The Cambridge Companion to Chomsky, Cambridge, 2005
At the time of writing, Noam Chomsky has produced over eighty books, hundreds of articles, and thousands of speeches. He has given thousands of interviews, written countless letters, and supervised scores of theses. He has made important, sometimes groundbreaking, contributions to three areas – linguistics, philosophy of mind and human nature, and politics. He set linguistics on a successful naturalistic, biologically oriented scientific course; his theoretical contributions continue to lead the field. Like Descartes, Galileo, and Hume, and unlike the eighteenth-century philosopher Kant and the great majority of philosophers thereafter, Chomsky is both scientist and philosopher, and his philosophical work is continuous with his scientific. His science of language and incipient science of mind offer a genuine prospect of coming to a biologically based grasp of human nature and of the way it allows for human understanding and action. His political work, like both Hobbes’s and Rousseau’s, seeks a foundation in a science of human nature, although with better prospects for developing such a theory – and for exploring its implications for political ideals and goals – than Hobbes’s misguided attempt to construct a causal theory of human action or Rousseau’s fanciful assays into a “state of nature.” And unlike both of them – and far too many contemporary political “theorists” – there is no sign in Chomsky’s political work that his views and critical analyses are driven by a wish for power.
One purpose of this volume is to offer to a general audience several people’s perspectives on Chomsky’s contributions in linguistics, philosophy of mind and human nature, and politics. The first chapter in each section provides an overview of Chomsky’s views in these areas. Succeeding chapters develop major themes. I sketch some of those themes and how contributors develop them near the end of this introduction. A sketch suffices: the chapters and organization are self-explanatory.
Chomsky the scientist of language
Linguists in the Chomskyan tradition think of themselves as natural scientists – not social scientists, and not engineers. It is important to see what this implies. Ordinary usage is little help. The term “science,” like “language,” has no unique use in everyday speech: people apply it to everything from physics to astrology. And – in part because being a scientist is associated with expertise, specialized knowledge, intelligence, etc. – the desire for social status and political authority leads to applying the label “scientist” to some questionable candidates. Given this, we cannot expect more than a few hints about what science for the Chomskyan is by looking to the practices that have been called scientific, or to the range of people who have been called (or call themselves) scientists.
A more reliable source is the history of science and the shapes of the sciences that are universally agreed to be successful – physics, chemistry, biology . . . Their subject matters and degrees of progress differ, as do their principles, experimental techniques, and outstanding problems. There are, however, enough similarities to draw a composite sketch, especially where the characteristics chosen agree with those from other reliable sources.
Another such source is what those who began the development of successful sciences said they were doing. Chomsky often mentions Galileo and Descartes in this regard (e.g. 2002b); he considers himself to be working in a tradition of philosopher-scientists that they began. These pioneers developed and applied recommendations for how to proceed in carrying out investigations of natural phenomena that led to what were for their times remarkable successes. Focusing on Descartes in his Discourse – a work that explains how Descartes came to his scientific principles, says what they are, and outlines what he accomplished by using them – it is striking that he divorces science from another kind of understanding of the world. No one uses scientific concepts in solving the myriad problems encountered in everyday life. Everyone, including the young child and the scientist, has and uses what Descartes called “bon sens” (“good sense”), a practical form of problem-solving capacity that Descartes considers innate – a gift from God. “Bon sens” is sometimes translated as “common sense,” and I will adopt that term. It is a capacity to deal with the problems of politics and commerce, doing the laundry, consoling a grieving friend, and putting out the dog.
While everyone relies on common sense, it does not assume a single form for all times and all circumstances. This is a benefit. Practical problem-solving must accommodate differences in method and individual style, different environments, cultures, and social organizations, and so on. To do so, common sense must rely on rich and productive native (innate) resources and a flexible form of mental organization. That is how it can arise so early in children and be so remarkably adaptable. Where Descartes said that bon sens is a gift from God, we are likely to say that its rich resources are biologically based. It is only thus that we – even the very young – can so quickly conceive, anticipate, and adapt to different environments, adopt (even if only in play) different social roles, and quickly change to meet unanticipated contingencies.
The contrast to science is instructive. Science is an intellectual project that – where successful at all – uses formal (mathematical) theory-construction techniques to focus on specific domains; it is guided by a desire for simplicity and – as Galileo and Descartes’s work shows – it places simplicity before “data”; it makes progress (often in jumps) over centuries through contributions from many people towards solutions to the theoretical problems it constantly confronts, revises, and refines; and it creates its own standards of intelligibility (Chomsky 2002b: 68) that are far from the practical concerns of common sense. Physics, for example, has taken centuries to develop, has advanced in spurts, uses mathematical techniques to describe a world populated by entities and processes beyond the ken of commonsense understanding; and while no doubt far from complete, it has obviously progressed well beyond Galileo’s and Descartes’s “mechanical philosophy.” This punctuated but deliberate pace is probably necessary because science does not rely on the rich and productive native systems, flexibly organized, that common sense utilizes. While it can and obviously does rely on apparently innate senses of simplicity and what counts as a good explanation and description (Chomsky 1980), construction of the theories that solve the problems science confronts requires invention, favorable conditions, and cooperative activity. That is why – with the exception of parts of mathematics – only rudimentary forms of natural science developed before the end of the sixteenth century. Unsurprisingly, it also takes a considerable amount of time and training for individuals to acquire sophistication even in a specific science; the full range of the developed sciences is out of the reach of everyone. Fortunately, science’s findings are not needed for survival, or even to thrive. No doubt doing laundry benefits from engineering applications of fundamental scientific principles – those that lead to variable-speed electric motors and front-loading washing machines. But for millennia people managed with technology that required only the engineering solutions offered by unaided common sense. They built bridges of various materials, annealed metals into Samurai swords, and constructed cathedrals. In sum, science brings the developed formal tools of highly focused inquiry to bear on theoretical problems; progress – relying as it does on invention – is usually slow. Using their common sense, people utilize native resources, perhaps in forms of practice that have led to practical success before, to deal with the immediate demands of everyday problems. We invent scientific tools to deal with bosons and genomes; we depend on native resources to critically assess the performance of an elected representative – or the intentions of an artist.
One of the characteristics of scientific practice that science’s history reveals is seeking a particular kind of objectivity – one that is universalized, so that it is not tied to person, circumstance, culture, or history. That notion of objectivity cannot serve the tasks that common sense deals with; commonsense understanding’s concepts are “designed” to serve matters of human interest – including those of perception and action. It should be no surprise, then, that science can lead to denials of the “obvious” claims of commonsense understanding. Sciences of the mind tell us that the colors we experience are products of our visual systems, not “on” things outside, and that languages – including their sounds and meanings – are native and in the head, not somehow outside the head, perhaps products and properties of communities and polities. Scientists often must ignore appearances (as in colors and words outside the head) and invent, using the tools that mathematics – much of which is invented too – provides. And they must measure progress not by how well a proposed change in a theory satisfies untutored opinion or “raw experience” but by improvements in description and explanation of the relevant phenomena and greater formal simplicity. Descartes – had he lived long enough – would have seen his contact mechanics refuted by Newton’s gravitational principle. The “obvious” idea that action and effect require contact seems to have its origin in common sense; it fails in science. Physicists learned long ago that the apparently obvious is at best a starting point. That lesson has been hard to learn with language, as we will see.
Chomsky’s science of language is a science in the Cartesian–Galilean tradition. It is a branch of the study of biology. It is a naturalistic science that provides an “abstract” description and explanation of a biological system found only in humans, the system that Chomsky calls “the language organ.” The language organ revealed by Chomskyan science of linguistics is far from the commonsense idea of a language as a social phenomenon. To reveal this organ, the science of linguistics had to develop standards of intelligibility that were consonant with those of the natural sciences, not with what some philosophers call “granny’s view” of language. The result, after several decades of work, is that the language organ appears to be remarkably simple in its “design.” This is unusual in biology, a domain that usually reveals what Jacques Monod calls the “tinkering” of evolution. Apparently, extending naturalistic science to the study of a biological system of the mind yields a fascinating result: language confirms Galileo’s and Descartes’s vision of a well-designed, elegant nature.
Descartes’s mechanics – even in its rather primitive form – conflicted with the “obvious” principles taught by church and universities. It conflicted with the teleological world of Aristotle, modified by the seventeenth century to suit Christian doctrine. That is why his and Galileo’s novel mathematical–mechanical theories of natural phenomena faced opposition from philosophical and religious systems that, like many today, take their task to be that of defending ideas that have their origin in common sense with its practical, not theoretical orientation. Successful sciences since Galileo and Descartes have continued to use simple, elegant, formal mathematical tools and invented theories and concepts to provide descriptively and explanatorily adequate theories of their domains. Like Galileo’s and Descartes’s sciences, they continue to be opposed, although some of the opposition is muted by the obvious success of the theories.
Chomsky’s naturalistic inquiry into language, like Descartes’s into cosmology, physics, optics, and neurophysiology, also gets opposition from the experts. Opposition comes from several fronts, but most seem to proceed on the assumption that language is not a natural phenomenon. They see language in terms of its use – perhaps as a set of social practices, a bunch of “tools” we have made to communicate, etc. In each case, one finds a version of what Chomsky calls an E-language approach to theorizing about language (external approach). Among the majority of philosophers, it appears as insistence that one or another form of the “obvious” idea that language is an institution created by humans to communicate – a “practice,” a product of history, a set of habits, an “interpretive medium,” a mode of communicating a speaker’s intentions. From psychologists, philosophers, and other cognitive scientists wedded to one version or another of what Steven Pinker (2002) calls the “blank slate” picture of the mind, it comes as the idea that it is a form of behavior that solves cognitive problems such as classifying, describing, etc. Any organism or device that displays the “same” behavior has a language, they believe, and they “get” it by whatever means (training, programming) the blank slate advocate employs. So getting an ape (on which see Petitto, this volume) or a machine to simulate the behavior “proves” that language is not a biological organ with which only humans are born.
Perhaps these experts succeed sometimes by some standards – it is not clear which – but not by those of naturalistic inquiry. Philosophers who think language is a social phenomenon that children learn from their community ignore the fact that languages are quickly acquired by very young children without training. The same is true of concepts of social role. If neither language nor concepts of social role (and much else) is taught, they must somehow be built into the child’s mind at birth; that is where to focus naturalistic inquiry. And blank slate advocates need to learn the elemental lesson that it is unwise to focus attention on sameness of behavior or “output” and the means by which these are induced. Even if one succeeds at getting a machine or ape to “speak” – it has not happened, probably for reasons that Descartes (and Alan Turing) pointed to – that is no proof that the systems that make this behavior possible (which is where Chomsky focuses his work) are the same as the ape’s or the machine’s. Manufacturing an excavator to dig ditches hardly proves that human gravediggers moving shovels with their articulated arms have hydraulic systems in their arms activated by diesel-powered compressors.
Chomsky on biology and evolution
To avoid confusion that might arise from speaking of Chomsky’s view that language is a biological organ, I need to mention the matter of evolution (for detail, see Jenkins 2000). Chomsky, like Richard Lewontin (1990), has little sympathy for current efforts (e.g. Pinker & Bloom 1990) to try to show that language – especially in the form of the basic computational system that links sounds and meaning to produce a discrete infinity of sentential expressions – is the product of some sort of natural selection that tracks increased reproductive advantages afforded to those who are (on the Pinker–Bloom story) better communicators. Chomsky does not doubt that language evolved, in some sense: it is biologically based and appeared in the human species. And he has no doubt that it has proven to be extremely useful to humans. But selection-for-communication, selection-for-some-function-or-another, and even selection, period, do not exhaust the field. Pace Pinker and Bloom, there are alternatives.
One problem with the attempt to show that language was selected by reproductive advantage is that humans as a species have been relatively stable for a long time – probably 100,000 to 200,000 years. So language in the form of the basic computational system that seems to be unique to us must have emerged somewhere between now and 100,000 to 200,000 years ago. (The best current guess is approximately 60,000 years when the migration from Africa began.) It is all but impossible to find evidence in observable phenomena for the selectional emergence of such a system. Perhaps we will someday identify the computational-system specific gene(s) that provides us our languages’ syntax; perhaps too by investigating remains we could say when this gene was introduced. But nothing would tell us why and how it developed. Speculation about selection-for-any-function of language’s computational system (its “syntax”) seems to be empty.
But that is not all that needs to be said. For one thing, while looking for a historical record seems hopeless, we can compare. Hauser, Chomsky, and Fitch do just that in their article in Science (2002). Comparison offers no immediate help to the Pinker–Bloom selection-for-communication cause, however: language’s basic computational system can produce a discrete infinity of sentences, and there is no current evidence that other species can “express” discrete infinities of elements of any sort. For example, no other species enumerates arbitrarily large numbers of elements in a set by counting them out. But perhaps we have not looked far enough. Most comparative studies focus on human and animal communication systems; perhaps the computational system is found elsewhere. In concluding their discussion, Chomsky and his co-authors suggest looking at other kinds of system: look, they suggest, for non-communicative systems that rely on a recursive computational procedure that provides for a discrete infinity. Perhaps they will be found in navigational systems, or those that “parse” social relationships. If there were such a system, it – or its homologue or analogue in the developing human species – might have been exapted (coopting a system adapted to serve another function) for language. Selection-for-communication would fail, but perhaps aspects of the selectional cause would be salvaged. It is a project worth trying.
In his own work, Chomsky suggests we look wider still. Darwin pointed out that selection is only one part of evolution. And there are other, non-selectional traditions of biological development and speciation. One such tradition is found in Stephen Jay Gould’s and Richard Lewontin’s suggestion (1979) concerning “spandrels” – structural consequences of other, perhaps selected, systems. Another tradition, perhaps related, goes back to Goethe and his discovery of an Urform for plant morphology (cp. Chomsky 2002a: 66). Goethe thought he had discovered a formula that predicted all (biologically) possible forms plants could take. If there is such a formula, it indicates that there lies in plant morphogenesis a physical factor that yields very different-appearing plant forms (and “species”), given slightly different “input” conditions. The formula and different physical conditions, not selection, would account for differences. That tradition was represented in the nineteenth century by several individuals in Europe and, in the twentieth century, versions of it appeared in mathematical form in the work of D’Arcy Thompson (1917) and Alan Turing (1952). Many have pursued their suggestions; there is a growing mathematical science of morphogenesis.
Chomsky sometimes suggests (2002b: 57) that the development of the language organ might be explained as a mathematical consequence of the kind of complex form of mental biology humans have. Even selection has to operate within the “channels” provided by basic physical processes, after all; and our language faculty appears to be too “perfect” a solution to linking sounds and meanings to be the result of selectional tinkering. Perhaps the computational system built into our languages is “anticipated” in those physical processes and structures and arises “by itself” when other systems are in place. This would allow that the computational system of language came about as a complete package, perhaps 60,000 years ago. Or perhaps language is a spandrel. Either way, we abandon the gradualist, “historical” form of development and biological differentiation (and the “tinkering”) that the selectional picture relies upon. We might even be able to find evidence.
In sum, while there is no doubt that language has proven to be an extremely useful biological faculty that has given the human species extraordinary cognitive advantages, there is little reason now – and there may never be – to hold that the computational core of the language organ developed slowly over a long historical period by virtue of affording reproductive advantages to successive generations of communicators.
On the unity of Chomsky’s thought
A person’s intellectual work as a scientist need not be connected to his or her political views – there is no reason, for instance, that a biochemist’s scientific work should have anything to do with her neoliberal views. But Chomsky’s linguistics and his political views seem to be special cases, particularly when one takes into account his philosophical?scientific work on the human mind and human nature.
One reason to look for connections and perhaps even a degree of convergence in all three areas of Chomsky’s work is that each has, in its own way, something to say about human beings. More narrowly, each focuses on distinctive features of human beings – on language, a biologically unique mental faculty; on our distinctive natures and minds with their limited but biologically unparalleled intellectual capacities for dealing with both practical and scientific problems; and on those apparently unique forms of social organization that we think of variously as polities, communities, societies, and?or cultures. No other organism creates for itself organized groups of non-kin individuals in ways that allow for cooperative, non-contact, coordinated ways to meet needs and solve problems.
Some non-Chomskyans think that there is a connection between language and culture or community. Philosophers as varied as Foucault and Putnam and psychologists as different as Piaget and connectionists share the assumption that language depends on the society, culture, community (etc.) in which one is raised. By “depends” I mean not just that children born into a group of Japanese speakers come to speak Japanese; everyone acknowledges that and tries to explain why it happens. Rather, they take language to be constituted by the society or community in which one is born. This idea often appears as the view that children are taught the language of their community by their elders: the elders know the “rules of correct usage?correct practice” (given relevant circumstances) and instruct by encouraging correct verbal responses and discouraging incorrect. People (as a group), over a long historical period, are believed to have invented the practices that define a specific community, culture, etc. and, while doing so (or perhaps in doing so), to have also invented language – another form of practice that happens to allow individuals to communicate, coordinate, etc. Individual creativity in the exercise of one’s intellectual powers does not figure in this story; the focus is on community practices?habits?rules for applying words correctly, etc. If defenders of this idea speak of human nature at all, they make it a historically conditioned notion: as people’s fundamental practices change, people’s social?cultural “natures” change. Alternatively, they might say that human beings are plastic (people are intellectual?cultural blank slates), so that human social?cultural natures are – unlike those of any other biological species – molded by the societies, cultures, etc. in which they are born.
Chomsky’s view of language and the mind reverses priorities. For him, human languages are not expressions of culture and society – in effect, human artifacts. They are, in a sense, expressions of our genes: all the existing and possible natural languages (not technical symbol systems, such as those found in the sciences) are biologically encompassed within what he calls “Universal Grammar.” If there is any dependency between language so conceived and society, culture, etc., it cannot make culture the condition of language. If anything, culture (etc.) depends on language. Suggesting that culture depends on language in this sense is not making a causal or deductive claim. The language organ does not secrete cultures or social arrangements. Rather, language provides the rich, unlimited set of conceptual structures (Chomsky informally calls them “perspectives”) and the opportunity to communicate them that humans need to conceive of alternative ways to solve the problem of how to live together to the benefit of all, to discuss and come to agreement on the options, and the like. In effect, language and our other cognitive resources, but especially language, make it possible to create cultures and much else. Adopting this point of view – that native conceptual tools, and especially language, must be in place before articulated conceptualization and understanding, much less discussion, can occur – another matter falls into place too. Individual creativity – a curiosity on the culture-first approach – can now be seen as benefiting from the infinite scope of linguistic output of which our systems are, in principle, capable.
To see why language should have a central role in making sense of how we come to create our diverse communities and cultures – and individual cognitive and expressive styles – it is important to keep in mind that humans are the sole species to have language. Many other species have communication systems. And some others also have the “performance” systems that are involved in human language: auditory perception and production (for speech), visual perception and aspects of articulatory shaping (for sign), plus aspects of those resources that Chomsky calls “conceptual and intentional” – those non-linguistic resources that can be brought to bear on circumstances to yield various forms of intelligent behavior. But no other species has the capacity to develop a potentially infinite, discrete set of mental “outputs” in the form of expressions or sentences that link perception-related configurations, whether sound or sign, with conceptual materials (Hauser, Chomsky & Fitch 2002). That is, no other species can produce – apparently at will – innumerable sets of sentences or expressions. Given the obviously central role of language in human thought and action, our distinctive mental capacities – found in both practical and theoretical problem-solving – may be due, in large measure, to language. And, with these capacities, we also can develop social organizations: we can plan, organize, decide to cooperate, and create institutions. It becomes quite plausible that culture and our various forms of social organization depend on language rather than the other way around. So we have one connection between the areas Chomsky works on: the science of language might well provide the key to what is distinctive to our minds and natures, to making sense of why we have the distinctive mental capacities we do and, in turn, making sense of how we can create our various forms of social organization.
Another kind of connection depends on the fact that views of human nature are always behind people’s attempts to justify their moral and political principles. In the background of every political and moral “ism” (including those largely indistinguishable forms of corporation-dominated plutocracy–oligarchy called “neoliberalism” and “neoconservatism”) one finds assumptions about human nature – about what human beings “are” and what they are or are not capable of. These views of human nature typically play a justificatory role. “That’s a silly view of democracy,” someone might say of the fully participatory form Chomsky favors, “people (of their natures) don’t have enough interest, intelligence, knowledge, or time to participate fully. We must give an elite managerial class the power to make decisions and run the economy, government, courts . . .” Someone else might say: “People are naturally aggressive and acquisitive. A good form of government must have full authority to restrict their unchecked exercise (a Hobbesian state of nature); we need authoritarian government to provide a form of rescue.”
While justification of this sort is common, few of those employing it bother to elaborate their view of human nature. And the connections between whatever degree of articulation one finds and the moral?political?religious . . . claims they are supposed to justify can be quite hazy. Moreover, there is little if any effort to show that one’s view of human nature is itself justified by the standards of empirical inquiry. A biologically based science of human nature would avoid these problems. Appeals to gods and revelation, or to what seems to be obvious to some group or another, are almost always self-serving efforts which reveal a desire to place or maintain oneself or one’s group in a position of power or authority. We need a detailed, objective view of human nature, and scientific inquiry can provide that. It alone can say what is distinctive about our natures – as opposed, say, to those of various other primate species.
A plausible way to focus such an inquiry would be to look for aspects of the human mind that are distinctive – for faculties or forms of mental organization that humans have that other primates lack. The faculty of language, clearly, is such. A science of language and of what language provides humans should thus have an important role in such a science. Not only does language seem to be unique to humans, but it also seems to contribute to a unique form of mental organization. The biological faculty?organ of language acts rather like a central cognitive system, allowing us to coordinate materials provided by other cognitive systems in ways that other creatures seem to be unable to manage. And, of course, it provides the conceptual tools to allow us to speak of anything, anytime. In these and other ways, it enables us to “solve problems” in a wide variety of manners. It is almost as if language allowed our minds to be “universal instruments,” to use Descartes’s Discourse terminology. So Chomsky’s science of language, even in its incomplete state, represents a good beginning to a science of the human mind and, thus, of human nature.
While the science of human nature is in its earliest stages, we can use what we have now to begin to think about how to craft a good society. We cannot move directly from biology – specifically, the biology of the human mind and what it provides us (for these are what make us distinctive in ways that we so obviously care about) – to a picture of an ideal. We must start by deciding what an ideal society should accomplish – what its function(s) should be. For social organizations are institutions, made by human beings, and a good one should fulfill its function well. A plausible suggestion is that the function of a human society is to meet not just the needs of survival, but those that are characteristic of the kinds of creatures we are. Call these characteristic needs of humans “fundamental distinctively human needs.” Now what the science of mind tells us about language and the rest of the mind, and about how people use these cognitive tools, can come into play. To be brief, because we have language and language seems to be the key to our extraordinary mental capacities, we alone seem virtually designed to be creative creatures. Our languages provide an unlimited range of “perspectives” (Chomsky 2000a: 150, 180), and these can be – and are – used to serve all sorts of purposes, including those of art and labor. Language’s unlimited range comes to play a role in virtually all our affairs – not just our thoughts and efforts to understand others, but in our jobs and everyday tasks, even putting out the dog. An ideal form of social organization must, then, give individuals ample opportunity to exercise their creativity. This need not mean that we must all become craftsmen and painters or composers. It might mean that if we labor with others in a factory, we have sufficient freedom and opportunity to fully contribute to all decisions that concern us, to bring about change, and to otherwise control the conditions under which we work. Or it might mean that operating an excavator, we not only do the job well and with a concern for those who will use what we do, but with a form of artistry.