syntax of Anaphora, published in 2004 by Oxford University Press, and The Syntax of (In)Dependence, published by MIT Press, also in 2004. In addition to new work on transitivity and reflexivity informed by his work for the Afranaph project, he is also working from a minimalist perspective on some fundamental architectural properties of the theory of syntax.
For more information about Ken Safir, please see his website.