EDITORS: Richard K Larson, Viviane Déprez, Hiroko Yamakido TITLE: The Evolution of Human Language SUBTITLE: Biolinguistic Perspectives PUBLISHER: Cambridge University Press YEAR: 2010
Karl Diller, Cann Lab (genetics), University of Hawaii at Manoa, and Linguistics program (Emeritus professor), University of New Hampshire
This book is an outgrowth of a 2005 symposium at Stony Brook in which the intent was to explore the implications of Hauser, Chomsky, and Fitch's paper ''The Faculty of Language: What Is It, Who Has It, and How Did It Evolve?'' (2002). That paper, hereafter 'HCF,' is reprinted as chapter 1. Then 14 chapters, including additional chapters by Hauser, Chomsky, and Fitch on their own, respond to issues raised by the original paper.
The basic intent of HCF and of this book is to explore how or whether generative linguistics can be reconciled with biological evolution. If it has been hard to fit generative grammar into evolutionary theory, it is because of the focus on the language of an ideal speaker-hearer, I-Language, in an idealized homogeneous speech community, specifically ruling out an evolutionary setting. The editors argue that in the principles and parameters model, ''the linguistics 'module' was unique in content among all our faculties... [and] the question of language origins could gain no empirical or theoretical traction'' (p. 3). In moving from the principles and parameters model to the minimalist model, they argue, generative grammar has opened itself to these questions of origin in an evolutionary setting.
The editors state that nearly all of the authors have been ''influenced in one way or another by the Hauser, Chomsky, and Fitch (HCF) paper'' (p. 1), but nevertheless the majority of chapters dispute various arguments of the HCF paper, such as whether 'recursion' is important for the origin of language, whether or not language is basically for thought or for communication, and whether natural selection and adaptation apply to the origin of language. There is enough difference in approach to provide some interesting and good scientific dialog.
The original Hauser, Chomsky and Fitch paper is well known for its distinction between the broad sense 'faculty of language' (FLB) and the narrow sense 'faculty of language' (FLN). FLN is ''the abstract linguistic computation system alone'' (p. 17), reduced in minimalist theory to the operation 'merge' which embodies recursion. FLB includes the 'sensory-motor' and 'conceptual-intentional' systems, which are linked together by the FLN. The peripheral parts of FLB are said to ''have an ancient evolutionary history, long predating the emergence of language'' (p. 23), whereas FLN is uniquely human. Human uniqueness is ''an overarching concern'' for HCF (p. 21), in contrast to the search for connections and the continuity of life in evolutionary thought. They argue that the narrowness and uniqueness of the FLN renders ''the status of FLN as an adaptation open to question'' (p. 23). The suggestion is made that even recursion evolved as a 'spandrel,' for some reason other than language -- for navigation, perhaps, for example.
Chomsky follows up these ideas in chapter 2, ''Some simple evo-devo theses: how true might they be for language?'' Starting off with the statement that ''Study of evolution of some system is feasible only to the extent that its nature is understood'' (p. 45), two thirds of the chapter are devoted to understanding language in terms of principle and parameters theory and minimalist theory. The upshot is that ''Evolution in the biological sense of the term would then be restricted to the mutation that yielded the operation Merge...'' (p. 61) which would enable ''capacities for complex thought, planning, interpretation, and so on'' (p. 59), but not would not, at first, involve communication. 'Externalization' (i.e. speech, sign language, lexicalization, communication, etc.) came later and ''may not have involved an evolutionary change -- that is, genomic change'' (p. 61). Accordingly, he argues, ''any approach to evolution of language that focuses on communication, or the SM [sensory-motor] system, or statistical properties of spoken language, and the like, may well be seriously misguided'' (p. 61). Chomsky envisions the sudden emergence of human language rather recently in the history of the human species, which, he says, ''would lead one to expect something more like a snowflake than the outcome of extensive Jacobian 'tinkering' over a long period'' (p. 59). Chomsky argues that the neural change for Merge, using Tattersall's words, ''probably had nothing to do with adaptation,'' and that ''the appearance of language and its anatomical correlates was not driven by natural selection'' (p. 59). Since adaptation and natural selection are key to the biologist's view of evolution, Chomsky's view of biological evolution is indeed minimal, when it comes to language.
''Your theory of language evolution depends on your theory of language'' is Ray Jackendoff's argument in chapter 3. Acknowledging the wider context of various theories of language resulting in different notions of what, if anything, is innate and special for language, Jackendoff contrasts his own theory of a parallel architecture for syntax and semantics with Chomsky's syntactocentric architecture of language. He gives a persuasive example of recursion in the visual system to show that recursion is not special to language (p. 69). The parallel architecture is much more compatible with the biologist's view of evolution. As he says, ''with the parallel architecture, one can imagine various scenarios in which the language capacity evolves in stages, each adding an increment to the system's communicative efficiency and flexibility...'' There is a fair amount special to humans, ''but it is motivated by its incremental adaptability'' (p. 71).
Since 'recursion' is central to the HCF paper, Tecumseh Fitch attempts to clarify the concept by discussing ''Three meanings of 'recursion': key distinctions for biolinguistics'' in chapter 4 -- usages from computer science, linguistics, and meta-mathematics.
In chapter 5, ''On obfuscation, obscurantism, and opacity: evolving conceptions of the faculty of language,'' Marc Hauser expresses his ''frustration'' at the critiques of the HCF paper, suggesting a ''wanton eagerness to misunderstand, misconstrue, or fabricate'' among the critics, or worse that ''the study of language evolution is doomed to endless obfuscation, obscurantism, and opacity!'' (p. 91). He then defends the HCF paper point by point. He concludes, ''I hope these clarifications help. If not, let me know and I will return to the pub for consolation'' (p. 99).
Three chapters in part two of this book, ''Language and the interface systems,'' extend well beyond the concerns of the narrow sense faculty of language. Peter Gärdenfors and Mathias Osvath discuss ''Prospection as a cognitive precursor to symbolic communication,'' arguing that the ''Oldowan culture, 2.6 -- 1.5 million years ago, constituted an ecological niche containing evolutionary forces that generated symbolic cognition. The long-ranging character of this culture made the use of prospective cognition, that is the skill to plan for future events and needs, beneficial'' (p. 104). An important result of their argument is that ''Coherent with evolutionary theory, this suggests that there was a gradual shift into symbolic cognition, in contrast to explanations relying on discontinuity'' (p. 114).
Michael Corballis argues in chapter 7 that language evolved before speech. Chimpanzees and other apes cannot be taught to mimic human speech, but they can be taught to communicate to a certain extent with sign language and other visual sign systems. Humans, then, were pre-adapted for a visual language. He argues that a gradual evolution from manual sign to speech could overcome what Robin Burlings has called the ''nearly fatal flaw'' (p. 119) of the gestural theory, of why one would switch from a perfectly adequate manual-visual language to an audible one, if mimicry and speech were nearly impossible for chimpanzees.
Dan Sperber and Gloria Origgi give ''A pragmatic perspective on the evolution of language'' in chapter 8. They contrast the ''code model'' of communication with an ''inferential model'' which goes well beyond the code to interpret the meaning of an utterance in context. They argue that the inferential model is necessary for a structured code to develop -- and for the code to change rapidly as it does in human language. The inferential model allows communication to precede grammar.
Part three of the book, ''Biological and Neurological Foundations,'' starts off with a discussion by Daniel Dor and Eva Jablonka of ''Plasticity and canalization in the evolution of linguistic communication: an evolutionary developmental approach.'' Plasticity is the ability of an organism or the brain to change in response to different environmental circumstances; canalization is the tendency to keep stable in the face of different environmental stimuli. They envision a co-evolutionary spiral which for the evolution of language is ''first and foremost a socially driven process.'' They say that ''Language not only adapted to the brains and minds of individual speakers, but the brains and minds of the speakers had to adapt themselves to language'' (p. 147).
Massimo Piatelli-Palmarini brings us back in chapter 10 to the generative viewpoint that ''I-languages are what one has to attempt to reconstruct the evolution of, when dealing with language evolution'' (p. 150). He argues that ''the evolution of language may not be the result of a cumulation of a host of smaller steps...'' and that ''communication may not have been AT ALL the driving force behind language'' (p. 161). He argues that ''Every inquiry into the evolution of language must be an inquiry into the evolution of the computational brain machinery capable of carrying out [the 'recursive handling' of] edge-features operations'' (p. 151).
Philip Lieberman's chapter (11) is titled ''The creative capacity of language, in what manner is it unique, and who had it?'' The focus, however, is on the neural and anatomical structures that underlie language and speech. The recursive properties of language he subsumes under ''reiteration'' which is mediated by sub-cortical structures. Reiteration for motor control (which extends to dancing) is the basis for reiteration in language. For speech, he argues that the length of the neck is a proxy for whether or not early humans had a modern supralaryngeal vocal tract enabling speech as we know it, where the horizontal and vertical parts of the vocal tract are about equal. He states that ''Speech, language, and some degree of cognitive flexibility surely were present earlier,'' but that the long neck evidence for the fully modern supralaryngeal vocal tract doesn't appear until some 50,000 years ago. The evidence is not yet published.
Studies comparing the language abilities of monozygotic and dizygotic twins are the focus of Karin Stromswold's chapter 12, ''Genetics and the evolution of language: what genetic studies reveal about the evolution of language.'' About two thirds of language disorders are said to be due to genetics, as opposed to environment, where language disorders are scored in such a way that one either has or doesn't have the disorder. For normal language variation, the methodological problems are more complex, and one finds genetic overlap with various non-linguistic abilities. Her conclusion is that ''results of genetic studies are most consistent with linguistic models and evolutionary theories that invoke both general and language-specific factors... [but] they cannot tell us the process by which language evolved'' (p. 190).
Part four of the book presents the Anthropological Context. Ian Tattersall focuses on ''behaviorally modern Homo sapiens'' in his chapter on ''A putative role for language in the origin of human consciousness.'' He argues that although anatomically modern humans appeared some 200,000 years ago, modern behavior appears from about 50,000 years ago epitomized by the cave paintings and bone flutes of the Cro Magnon in Europe and the systematic elimination of other primate species of Homo. The argument is that ''The most plausible candidate for this cultural stimulus is the invention of language, an activity that is virtually synonymous with our symbolic reasoning ability'' (p. 197). Arguing against the linear and gradual model of the mid-twentieth century ''Evolutionary Synthesis,'' he says that ''in evolution form has to precede function, if only because without form there can be no function. Indeed ... any novelty must arise as an ''exaption,'' a new variant arising independently of any novel function for which it might be co-opted'' (p. 197). His conclusion is that ''Remarkable as our species Homo sapiens undoubtedly is, it was not fine-tuned by evolution for symbolic cognition and language'' (p. 198).
The strongest critique of Hauser, Chomsky and Fitch's view of language evolution comes in Derek Bickerton's chapter, ''On two incompatible theories of language evolution.'' Bickerton argues that HCF is ''a model so abstract that any account of language evolution in terms of process is ruled out. That argument makes sense only if the emergence of language is *not* a process, but an instantaneous event'' (p. 200). Bickerton proposes an alternative: a process starting with a protolanguage characterized by ''Sequence'' as opposed to ''Merge'' in which words could have been linked together like ''beads on a string'' without grammar. His scenario imagines an environmental pressure for a ''communicative system capable of displacement,'' as in the system of bees and ants for recruitment to exploit distant but rich food resources. Then ''Once established, protolanguage itself became a selective pressure for its own expansion, and for mechanisms that would then regularize, and thus automate and disambiguate, ever-lengthening propositions -- mechanisms such as hierarchical structure and recursion. At the beginning of the process, human ancestors had only a standard primate conceptual system, but this co-evolved with language to gradually create the rich and complex system that characterizes language today'' (p. 210).
In the final chapter, ''On the evolution of language: implications of a new and general theory of human origins, properties, and history,'' Paul Bingham rejects the idea that a mutation for merge or recursion was necessary: ''there is no reason whatever to suppose that any qualitatively new neural/cognitive capability was necessary to initiate the evolution of human language (or communication more generally)'' (p. 223). The evolution of language is put squarely into the context of biological adaptations: ''language is merely one element of a more complex 'total package' of adaptations for uniquely human elite information exchange'' (p. 222). This generously wide adaptational context is shrunk back, however, to a narrow rate-limiting step in the evolution of language, ''the management of the conflict of interest problem'' in the context of selfish genes and selfish animals (p. 223). Lower animals, he argues, could not evolve wide circles of cooperation with non-kin because free riders always did better in the short run. The problem is to impose a cost on free riding that exceeds the benefit. In a kind of 'War is Peace' scenario, he argues that ''there is only a single viable way to impose such costs on a free rider -- to inflict violence'' (p. 217). The key, he argues, is the elite ability to throw stones accurately so that the cooperators can ostracize the free rider with violence from a distance, with relatively little danger to themselves. He argues that evidence for elite throwing is present in late pre-human australopiths, followed quickly by uniquely human social cooperation and 2 million years of brain expansion. We might be wary, though, of the exclusiveness of Bingham's argument: that for the evolution of language, the conflict of interest is ''the *sole* limitation'' on the exchange of information; that ''NO OTHER FACTOR than solution of the conflict of interest problem and NO NEW CAPABILITY (ability to generate and comprehend 'combinatorial symbolism' or a new level of individual intelligence, for example) need precede evolution of elite, symbolic communication ... which will evolve WITHOUT THE NECESSITY OF ANY OTHER PRE-CONDITION'' (pp. 215-216, EMPHASIS his).
The chief flaw of this book is its narrowness of scope. The editors express the breathtakingly provincial opinion that ''HCF and related efforts have signaled a symbolic lifting of the ban on investigations into the origin of language officially imposed by the Linguistics Society of Paris in 1871 (p. 1).'' This seems to overlook a great deal of work on the origin of language, including important attempts even within the generative linguistics framework to reconcile generative linguistics and biological evolution such as Pinker and Bloom (1990), twenty years ago, and Lenneberg (1960), early on in the generative enterprise. Likewise there is almost no acknowledgement of such processes as grammaticalization and of the many varieties of cognitive and functional linguistics that have had no disconnect with evolutionary theory. There is little acknowledgement of the 150 years of interaction between neurolinguistics and evolutionary theory, going back to the correspondence between Charles Darwin and Paul Broca in the early 1860s.
It should be noted in passing, that the Paris ban on discussions of the origin of language had an ideological anti-scientific motivation. The date was actually 1866, not 1871, and the ban was one of the original bylaws of the Société de Linguistique de Paris (SLP). As we see on the SLP website, the SLP was founded originally to support catholic and monarchist ideas and to counteract positivist thinking. Specifically it was organized in order to oppose the Société d'Anthropologie de Paris which had been founded by Broca in 1859 (http://www.slp-paris.com/index.html). Ten years later, in 1876, the Linguistic Society of Paris was reconstituted on a less ideological basis as a learned society with a new set of bylaws. The original bylaw banning discussion of the origin of language cannot be taken seriously.
The most important question raised in this book is whether language (involving maybe a mutation for 'merge') emerged recently with ''behaviorally modern humans,'' long after the emergence of anatomically modern Homo sapiens, or whether language coevolved with the changes seen in brain anatomy in the 2.5 million years since the pre-human australopithicines. Arguing for the recent scenario, Tattersall talks of the ''invention'' of language. Does this imply that language was simply a cultural phenomenon, not a biological one? The HCF paper argues that natural selection and adaptation were not relevant to the evolution of language. Does that imply that the biological basis of language was not shaped specifically for language? The majority of chapters prefer a long term evolutionary scenario with full language present at the emergence of anatomically modern humans. The notion of a cultural revolution 50,000 years ago when humans became ''behaviorally modern'' has been convincingly refuted by the archeological evidence presented by McBrearty and Brooks (2000). The genetic evidence likewise does not support the recent date (Diller and Cann 2009). Modern humans had dispersed to all corners of Africa and into the Middle East by 100,000 years ago, to South Asia before the Toba volcanic eruption 74,000 years ago, and to Australia perhaps by 60,000 years ago. Once the human population had dispersed into widely isolated subgroups, chances were minimal for a mutation for 'merge' spreading universally to everyone.
A second major question is whether language emerged for thought with communication added on later, or whether it evolved as a system of communication from the start. The majority of chapters see language as an evolved system for communication.
A third question is the special status of 'recursion' in language. This is clearly controversial: Jackendoff shows evidence of recursion in visual thinking, and others see recursive processes as basic to mammalian cognition.
The majority of chapters do not accept Chomsky's argument that the broadly construed faculty of language was not subject to biological evolution, and that ''Evolution in the biological sense of the term would then be restricted to the mutation that yielded the operation Merge...'' (p. 61). Thus even without critiques from the many varieties of cognitive and functional linguistics, the Hauser, Chomsky and Fitch article does not fare well.
Chomsky envisions the emergence of a neural change for 'merge' that is sudden and perfect, ''more like a snowflake than the outcome of extensive Jacobian 'tinkering' over a long period'' (p. 59). There is a phrase from Chomsky's 1959 review of Skinner's 'Verbal Behavior' that may be appropriate here: the speculation about this magical mutation for 'merge' late in the history of Homo sapiens may be just ''a kind of play-acting at science'' (Chomsky 1959: 39 ).
Chomsky, Noam (1959). A review of B. F. Skinner's 'Verbal Behavior' (New York: Appleton-Centuray-Crofts Inc, 1957) in Language, 35:1, 26-58. Reprinted in Fodor and Katz (1964).
Diller, Karl C. and Rebecca L Cann (2009). Evidence against a genetic-based revolution in language 50,000 years ago. In Rudolf Botha and Chris Knight, eds., The Cradle of Language. Oxford: Oxford University Press.
Fodor, Jerry A., and Jerrold J. Katz, eds., (1964). The Structure of Language: Readings in the Philosophy of Language. Cambridge: MIT Press.
Hauser, Marc D., Noam Chomsky, and W. Tecumseh Fitch (2002). The Faculty of Language: What is it, Who has it, and How did it Evolve? Science 298 (22 November 2002), 1569-1579.
Lenneberg, Eric H (1960). Language, Evolution, and Purposive Behavior. In S. Diamond, ed., Culture in History: Essays in Honor of Paul Radin (New York: Columbia University Press); Revised and extended version published as The Capacity for Language Acquisition, in Jerry A Fodor and Jerrold J Katz, eds. (1964).
McBrearty, Sally, and Alison S. Brooks (2000). The revolution that wasn't: a new interpretation of the origin of modern human behavior. Journal of Human Evolution 39:5, 453-563.
Pinker, Steven, and Paul Bloom (1990). Natural language and natural selection. Behavioral and Brain Sciences 13:4, 707-784.
ABOUT THE REVIEWER
ABOUT THE REVIEWER:
Karl Diller is currently working on the evolutionary genetics of language
in the Cann Lab (genetics) in the PhD program in Biomedical Science (Cell
and Molecular Biology), John A. Burns School of Medicine, University of
Hawaii at Manoa, Honolulu HI. He has a PhD in Linguistics from Harvard
University, and is professor emeritus of linguistics at the University of