INNATE IDEAS, NATIVISM

Nativism, or the doctrine of innate ideas, is the thesis that human beings possess at least some substantive knowledge innately. The doctrine has long been the subject of intense controversy among philosophers, and since the late 1950s, among cognitive scientists as well. It is generally understood in opposition to the doctrine of empiricism, according to which all substantive human knowledge derives from sense experience.

Proponents of nativism argue that experience alone cannot account, either de facto, or in principle, for the extent and specific content of human knowledge. Arguments of the first type focus on the type and amount of information actually available to a human being during a given period of empirical experience, and purport to show that the information contained in that experience is insufficient to account for a person's manifest cognitive achievements at that point. Such arguments have become known by the term introduced by linguist Noam Chomsky as "poverty-of-the-stimulus" (POS) arguments, and constitute the most common way of defending nativist hypotheses (Chomsky 1957, 1965, 1975). Arguments of the second type are transcendental in character. They purport to show that certain kinds of knowledge are preconditions of empirical learning, and could not, therefore, be the products of such learning. Philosopher Jerry Fodor, who posits an innate system of representation, is probably the most well-known contemporary author of an argument of this type (Fodor 1975). In both kinds of argument, nativists tend to focus on the particular mechanisms of knowledge acquisition presumed or posited by their opponents.

Opponents of nativism also employ a variety of strategies. Some attempt to rebut POS arguments, by showing that the relevant body of empirical experience does, in fact, contain the right types and amount of information to account for human knowledge, and that actual human learning mechanisms can, in fact, extract it. Others critics attack the doctrine directly, arguing either that the notion of innate knowledge is hopelessly obscure, or that it cannot be made scientifically respectable. Still other opponents contend that the doctrine properly understood is trivial—that it posits nothing an empiricist need reject.

Indeed, philosophical discussion of nativism has tended to focus on the meaning of the doctrine, and in particular, the operative terms, "knowledge" and "innate." Philosophers have traditionally made a provisional distinction between declarative, or theoretical knowledge—"knowingthat- "—and practical knowledge—"knowing-how." It is one thing to attribute to someone practical knowledge of, for example, logic, and quite a different thing to attribute to her declarative knowledge of logic. Normal human beings all have the ability to reason in accordance with the laws of logic, whereas few of us can articulate those laws, or even recognize them if they were laid out explicitly before us. Accordingly, the claim that some practical ability is innate is, on its face, different from the claim that some bit of declarative knowledge is.

The term "innate" has both a categorical and a dispositional sense. In the categorical sense, a trait is innate only if it is manifestly present at the creature's beginning; in the dispositional sense, a trait can be innate if the creature is disposed to manifest it under normal conditions. Now it may well be that a newborn human infant lacks the ability to reason—the relevant neuronal connections may not yet have formed—so that practical knowledge of logic is not innate in the categorical sense. Still, a normal infant is disposed to acquire the ability to reason, in normal circumstances. That is, the infant is so constituted that the necessary neurological connections will be formed, provided the infant gets enough to eat, suffers no head injuries. Practical knowledge of logic should therefore be counted as innate in the dispositional sense. The same point can be made with respect to declarative knowledge: It could be innate in the dispositional sense even if it is not innate in the categorical sense.

There are further complications. Knowledge can be either explicit or implicit. While it could be argued that few individuals have explicit declarative knowledge of the laws of logic, it is clear that everyone has—in virtue of being a competent reasoner—implicit knowledge of those laws. Contemporary cognitive science introduces another wrinkle: Knowledge can reside at the "personal" or at the "sub-personal" level. Personal-level knowledge is knowledge that can be consciously accessed, and that can play a role in conscious deliberation. Sub-personal knowledge is posited to reside in cognitive "modules"—subsystems that perform cognitive operations that are inaccessible to consciousness, and that are isolated from personal-level beliefs and desires. Many philosophers find the notion of sub-personal "knowledge" oxymoronic, and yet the notion has come to play an increasingly important role in contemporary cognitive science.

Contemporary cognitive science and cognitive ethology (the study of nonhuman animal cognition and behavior) have generated a great deal of support for the view that much apparently intelligent animal behavior is to be explained by natively structured cognitive "modules" containing domain-specific information and algorithms. Explanations of this sort have been posited for phenomena as diverse as bird navigation systems and bee foraging strategies (Gallistel et al. 1991). Psychologist Steven Pinker suggests that this conception of "instinct" provides the best way to understand the species-specific human capacity for language (Pinker 1994).

Nativism as it is understood in philosophy and contemporary cognitive science is a theory of human universals, of species-wide characteristics. It should therefore be sharply distinguished from "biological determinist" theories—theories that purport to explain individual human differences with respect to such traits as intelligence and criminality in terms of innate and supposedly immutable genotypic differences.

History of the Debate

The basic outlines of the debate were set in ancient Greece, and were further elaborated in the early modern period. Plato employed a POS argument for a version of nativism, his doctrine of "recollection" (anamnesis ). In Meno, Plato's character Socrates elicits from an uneducated slave boy the solutions to a series of problems in geometry, culminating in the identification of an irrational number, the square root of two. Since the boy could not have learned it, Socrates argues, the relevant information must have already been present within his soul, needing only the stimulus of Socrates's questioning for it to be "recollected." Socrates goes on to argue that the origins of this knowledge, which every human being could "find … within himself" (Plato 1981, p. 75), are prenatal, the result of the soul's early encounter in a previous life with the objects of eternal truths.

Aristotle rejected Plato's doctrine of innate ideas, averring that the mind is initially blank. Universals, or "intelligible forms" can only be grasped through experience, by abstraction from the "sensible forms" delivered to the mind through sense perception (Posterior Analytics, Bk. II). This Aristotelian empiricism was embraced and elaborated by the Scholastics in the Middle Ages. According to Aquinas, apparently a priori knowledge such as that of Plato's slave boy could be accounted for in terms of the mind's detection of its own operations. Although the mind natively has the propensity to perform these operations, it is not until the mind is stimulated by the provision of the sensible forms imparted through perception that such operations actually occur. In this sense, then, even the apprehension of one's own inner mental life is indirectly dependent upon sensory experience (Adams 1975).

René Descartes revived nativism. He rejected the Aristotelian/Scholastic doctrine that our ideas of perceptual qualities resemble or share a form with the qualities of the bodies that occasioned them. On Descartes's view, perception, like all causal processes, involved the movement of tiny bodies or corpuscles, the properties of which bore no resemblance to the ideas of sensory qualities in which the process culminated. In that case, Descartes argued, our ideas of color, shape, movement could not have been imparted by sensation, but must rather be innate, and merely occasioned by sensation. In calling such ideas innate, Descartes was not claiming that they were manifestly present in the mind since birth. Rather, he meant that our minds are innately constituted in such a way that sensory experiences of certain sorts reliably give rise to ideas of certain sorts (Descartes 1648/1969).

Although Descartes did not believe that ideas must share forms with the objects they represented, he did accept a principle that constrained the relation between the two. This principle, sometimes called the "principle of proportionate reality," formed the basis of a type of POS argument, albeit an a priori one. The principle states that a cause must contain at least as much reality as its effect. Applied to the special case of the relation between ideas and their causes, Descartes claimed that the principle entails that the cause of an idea must possess at least as much "formal" reality as the "objective" (or representative) reality contained in the idea.

There are three possibilities, Descartes says, as to the causes of those of our ideas that are involved in judgments, that is, that admit of truth or falsity: they can be (a) caused by something outside of ourselves ("adventitious" ideas), (b) fabricated by ourselves ("factitious"), or (c) innate. Now our idea of God is an idea of an infinite, eternal, omniscient, omnipotent and perfectly benevolent being. Since the amount of objective reality in such an idea is infinite, the cause of this idea itself must be something infinite and perfect. No finite substance could be the cause, and hence it could not have been caused by a substance exeeeee e external to myself, nor could I have caused it myself. Only an infinite, perfect being could have caused it, i.e., only God Himself. Furthermore, since this idea does not come upon me "unexpectedly" as do ideas that come through the senses, and since it is not within my power to change it as is the case with ideas that I originate, my idea of God, Descartes concludes, must be innate, implanted by God at the beginning of my existence (Descartes 1641/1969).

Descartes also made two empirical arguments for the domain-specificity of the human language capacity, arguments explicitly cited by Noam Chomsky in his twentieth-century defense of the same conclusion. In Discourse on Method (1637/1969) Descartes sought to establish the existence of a "special faculty" in human beings that accounted for our capacity, distinctive within the animal kingdom, for creative language use. He considers the objection, offered by a hypothetical interlocuter, that human linguistic competence might be due to a merely quantitative difference in some general ability between human beings and other animals, rather than, as Descartes's view had it, a difference in kind.

Descartes's first response is to point out that even the "stupidest" human children acquire language without difficulty, whereas not even the most intelligent infrahuman animals are able to acquire it at all. (It is noteworthy that Descartes's judgment about the linguistic incapacity of infrahuman animals was confirmed during the 1970s, when a variety of researchers attempted, without success, to teach American Sign Language to apes and chimpanzees. Despite assiduous training, not one of the otherwise highly intelligent and creative animals even approached the linguistic achievements of the average deaf human three-year-old (Pinker 1995). This response, however, is vulnerable to the following rejoinder: It may be that there simply is no overlap between humans and other animals in the ranges of variation in the relevant ability, that the "stupidest" human has a far greater amount of this hypothetical general ability than does the "most perfect parrot or monkey." Descartes's second argument speaks to that possibility: if human linguistic ability is a manifestation of a more general ability, then we'd expect that human beings would excel above animals in all activities to the same extent we exceed them in the practice of communication. But this is not what is observed (Descartes 1637/1969).

John Locke, like Aristotle, held that the mind is initially only a "white paper void of all characters, without any ideas" (1689/1979, p. 2). He explicitly rejected Descartes's doctrine of innate ideas. It was unnecessary, he argued, to posit "innate principles" in order to account for any feature of human knowledge, when the humanly "natural faculties" of sensation and reflection could be shown to be quite sufficient. He particularly objected to the view that ideas of sense were innate, arguing that "it would be impertinent to suppose the ideas of colors innate in a creature to whom God hath given sight, and a power to receive them by the eyes from external objects" (1689/1979, p. 1). But he was equally adamant that truths of reason became known only through experience. The faculty of reflection was sufficient to ensure that we recognize such propositions as true as soon as we apprehend them.

Locke's Essay was meant to provide a systematic explanation of the origins of all human knowledge from these raw materials of human faculties and simple ideas of sense, and thus was addressed to Descartes's POS arguments. But Locke also had principled reasons for rejecting the doctrine of innate ideas, reasons that echo today in contemporary empiricists' objections to empirical nativist theories. To begin with, he argued, these allegedly innate principles are not in fact universally known—"children and idiots have not the least apprehension … of them" (p. 4)—and so could not have been imprinted upon the human soul at birth. If it is replied that the principles could have been imprinted without the subjects' being aware of them, Locke argues, then the doctrine of innate ideas is trivialized. If we can make sense of the notion of an unperceived idea, Locke argues, it can only be as an idea that we have the capability of acquiring. If the doctrine is understood in this way, however, it becomes trivial, because any truth a person can come to know in the course of a lifetime would count as "innate" in this sense. And in any case, even if "universal consent" were established, nativism would not provide the only explanation.

Locke's positive account of the acquisition of ideas and the formation of knowledge relied on posited mechanisms of association and abstraction for the accumulation and manipulation of sensory impressions into general ideas, abstract ideas, and judgments. Subsequent empiricist models of concept acquisition and learning have followed Locke's model in essentials.

Gottfried Leibniz, in his New Essays, attempted a systematic rebuttal of Locke's critique of the doctrine of innate ideas. Leibniz first addressed Locke's in-principle objection to the notion of an innate, but unperceived idea. Leibniz argues that Locke is operating with too restrictive a notion of "knowledge" if he does not acknowledge the existence of implicit or unconscious knowledge. And to Locke's objection that allowing implicit knowledge would trivialize the doctrine of innate ideas, Leibniz responds that a distinction can and must be made between sciences like arithmetic and geometry, that we can "construct for ourselves … in our private room … without learning through sight or even touch the truths which we need," and those which require sensory experience (Leibniz 1704/1975). Furthermore, Leibniz argues, there is a difference between the mind's actually possessing structure, of a sort that permits the generation, a priori, of knowledge, and the mind's simply having the potential for acquiring truths. This disagreement between Locke and Leibniz about the nature and significance of innate mental structure has echoes in contemporary philosophical debates.

Leibniz then turned his attention to Locke's positive account of the development of knowledge, and here makes an argument of the type this entry has termed "transcendental." Leibniz focuses on two concepts repeatedly cited by Locke as examples of ideas that could not possibly have been innate, but must be acquired through experience, namely impossibility and identity. According to Locke, these ideas can only be the product of the comparison of various specific sensory experiences, such as the taste of a normal nipple, versus one rubbed with wormwood. But this, according to Leibniz makes no sense: abstraction cannot explain the acquisition of these concepts, because abstraction presupposes the ability in the subject to classify experiences as similar or different, and thus as possessing the concepts same and different. Possession of such concepts is a precondition of, and thus cannot be the result of empirical learning (Leibniz 1704/1975). Ironically, a similar argument was made by the empiricist David Hume for the innateness of the principle of induction. The notion that the future tends to resemble the past is not one, Hume argued, that could be acquired through experience, because we need to presume that principle in order to take experience as evidence for anything at all (Hume 1748/1977).

Immanuel Kant, from whom my use of the term "transcendental" is borrowed, argued similarly that empirical experience as we know it—"intuitions"—are only possible because of innate forms and structures that characterize our perceptual and intellectual capacities. Space and time, he argued, are not features of reality considered "in itself" but are rather the "a priori" forms of perception. Perceptual experiences are only cognitively available to us because of the "pure categories of the understanding," highly general concepts like object and cause, that allow us to utilize experience to form judgments (Kant 1781/1787/1929).

Current Controversies

At the beginning of the twentieth century, under the influence of logical positivism, many philosophers rejected not only the doctrine of innate ideas, but the very notion of the mind. Hypotheses about mental structures and processes were held to be at best unverifiable, and at worst, unintelligible; the only possible "science of the mind" would be, ironically, a science of behavior. J. B. Watson and B. F. Skinner developed what they claimed was a fully general account of learning, applicable to all behavior, whether "intelligent" or reflexive, human or infrahuman. The basic mechanisms posited by behaviorists—classical and operant conditioning—involved the evocation, shaping and reinforcement of patterns of behavior. Although behaviorists eschewed any reference to the mental, they did tacitly accept Leibniz's point against Locke. Accordingly, they posited what philosopher W. v. O. Quine called an "innate similarity space"—a disposition to treat stimuli as falling into similarity classes, manifested by "stimulus generalization," the transfer of a reinforced response to similar but novel circumstances, a precondition of the learning of complex behavior (Skinner 1953).

Skinner thought that behaviorist principles could and would account for even the most complex behaviors acquired by human beings in their lifetimes, including the mastery of language. His book Verbal Behavior was the first, and to date, the last, effort by an anti-nativist, anti-mentalist to provide a systematic and relatively detailed explanation of the acquisition of human language (Skinner 1957). But in 1959, Noam Chomsky published a devastating review of the book in which he showed that the theory faced a fatal flaw: It was either grossly empirically inadequate or else devoid of empirical content (Chomsky 1959).

Chomsky had outlined his own positive account of language acquisition in 1957, an account that was iconoclastic at the time in reviving not only mentalism, but nativism. Unlike Skinner, who derived his model of language learning top-down, Chomsky urged a naturalistic approach to the study of language, one that focused on the actual conditions in which children acquired their linguistic competence. This focus quickly revealed several important points: (1) children attain language without any explicit instruction; (2) language is almost universally acquired, even for blind children (and deaf children if they are given access to signed language) within the first five years of life; and (3) the body of evidence available to children during acquisition—the body of "primary linguistic data"—is badly "impoverished" relative to the body of information eventually mastered. In particular, the data do not contain "negative evidence"—information that certain constructions are not licensed. This is highly significant, as there seem to be certain kinds of ungrammatical constructions that never appear among children's early mistakes. Given all this, Chomsky concluded that language acquisition was not the result of some general learning mechanism operating on sensory data, but that it rather involved, a domain-specific cognitive mechanism that embodied constraints on the forms human linguistic systems can take—the "language acquisition device" (LAD). This natively specified set of constraints, "universal grammar" (UG), greatly simplifies the acquisition task by sharply constraining the set of candidate grammars a child must consider in response to data and by filling in gaps left by experience.

The POS paradigm has since been applied in many other ways within cognitive science. Psychologists Barbara Landau and Lila Gleitman have shown that blind children acquire verbs of sight (for example, "look" "see") in much the same way as sighted children, though, obviously, in the absence of any visual experience (Landau and Gleitman 1985). Gleitman has argued that children's mastery of semantics (as opposed to syntax, which is Chomsky's focus) cannot be explained on the basis of experience, showing that the most common verbs in human language have no distinctive profiles of contingency of usage from which the child could infer their meanings (Gleitman 1990). These findings directly contradict the empiricist accounts of meaning acquisition found in Skinner and Quine (Skinner 1957, Quine 1973).

Elizabeth Spelke, pioneering an experimental paradigm for studying infant cognition, has provided empirical support for the views of Leibniz and Kant that our basic conceptual organization of the external world is natively specified, as a precondition of all other empirical learning (Spelke 1995). Evidence from the study of autistic persons has led many researchers to posit an innate "theory of mind" that enables non-autistic persons to interpret the facial expressions, gestures, intonation patterns of their fellow human beings in a reflexive way, and to effortlessly generate appropriate hypotheses about their likely intentions, desires, and reactions (Frith 1992, Baron-Cohen 1995). Empiricist critics of this so-called "theory-theory" approach to our capacity for psychological understanding argue that we do not need innate psychological knowledge; that rather, we utilize our capacity to mimic our conspecifics to run "simulations" of other individuals' psychologies, and thus to learn empirically what is going on in their heads (Gordon 1986, Heal 1994, Goldman 2006).

Philosophical critics of Chomsky have challenged the intelligibility of his model of mind and his conception of innate knowledge, raising many of the points raised by Locke and other empiricists. Gilbert Harman and Hilary Putnam each have argued that Chomsky's arguments demonstrate, at best, the existence of native principles of induction governing the learning of language, a thesis any empiricist could accept. Putnam also argued that nativism is not the only or the best explanation for the existence of linguistic universals, or the ubiquity of acquisition (Putnam1967, Harman 1967). Chomsky replies that the real issue is whether there are specialized mechanisms that in some sense embody domain-specific information, not how that information is embodied (See also Katz 1966). He insists that the notion of "innate knowledge" should be understood as an abductive posit in a scientific theory, with the details to be worked out as part of the relevant scientific investigation, as is usual in other branches of science (Chomsky 1969). Pinker and others have stressed, in response to Putnam's second point, that it is the poverty of the stimulus, rather than universality, that provides the strongest support for nativism. Alvin Goldman considers and responds to the objection, first found in Locke, that mere innateness cannot secure the justification required for knowledge, arguing that an externalist epistemological framework that takes account of selectional processes that account for native beliefs or biases can provide the systematic warrant needed for knowledge (Goldman 2006).

On the empirical side, critics have focused, appropriately, on the POS argument. Some have attacked the logic of the POS argument, contending that many cognitive achievements are made in the absence of apparently needed evidence, with there being no reason to think the relevant knowledge is innate. Fiona Cowie, for example, argues that she acquired the concept "curry" without explicit instruction, without negative evidence, but that that is no reason to think the concept is innate (Cowie 1998). Others contend that the primary linguistic data are richer than Chomsky has supposed. For example, it has been widely documented that parents and other speakers interacting with infants tend to produce a simplified and pedagogically friendly version of human language, dubbed "Motherese," that may provide the child with a salutarily biased sample of the language (Snow 1972). Other critics argue that our cognitive resources for extracting information from the environment are more powerful than Chomsky supposed. These latter critics include advocates of a new empiricist model of mind called "connectionism." Connectionists reject the view that cognition involves the manipulation of structured representations, arguing instead that the mind is a vast network of neural nodes, capable, given suitable "training," of detecting and responding, at levels not accessible to consciousness, to very subtle regularities in the data stream (Elman et al. 1996).

Chomsky's theory is computationalist—that is, it presupposes that the mind is, inter alia, an information processing device. In 1975, philosopher Jerry Fodor argued that not only Chomsky's, but all successful and fruitful psychological theories were tacitly committed to this model, and that the model itself carried a heavy ontological commitment. Computations, Fodor argues, require a medium of computation. Since the acquisition of natural language is one of the processes that is, ex hypothesi, computational, natural language cannot itself be the medium in which the relevant computations take place. There must be an antecedently existing medium, an innate "language of thought" (LOT), with at least as much syntactic complexity and expressive power as needed to fully represent natural languages (Fodor 1975).

Fodor's view on the latter point has developed. Initially, he claimed that the argument showed that all concepts were innate, but he has recently modified this to the claim that no concept is learned, leaving open the possibility that there are non-computational, and hence non-psychological means for acquiring concepts (Fodor 1997). The argument that no concepts are learned is transcendental in character, and simple to state: (1) Concept learning (if there is such a thing) would involve the formulation, projection and confirmation of hypotheses as to the extension of the concept to be learned; (2) Such processes presuppose the means to represent the extension of the concept to be learned; (3) But any system that has the means to represent the extension of a concept ipso facto possesses the concept. Therefore, (4), one cannot learn a concept without already possessing that concept; hence (5) concept learning is impossible.

Fodor's critics include many of the philosophers who have challenged Chomsky. Critics have charged that the LOT leads to an infinite regress of languages and interpreters (Dennett 1975, Harman 1975). Connectionists have argued that Fodor's fundamental assumption—that mental operations involve the manipulation of structured symbols—is mistaken, and that there is therefore no need to posit an innate medium of computation (Clark 1993). Other critics have argued that Fodor has constructed a false dilemma about concept learning: that there are plausible models that do involve a rational extraction of information, and that should thus count as learning, but that do not employ the hypothesis-confirmation model that Fodor presumes to be the only alternative to non-psychological triggering (Margolis 1998, Cowie 1998).

Fodor is also responsible for revitalizing the theory of mental modularity (Fodor 1983). According to Fodor, processing in each sensory modality, as well as linguistic processing, takes place in specialized functional regions of the brain, equipped with proprietary algorithms, memory, and computational vocabulary. Such processing is characteristically fast, automatic, and "informationally encapsulated"—insensitive to information from outside the module, typically information stored in central systems. The character of these modules is specified in the genome, although their development may require experiential inputs at crucial stages. Contemporary anthropologists and social psychologists, along with some cognitive psychologists, have become increasingly interested in developing modular explanations for a wide variety of human cognitive and psychological traits, from reasoning about social contracts to men's alleged aesthetic preference for firm breasts in women (Barkow et al. 1992; see Pinker 1997, 2002 for an overview). Ironically, Fodor has emerged as the most vocal critic of this line of thought, objecting both to the strict adaptationist presumptions of the modularists' methodology, and to what he regards as the emptiness of the conception of modularity they employ (Fodor 2000). The debate beginning with Plato will no doubt continue as philosophers, psychologists, and cognitive scientists try to bring fresh insight to the issue of innate ideas.

See also Aristotle; Artificial and Natural Languages; Chomsky, Noam; Connectionism; Dennett, Daniel C.; Descartes, René; Fodor, Jerry A.; Harman, Gilbert; Hume, David; Innate Ideas; Kant, Immanuel; Leibniz, Gottfried Wilhelm; Locke, John; Plato; Putnam, Hilary; Quine, Willard Van Orman; Skinner, B. F.; Thomas Aquinas, St.

Bibliography

Adams, Robert Merrihew. "Where Do Our Ideas Come From?" In Innate Ideas, edited by Stephen P. Stitch. Los Angeles: University of California Press, 1975.

Aristotle. Posterior Analytics (350 BCE). In The Basic Works of Aristotle, edited by Richard McKeon. Oxford: Oxford University Press, 1941.

Barkow, J. H., L. Cosmides, and J. Tooby. The Adapted Mind: Evolutionary Psychology and the Generation of Culture. New York: Oxford University Press, 1992.

Baron-Cohen, Simon. Mindblindness: An Essay on Autism and Theory of Mind. Cambridge, MA: MIT Press, 1995.

Chomsky, Noam. Aspects of the Theory of Syntax. Cambridge, MA: MIT Press, 1965.

Chomsky, Noam. Reflections on Language. New York: Pantheon Books, 1975.

Chomsky, Noam. "A Review of B. F. Skinner's Verbal Behavior." Language 35 (1) (1959): 26–58.

Chomsky, Noam. Syntactic Structures. The Hague: Mouton, 1957.

Clark, Andy. Connectionism, Concepts, and Representational Change. Cambridge, MA: MIT Press, 1993.

Cowie, Fiona. What's Within? Nativism Reconsidered. Oxford: Oxford University Press, 1998.

Descartes, René. Discourse on the Method of Rightly Conducting the Reason (1637). Translated by E. S. Haldane and G. T. R. Ross. In The Essential Descartes, edited by Margaret Wilson. New York: New American Library, 1969.

Descartes, René. Meditations on First Philosophy (1641). In The Essential Descartes, edited by Margaret Wilson. New York: New American Library, 1969.

Descartes, René. Notes Directed Against a Certain Programme (1648). In The Essential Descartes, edited by Margaret Wilson. New York: New American Library, 1969.

Dennett, Daniel. "Brain Writing and Mind Reading." Minnesota Studies in the Philosophy of Science 7 (1975): 403–415.

Elman, J. L., E. A. Bates, M. H. Johnson, and A. Karmiloff-Smith et al. Rethinking Innateness: A Connectionist Perspective on Development. Cambridge, MA: Bradford Books/MIT Press, 1996.

Fodor, Jerry. Concepts: Where Cognitive Science Went Wrong. Oxford: Oxford University Press, 1997.

Fodor, Jerry. The Language of Thought. New York: Crowell, 1975.

Fodor, Jerry. The Mind Doesn't Work That Way: The Scope and Limits of Computational Psychology. Cambridge, MA: MIT Press, 2000.

Fodor, Jerry. The Modularity of Mind. Cambridge, MA: MIT Press, 1983.

Frith, Uta. Autism: Explaining the Enigma. Oxford: Blackwell, 1992.

Gallistel, C. R., A. L. Brown, S. Carey, R. Gelman, and F. C. Keil. "Lessons from Animal Learning for the Study of Cognitive Development." In The Epigenesis of Mind: Essays on Biology and Cognition, 3–36. Hillsdale, NJ: Lawrence Erlbaum Associates, 1991.

Gleitman, Lila R. "The Structural Sources of Verb Meanings." Language Acquisition 1 (1) (1990): 3–55.

Goldman, A. Simulating Minds: The Philosophy, Psychology, and Neuroscience of Mindreading. New York: Oxford University Press, 2006.

Gordon, Robert. "Folk Psychology as Simulation." Mind and Language 7 (1986): 11–34.

Harman, Gilbert. "Language, Thought, and Communication." Minnesota Studies in the Philosophy of Science 7 (1975): 270–298.

Harman, Gilbert. "Psychological Aspects of the Theory of Syntax." Journal of Philosophy LXIV (1967): 75–87. In Innate Ideas, edited by Stephen P. Stitch. Los Angeles: University of California Press, 1975.

Heal, Jane. "Simulation vs. Theory-theory: What Is at Issue?" In Objectivity, Simulation, and the Unity of Consciousness, edited by C. Peacocke. Oxford: Oxford University Press, 1994.

Hume, David. An Enquiry concerning Human Understanding (1748). Indianapolis, IN: Hacket Publishing Co., 1977.

Katz, Jerrold J. The Philosophy of Language. New York and London: Harper and Row, 1966. Excerpted in Innate Ideas, edited by Stephen P. Stitch. Los Angeles: University of California Press, 1975.

Kant, Immanuel. The Critique of Pure Reason (1781a/1787b). Translated by Norman Kemp Smith. New York: Macmillan, 1929.

Landau, Barbara, and Lila R. Gleitman. Language and Experience: Evidence from the Blind Child. Cambridge, MA: Harvard University Press, 1985.

Leibniz, G. W. F. New Essays on Human Understanding (1704). Excerpted in Innate Ideas, edited by Stephen P. Stitch. Los Angeles: University of California Press, 1975.

Locke, John. An Essay concerning Human Understanding (1689). New York: Oxford University Press, 1979.

Margolis, Eric. "How to Acquire a Concept." Mind and Language 13 (1998): 347–369.

Pinker, Steven. The Blank Slate. New York: Viking/The Penguin Group, 2002.

Pinker, Steven. How the Mind Works. New York: Norton, 1997.

Pinker, Steven. The Language Instinct. New York: Harper Collins, 1994.

Plato. Meno. Translated by G. M. A. Grube. Indianapolis: Hackett Publishing Co., 1981.

Putnam, Hilary. "The 'Innateness Hypothesis' and Explanatory Models in Linguistics." Synthese 17 (1967): 12–22. In Innate Ideas, edited by Stephen P. Stitch. Los Angeles: University of California Press, 1975.

Quine, W. v. O. The Roots of Reference. La Salle, IL: Open Court, 1973.

Skinner, B. F. Science and Human Behavior. New York: The Free Press, 1953.

Skinner, B. F. Verbal Behavior. Acton, MA: Copley, 1957.

Snow, C. E. "Mothers' Speech to Children Learning Language." Child Development 43 (1972): 459–465.

Spelke, Elizabeth. "Initial Knowledge: Six Suggestions." Cognition 50 (1995): 433–447.

Stich, Stephen P., ed. Innate Ideas. Los Angeles: University of California Press, 1975.

Wilson, Margaret, ed. The Essential Descartes. New York: New American Library, 1969.

Louise M. Antony (2005)