Learning from Animals?

Examining the Nature of Human Uniqueness

Louise S. Röska-Hardy, Eva M. Neumann-Held, Louise S. Roska-Hardy, Eva M. Neumann-Held, Louise S. Röska-Hardy, Louise S. Roska-Hardy, Eva M. Neumann-Held

• Publisher: Psychology Press
• Year: 2008
• eBook ISBN: 9781135430238
• Edition: 1
• Topic: Psychology
• Subtopic: Applied Psychology
• Index: Psychology

Human language, cognition, and culture are unique; they are unparalleled in the animal kingdom. The claim that we can learn what makes us human by studying other animal species provokes vigorous reactions and many deny that comparative research can shed any light on the origins and character of human distinctive capacities. However, Learning from Animals? presents empirical research and an analysis of comparative approaches for an understanding of human uniqueness, arguing that we cannot know what capacities are uniquely human until we learn what other species can do.

This interdisciplinary volume explores the prospects and problems of comparative approaches for understanding modern humans' abilities by presenting: (1) the latest findings and theoretical approaches in primatology, comparative psychology, linguistics, and philosophy; (2) methodological reflections on the prospects and challenges of understanding human capacities through comparative research strategies; and (3) discussions of conceptual and ethical issues.

This is the first book to address the issues raised by comparative research from such a diverse perspective. It will therefore be of great interest to students, researchers, and professionals in comparative psychology, linguistics, primatology, biology, and philosophy.

Part I

Language

1 Prolegomena to a science of biolinguistics

W. Tecumseh Fitch

In the first years of the new millennium, the word “biolinguistics” has rather suddenly come into use as an umbrella term for various biological approaches to the study of human language. At least three recent books have “biolinguistics” in the title (Givón, 2002; Jenkins, 2000, 2004), a new journal with that name has just been founded, and the first Laboratory of Biolinguistics (Riken Brain Science Institute, Japan) is producing its first generation of PhD students. Based simply on the divergent contents of the books just mentioned, this nascent field is broad in its interests and incorporates diverse viewpoints, both about what language is and how it should be studied. Despite numerous disagreements, what the scholars embracing this term all have in common is the core belief that the human capacity to acquire and use language is an aspect of human biology, and that it can thus be profitably studied from a biological perspective. While this core assumption of biolinguistics is not a new idea (Chomsky, 1965; Darwin, 1871; Lenneberg, 1967; Lieberman, 1975), it appears to be one whose time has come. The purpose of the current chapter is to survey the potential of this new field, and to highlight some problems that stand in the way of progress.

It is certainly an opportune time for scientists interested in human cognition to adopt a biological perspective, since the tools available to support empirical inquiry into all aspects of biology have suddenly become so powerful. Human brain-imaging techniques are now widely available that allow us to examine neural function noninvasively in normal subjects. After a decade or so of somewhat self-indulgent neophrenology, this field shows signs of maturing into a promising endeavour with important advantages over the patient-based approaches to neurolinguistics that preceded it. These tools will help map the functional circuits underlying language competence, and ultimately help point the way to the underlying computations that are of central interest. Behavioural techniques for investigating language and related cognitive functions, including eye tracking and looking time techniques, are unveiling a complex cognitive world in infants and nonverbal animals that stands in sharp contrast to their limited communicative ability. These accumulating results belie the belief that language is a prerequisite for any form of complex conceptual processing. Finally, and perhaps most profoundly, the revolution in molecular genetics has produced genome sequences of humans, chimpanzees, dogs, and many other species, and gene sequences of humans turn out to be nearly identical in many cases with homologous genes in chimps, mice, flies, and even yeast. We can now both observe and control gene expression in animal models, and the new results from developmental molecular genetics necessitate profound changes in traditional conceptions of “innateness”.

These and other new techniques are generating a flood of empirical data relevant to age-old questions about the development and evolution of language and the mind, data that often demand fundamental changes in our way of thinking about these problems. More than ever before, the biological approach to language has much to offer the linguist, psychologist, anthropologist, and philosopher. Problems that once seemed insuperable, such as interactions between ontogeny, cultural “evolution”, and phylogeny, are slowly yielding to concerted theoretical and empirical effort (Deacon, 1997; Kirby, Dowman, & Griffiths, 2007; Kirby, Smith, & Brighton, 2004; Steels, 1999; Tomasello, 2001). But there is trouble in this potential interdisciplinary paradise, and despite considerable grounds for optimism, it is by no means certain that the new biolinguistic approach will be as successful and productive as it deserves to be. The problems facing a future science of biolinguistics come in two flavours. The first, and the less challenging intellectually, are essentially sociological problems concerning terminology, disciplinary turf wars, and struggles for dominance. A reliance on oversimplified models and outmoded distinctions is another important sociological impediment to progress. Although these problems are easily diagnosed, they may be difficult to solve. Fortunately, some of the more deeply entrenched and recalcitrant disciplinary divides and outmoded approaches seem to be breaking down, and I am optimistic that the next generation of young biolinguists, for whom disciplinary boundaries are more fluid, will eventually leave many of these sociological problems behind.

The second class of problems involve far more profound theoretical difficulties, and constitute some of the most serious intellectual challenges of our time, or indeed that science has ever faced. It is on these difficulties that I focus in this chapter. I see three broad areas of conceptual challenge, each of them related to the others. All three demand fundamental theoretical and empirical progress before we can hope to understand the biological basis for language. The first challenge is neuroscientific: Despite huge progress, at a basic level, we still do not understand how brains generate minds. This is as true of a dog’s brain as for a human’s, and it is true of very basic aspects of cognition, such as vision and motor control, along with language. The most fundamental neurolinguistic questions concern the basic computations underlying language use, and their specific neural basis. Current attempts to address this question remain on a shaky theoretical footing. The second challenge concerns genes and development: How do genes control the development of a single-celled zygote into the trillions of integrated cells comprising a complex behaving organism? Again, great progress has been made, and the new epigenetic paradigm allows us to reject long-reigning models of the genome as blueprint. However, the complex and circular nature of epigenesis, and the resultant causal indirectness of development, still pose serious conceptual challenges. Although we now understand in some detail how physical structures like the vertebrate limb develop, the principles underlying brain development and evolution remain only dimly understood.

Finally, while the neurocomputational and developmental difficulties are basically biological, and apply to any aspect of cognition, the last and most profound difficulty concerns language more specifically. This suite of problems concerns questions of meaning. Put simply, we have a good theory of information (Shannon information theory), but we lack anything even approaching a good theory of meaning (what I intend with this information/ meaning distinction will become clear below). Problems of reference, relevance and context-dependent interpretation remain central unresolved issues in the philosophy of mind. The first two problems have matured to a stage where they appear to be accepted as problems of the empirical natural sciences, but these last problems remain in the philosophical category (we don’t even know how to devise experiments to help sort the issues out). Although these unsolved semiotic challenges pose problems for any aspect of cognition, e.g., what is it that happens when an organism interprets some stimulus as “meaningful”, they become particularly acute when discussing language, which is that aspect of cognition centrally concerned with meaning.

Recent reviews of new approaches and data in biolinguistics are already available (Fitch, 2005b; Johansson, 2005). Therefore, my goal here will rather be to outline and clarify the problems facing the field. As one interested in seeing this field grow and flower, I intend my critical comments to be constructive. I have been working in “biolinguistics” (without knowing it) for the last 20 years, since my decision as a young marine biologist to refocus my efforts on the evolution of language (Fitch, 1994; Fitch, 2000b, 2002b, 2005b). Although I remain optimistic, I have become acutely aware of the difficulties facing the field, in part because successes in various areas have brought the remaining problems into sharper focus. Through my involvement in a recent interdisciplinary foray in biolinguistics (Hauser et al., 2002) and the rhetorically charged debate that followed (Fitch, Hauser, & Chomsky, 2005; Pinker & Jackendoff, 2005), I have also developed a healthy, if depressing, awareness of the sociological problems that await attempts at interdisciplinary bridge building.

In this chapter I will briefly discuss the sociological problems and disciplinary strife that arise from choices in terminology and differing conceptions of “language”. These pose important but soluble problems for those with a bona fide interest in solutions and will not be my core focus here. I will then outline and clarify some of the deeper intellectual challenges facing biolinguistics, discussing why many currently popular models and metaphors for understanding genes, brain, and language need to be abandoned if we hope to make substantial progress. In some cases I will also try, tentatively, to sketch approaches to the problem that appear to me to offer promise. But I will be satisfied if the reader, accepting my critique of the “state of the art”, rejects my proposals for remediation. Each problem alone is extremely difficult and, combined, as they must be in biolinguistics, even guessing at plausible answers is difficult. Thus, as with any prolegomenon, my focus here is making the problems sharp and clear, rather than offering solutions.

SOCIOLOGICAL CHALLENGES: DISCIPLINARY DISCORD AND TERMINOLOGICAL DEBATE

The chance that the key ideas of any professional scholar’s work are pure nonsense is small; much greater the chance that a devastating refutation is based on a superficial reading or even a distorted one, subconsciously twisted by a desire to refute.

(Langer, 1962, p. ix)

This wise insight accurately diagnoses much contemporary “debate” in biolinguistics, particularly concerning the evolution of language. I know of no other field where scholars seem so ready to champion their own pet hypothesis uncritically, while rejecting those of others as absurd or ludicrous. I confess to finding some of the proposals in the literature uncompelling or even ridiculous on first reading, e.g., Calvin’s “throwing Madonna” hypothesis (Calvin, 1983), or the “aquatic ape” hypothesis (Morgan, 1997), but further reading and thought have convinced me that some valuable insights, and probably germs of truth, are to be found in such ideas—for one willing to put in the work of understanding them. Unfortunately, such willingness is too often in short supply, and debate in the biology and evolution of language frequently reduces to either misrepresentation (dismissals based on straw-man caricatures) or arid terminological debates (I dislike the term X for some trait and propose term Y for the same thing). Often the two are combined. This syndrome is particularly true of criticisms of Noam Chomsky, whose ideas so many scholars apparently love to hate. In my opinion, once placed in context and properly understood, most of Chomsky’s scattered statements about both language evolution and its biological bases either are rather uncontroversial statements that any modern biologist studying (say) limb development would accept as a matter of course, e.g., that there must be various biological constraints upon the development of the language system, or present unpopular alternative hypotheses that deserve more careful consideration, e.g., language as a tool for thought rather than communication. Outside of his technical linguistics work, Chomsky’s main contribution to biolinguistics is his long championing a scientific approach to language as a biological phenomenon (Chomsky, 2005). One will search in vain in Chomsky’s own writings for the naïve conceptions of Universal Grammar for which he is so often mistakenly pilloried—one reason his critics typically quote his few scattered statements out of context, if they quote them at all.

My purpose in this chapter is neither to champion nor attack Chomsky’s conception of language (for this see Jackendoff, 2002; Jenkins, 2000; Lieberman, 2000)—but rather to argue that such discussions too often miss or leave unmentioned deeper commonalities of viewpoint and approach shared by most contemporary theorists interested in the biology of language. In the next sections I will try to look past the terminology at some uncontroversial facts about the biology of language, briefly discussing the terminological controversies they have driven. My purpose is to shed the rhetoric and move into the conceptual heart of biolinguistics. This will set the stage for the main part of the chapter, where I discuss the core outstanding conceptual difficulties in detail.

“The human capacity to acquire language”: The core explanandum

The central research topic in biolinguistics is a characterization and explanation of the human capacity to acquire and use language. That this is an aspect of human biology is made clear by the everyday fact that any normal child raised in a human household will quickly, and apparently effortlessly, acquire the language(s) of its family and community, while no nonhuman animal will do the same. The pet dog or cat may learn quite a bit about the social and practical aspects of life in a human household, and often to recognize a few dozen spoken words of the local language, but its abilities to express its own thoughts using this language are little different from those of a potted plant in the living room. Perhaps more surprisingly, a chimpanzee raised in a human home will not spontaneously do much better: Even with long concerted training apes learn to produce only an indistinct handful of spoken words (Hayes, 1952). Although use of the manual/visual modality via sign or symbols helps apes considerably (Gardner & Gardner, 1969; Premack, 1971), the adult ape still cannot progress to anything like the level of a 5-year-old child, and its “linguistic” utterances will be mostly confined to requests for tickles or treats. Without belittling the accomplishments or value of such experiments (cf. Savage-Rumbaugh, 1986; Savage-Rumbaugh et al., 1993), it is important to acknowledge these limitations as well-replicated biological facts.

Clearly, immersion in a linguistic environment is not enough for spoken language to develop in most organisms. There must therefore be something about human children which differentiates them from other species, and this something provides one of our core explananda in biolinguistics. We might gloss this neutrally as “the human capacity to acquire language”. In generative linguistics following Chomsky this capacity is traditionally called the “Language Acquisition Device”, and a characterization of its properties termed “Universal Grammar”, reviving a seventeenth-century term. Universal Grammar (before Chomsky) simply designated those aspects of human language competence which, because they are shared by all humans and all languages, went unmentioned in traditional grammars (Allan, 2007; Chomsky, 1966). For example, the notion that words exist and have specific meanings does not need to be specified in a grammar of French—it can be taken for granted. But this is precisely the sort of fact that does need to be explained by a successful biological approach to language. The original usage of the term made no particular claims about the nature of this competence, e.g., that it was specific to language, or conversely a general aspect of human cognition, nor did Chomsky’s revival of the term, which is quite neutral on such questions by my reading. However, both “LAD” and especially “Universal Grammar” arouse suspicion and rejection from scholars who nonetheless accept that such a human-specific biological capacity exists (Lieberman, 1998a; Tomasello, 1999, 2005b). A huge amount of ink has been shed rejecting the term “Universal Grammar”, even by people who accept without question that a biologically based capacity to acquire complex language fully is a uniquely powerful birthright of any normal human, but no known animal. The substantive debate concerns not the existence of such a human capacity for language acquisition, which is abundantly clear regardless of terminology, but rather its nature, e.g., the degree to which it is specific to humans, or to language.

There remains, today, no widely accepted term for this central aspect of human biology, despite the consensus about its existence. A recent attempt to break the resulting terminological logjam by introducing two new terms—the faculty of language in broad and narrow senses (FLB and FLN; Hauser et al., 2002)—unfortunately elicited similar reactions (Pinker & Jackendoff, 2005), although FLB was specifically and explicitly intended to capture a much broader and more inclusive conception of the language capacity than the one connoted by LAD or UG. FLN was intended to have a considerably narrower scope, perhaps even denoting an empty set, but has been read simply as “language” by some and “UG” by others. The term “language instinct”, popularized in Pinker (1994) has been rejected equally vehemently (e.g., Tomasello, 1995a). Frankly, it is unclear to me whether any acronym or shorthand version of the “human capacity to acquire language” will escape a similar rhetorical assassination. Perhaps the field of biolinguistics will have to do without any such term for the time being (although I would personally vote for “language acquisition capacity” as a relatively neutral designation).

“Innate knowledge”

A similar terminological morass surrounds the term “innate”, and particularly the concept of “innate knowledge”, although the problems here are at least partly substantive rather than terminological. The deep conceptual problem ultimately stems from the complexity of epigenesis, the complex interaction between developmental programmes and the internal and external environment of the developing organism (cf. later). But the terminological problem hinges on what we are prepared (or inclined) to call “knowledge”. Knowledge is prototypically a representational state of adult minds, implemented somehow in their brains. We know enough about neuroscience to say that this implementation will involve the morphology of individual neurons, their connections with other neurons, and the computational activities these neural circuits engage in. From this perspective, it would be odd to ascribe “knowledge” to genes, or to the just-fertilized egg. But what about the newborn infant’s “knowledge” of language? Here we are on uncertain ground, for the child is certainly born with a brain, equipped with proclivities to attend preferentially to certain things (like human voices) and not others (like dog barks or engine noises). Even at birth the newborn already expresses preferences f...

Table of contents

• Cover Page
• Title Page
• Copyright Page
• List of contributors
• Foreword
• Acknowledgements
• Introduction—Issues and themes in comparative studies: Language, cognition, and culture
• PART I LANGUAGE
• PART II COGNITION
• PART III CULTURE
• Postscript: Human uniqueness in a comparative perspective
• References