A vague notion of what would later be called "archetype" began to structure Jung's thinking as early as 1910, when he was combing anthropology and mythology in search of an organizing "nuclear complex" within the psyche. He was working on the hunch that all of us humans have a fundamentally similar manner of organizing and interpreting the world we inhabit and the situations we fall into. It seemed that mythology would hold the key -- for it was ancient, held sacred by entire societies and cultures, and strove to explore the fundamental questions: who we are, where we come from, what our purpose is, the "demons" we fall in love with, the absurdity of death, and the kinds of nonsense we concoct to distract ourselves. [1]

Jung on the Nature of the Archetypes

          Although one reviewer found more than thirty different definitions of archetype in a single volume (CW9i) of Jung's Collected Works (R. F. Hobson, 1980), there is at least one stable fact. The archetype is always some sort of structuring principle that lies outside of everyday consciousness and, when it emerges suddenly, exceeds all subjective expectations. Running into such an archaic reality, Jung claimed as early 1911, was as surprising as encountering a 2000 year-old Corinthian column standing on a modern street corner -- the last thing we would expect, and yet disturbingly familiar.

Just a moment ago we were given over to the noisy ephemeral life of the present, when something very far away and strange appears to us, which turns our attention to things of another order . . . that on this very spot where we now run busily to and fro a similar life and activity prevailed two thousand years ago in somewhat other forms; similar passions moved mankind, and a man was likewise convinced of the uniqueness of his experience (CWB: 1).

Sometimes we seem to find "archetypes" outside ourselves as fascinating objects and people, but even so, something inside us enables our seeing them as fascinating. An archetype determines our perceptions and our behavior, and often without our awareness. Encountering puberty is a perfect example of archetypal experience. Every one of us "remembers" the same package of experiences our ancestors have been undergoing since time immemorial. When it has not yet happened, we are incapable of imagining it. But once it hits us, we can barely remember how things used to be.

          Falling deeply in love may be the most commonly accessible of such archetypal experiences forcibly to alter the attention of adults. We find that our awareness is radically transformed, the world imbued with a new light, our beloved elevated to quasi-divine status, and the two of us often impressed and embarrassed by such paranormal experiences as telepathy. [2] Every time eros invades our life, the experience seems unique and unrepeatable -- the first time it has occurred in the history of the world. And yet, it is as old as time. It is a typical human experience of great power and significance that can change our destiny over night.

Lamarckian flirtations. Jung's first efforts to formulate a theory of archetypes borrowed heavily from the history of philosophy as well as the sociology and anthropology of the late nineteenth century. Indeed, he is often at pains to point out that he did not invent the idea of archetypes, that it has a history, including Plato, Augustine, Malebranche, Bacon, Herbert of Cherburg, Descartes, Spinoza, Kant, Lévy-Bruhl, Hubert & Mauss, and Adolf Bastian. [3] In some places, he specifies that it is precisely the imagination that is structured by the archetypes of the unconscious -- just as space, time, and the categories structure the Kantian consciousness. [4]

          At a time when Darwinian conceptions were still rather fluid and in dispute, Jung borrowed language and imagery that today leaves him vulnerable to the suspicion that he believed powerful images and ideas could be inherited by one's children -- as though an image of great power could somehow cause a precise mutation in one's sperm or ova. Today we are much more conscious of how absurd that notion is than were social scientists between 1880 and 1920. Gregor Mendel's genetics [5] were not rediscovered until 1900 and their application to evolutionary theory had yet to be worked out, and could not be really understood until the structure of DNA was discovered in 1953.

          Thus, in 1920, Jung was using several problematic terms to characterize what later came to be called the archetype:

From the scientific, causal perspective the primordial image can be conceived as a mnemonic deposit, an imprint or engram (Semon), which has arisen through the condensation of countless processes of a similar kind. . . . (CW6: 748)

Seven years later, he said the unconscious itself was the "totality of all archetypes," which he described as "the deposit of all human experience right back to its remotest beginnings" (CW8: 339). In 1928, he said our ancestors had "traced these paths," and that every time an experience of that impressive type "breaks through" it opens up "an ancient riverbed" (CW8: 100). "Endless repetition has engraved these experiences into our psychic constitution" (1936, CW9i: 99).

          These passages leave no doubt that Jung is sincerely trying to align his incipient concept of archetype with the process of evolution. But the images used imply a Lamarckian rather than a Darwinian process. Talk of images (primordial or otherwise), mnemonic deposit, the accumulation of generations of experience, and engraving our constitution like a riverbed all seem to imply that what happens to the individual has a slight but determining effect upon its offspring -- rather like that famous stereotype of the giraffe's neck that grew incrementally longer, generation by generation, as individual animals stretched higher and higher for tender leaves. Today, no biologist would entertain an evolutionary theory based on any sort of athletic or mental achievement that could affect the configuration of the DNA in our reproductive cells. The idea of carving out a riverbed or contributing to a species' "memory traces" can hardly be taken seriously as a useful metaphor for the evolutionary process.

          Confusingly enough, however, Jung also sought to disavow the language of these published statements. He insisted that the term "primordial image" (borrowed from Henri Hubert and Marcel Mauss, 1909) and the term "collective representation" (borrowed from Lucien Lévy-Bruhl, 1923) did not have anything to do with inheriting images or ideas, themselves -- but rather "of having the possibility [to generate and entertain] such ideas" (1917/43, CW7: 101, 104). What was inherited was, he said in an inspired fit of ambiguity, "an organ of psychic energy" (1920, CW6: 754). Overlooking the inconsistency in his own language, he insisted to the end of his life that his "critics have incorrectly assumed that by archetype I mean `inherited ideas,' and on this ground have dismissed the concept of the archetype as a mere superstition" (CW18: 524).

Archetype and instinct. Our concern, therefore, is to determine what an archetype would have to be if it were to be more than a "mere superstition," indeed a concept harmonious enough with modern biology to aspire to the status of a genuine "human science." It would have to be inherited with our DNA and give rise to typical brain structures that correlate with the activities ascribed to the archetype. Furthermore, it would have to have identifiable precursors in evolutionary history, as seen in primates, mammals, and even "lower" animals. Fortunately a great many of Jung's assertions about human mental inheritance seem very much in tune with natural selection. As early as 1918, he said:

We receive along with our body a highly differentiated brain which brings with it its entire history, and when it becomes creative it creates out of this history -- out of the history of mankind . . . [i.e,] that age-old natural history which has been transmitted in living form since the remotest times, namely the history of the brain-structure (CW10: 12).

More vividly put thirty years later: "It is more probable that the young weaver-bird builds his characteristic nest because he is a weaver-bird and not a rabbit. Similarly, it is more probable that man is born with a specific human mode of behavior and not with that of a hippopotamus or with none at all" (CW8: 435). "The child brings with it an organism [the brain] ready to function in the same way as it has functioned throughout human history" (Ibid., 589).

(1961) Just as the human body represents a whole museum of organs, with a long evolutionary history behind them, so we should expect the mind to be organized in a similar way . . . I [mean] its biological, prehistoric, and unconscious development beginning with archaic man, whose psyche was still similar to that of an animal (CW18: 522).

We are born with a brain or mind whose structure is inherited and designed to respond in determined ways to certain kinds of situation. It is not, as the behaviorists had been saying for decades, a "blank slate" (tabula rasa) on which experience writes all that we ever know. Rather the "instincts . . . engender peculiar thoughts and emotions" that express "ever recurring patterns of psychic functioning" (1958, CW18: 1271). [6]

          That mention of "instinct" became crucial to Jung's discussion of the archetypes. Already in 1912, he implies as much when he cites "vestiges of obsolete functions" that still occur in our minds today (CWB: 36). It appears that he was referring at that time to the now largely [7] repudiated "biogenetic law" of recapitulation advanced by Ernst Haeckel around 1860, "ontogeny recapitulates phylogeny." The motto means that the embryos of evolutionarily more advanced animals pass through stages of development which "recapitulate" in historical order the anatomy of all the lower phyla. By 1927, however, Jung was able to express the essence of his earlier and biologically inaccurate statement in vivid imagery that is certainly less objectionable: "This whole psychic organism . . . preserves elements that connect it with the invertebrates and ultimately with the protozoa. Theoretically, it should be possible to "peel" the collective unconscious, layer by layer, until we come to the psychology of the worm, and even of the amoeba" (CW8: 322). Clearly what he meant to say is that behind the mental activities that we find most distinctively human and "advanced" lies an evolutionary history. If it had been known at the time, Jung would surely have been delighted to cite the brain-like functions in the E. coli bacterium in support of his intuition:

The mechanisms whereby they [E. coli] sense, remember, and move about their environment provide an excellent model for the basic features of nervous systems, albeit in an organism contained within a single cell and lacking a brain in a conventional sense (Allman, 2000: 3).

          It is still true today that instinct remains a fairly loose notion, "a biological phenomenon of immense complexity, not much more than a borderline concept of quite indefinite standing for an unknown quantity" (1932, CW11: 493). Nevertheless it remains useful to speak of "instincts" or "drives" [8] insofar as these terms always evoke the notion of patterns of behavior that are inherited -- very similar to what Jung ascribes to the archetypes. In 1919, he provided parallel definitions for instinct and archetype: "Instincts are typical modes of action" (8: 273), while "archetypes are typical modes of apprehension" (8: 280); instinct and archetype "determine one another" (8: 271). What he means is that the instinct drives the behavior pattern, while the archetype apprehends the environmental and/or physiological conditions under which the instinctual behavior is an appropriate response. There will be no instinctual behavior pattern unless the archetype "apprehends" these necessary conditions.

          According to his favorite example, every human archetype functions like those primitive and unvarying mechanisms that drive the yucca moth and leaf-cutter ant to perform highly complicated activities to fertilize their eggs and provide for their survival, despite the fact that their adult sexual forms are so short-lived that they never have a chance to observe the behavior before nature requires them to carry it out. Thus archetypes shape innate tendencies that predate all learning. In insects like the yucca moth and leaf-cutter ant, in fact, the innate releasing mechanism is identical with the archetype and functions in place of learning. While an instinct "drives" them to reproduce, a closely related archetype enables them "recognize" the appropriate season and the specific plants whose parts are used for depositing the eggs and feeding the future larvae. [9] Insects carry out such behavior patterns with invariable precision, and with nothing like what we would call "consciousness." Evolution has, therefore, provided increasing complexity as we proceed "upward" from phylum to phylum, allowing for a progressively wider range of freedom in pursuing instinctual patterns.

          Another aspect of the instinct/archetype relation Jung proposed in 1919 is based in this higher complexity, where the archetype "might suitably be described as the instinct's perception of itself or as the self-portrait of the instinct" (CW8: 277). We can understand this in at least two senses. First, the reproductive archetype of the yucca moth apprehends the flowering yucca plants as "affording" it the opportunity and necessity of reproducing. [10] Thus it sees, smells, and feels the blossoming plant as an archetypal image, the trigger that fires the instinctual pattern, i.e., the "self-portrait" of the instinct. Meanwhile, it seems extremely doubtful that the moth can imagine the shape and smell of the yucca flower at a time when it is not in bloom. Here is where the second sense of the archetype as portrait of its instinct comes in. We humans can do such things in the absence of environmental conditions. We dream of Jeannie with the light brown hair, undress Britney in our minds, and are sometimes arrested by the evanescent smell of Larry's skin even though Larry himself is a thousand miles away on a business trip. The imagined archetypal image represents the direction in which the instinct moves us; and because we humans are endowed with a highly flexible and nearly inexhaustible imagination, such instinctual reactions can appear before our mind's eye in symbolic fashion, "as ideas and images, like everything that becomes a content of consciousness" (CW8: 435).

(1956) Though "instincts" or "drives" can be formulated in physiological and biological terms, they cannot be pinned down in that way, for they are also psychic entities which manifest themselves in a world of fantasy peculiarly their own. They are not just physiological or consistently biological phenomena, but are at the same time, even in their content, meaningful fantasy structures with a symbolic character. An instinct does not apprehend its object blindly and at random, but brings to it a certain psychic "viewpoint" or interpretation, for every instinct is linked a priori with a corresponding image of the situation . . . (CW14: 602).

          Thus it is that not only our "behavior patterns," but also our dreams, fantasies, illusions, hallucinations, art work, religion--all are structured by the archetypes. Although no part of our experience is free of such structuring, and so subtly that we rarely notice it, there are also times when we undergo disturbing changes in our awareness due to a new archetype interrupting our everyday waking lives. Such moments bring a specific emotional charge that may enthuse us, disorient us, or put us "under a spell." [11] Depending on the nature of the archetype and the sector of human life it directs us toward, we may find ourselves hyper-alert, dreamy, overwrought, or in some other "altered" state of consciousness.

          Although Jung was not a systematic thinker and juggled a variety of different images and ideas to sketch in broad strokes what he intuited as an "archetype," he nevertheless provided a clearly discernible core of notions that enable our positing the archetype as a module of inheritance that is recognizable in typical patterns and images. It is an aspect of instinct, i.e., the ability of the instinct to recognize appropriate conditions in which to generate its distinctive behavior pattern. Subjectively, it manifests as a powerful emotional charge that invests what we see with overwhelming significance. And, although it manifests in the lower phyla as automatic and inflexible patterns, greater brain complexity gives animals increasingly greater freedom in adapting those patterns to individual circumstances.

          What still remains to describe is the full range of a single archetype: (1) how it is the same in all humans, (2) but varies in terms of the phenomena it generates, (3) what structures within the brain are associated with it, (4) where the evidence lies for its being inherited, and (5) how it appears in primates and lower animals. Jung never undertook such a study. But the kind of research that has been going on in the decades since his death has given us precisely the sort of evidence we need to appreciate the archetypes as consilient with modern biology.

The Language Archetype

          It has been standard practice for Westerners to pretend that we are not really animals, to doubt and even deny that we have evolved along with the rest of nature, to hold that "they" have instincts while we have culture and free will. Surely (we have long claimed) our capacity for language "proves" all this. William James (1890: 737) vigorously denied this cultural assumption of ours more than a century ago: "[N]o other mammal, not even the monkey, shows so large an array" of instincts as "the human species." If an instinct is an unconscious, largely impulsive tendency, universal in a species, that makes us do things even when we do not know why -- language must be one of those that has enlarged the human "array." The language instinct (or archetype) unleashes an amazing cascade of effects quite suddenly around the age of three years. The timing is as precise as the appearance of our teeth, first the milk teeth and then the permanent ones. What amazes us about language and tempts us to deny its instinctual nature is the fine instrument we can make of it. But at bottom, it is an instinct and an archetype like any other.

          We inherit archetypes that give us the capacity to learn some things rather than others, to learn some things more quickly and easily than others, and to learn some things at some times and other things at other times. Language is one of those things -- an "innately guided behavior." For just as bees are innately guided to flowers but have to learn the details of the particular ones that grow in their territory, so we are innately guided to pay attention to vocabulary, grammar, and syntax -- though we have to learn the details of the particular language spoken by our family (Aitchison, 2000: ix). To say that language is an archetype rather than an instinct is to emphasize the "apprehending" and purposive aspect of language. From our earliest babblings to our most eloquent persuasions, we are innately guided to pay attention to nuance. Pull is not the same as bull; and cracked is considerably less than shattered, though both are broken. We are constantly apprehending and discriminating. The archetype is the instinct's "self-portrait" and its guide. Archetype is not something added on to instinct, it is instinct's apprehending function, whereby it guides us to some things rather than others.

The language archetype as guide. Stephen Pinker, who directs the Massachusetts Institute of Technology's Center for Cognitive Neuroscience, has written three books in the last decade (1994, 1997, 2003) that contribute significantly to the project of revealing how consilient Jung's archetypal theory is with the modern biological paradigm -- even though Jung himself is not mentioned anywhere in the 1600 pages that comprise those volumes. In The Language Instinct (1994), Pinker has described in detail how we are innately guided to become linguistic beings. [12] In doing so he reveals the nature and operation of a single archetype, one that is more completely known than most, and one that is distinctively human. Pinker's account allows us to set up a model for what to expect of an archetype. Later on, this will enable us to reach some wider ranging conclusions.

          It seems obvious that every healthy human infant is driven to mimicry and compulsive babbling which gradually sorts itself out into recognizable words -- and not without plenty of helpful interchanges with parents and siblings. By the time we are ready to declare that baby is "learning to talk," a great deal has already taken place. A language archetype has been guiding baby's attention in the direction of our loquacity. And, though the learning of English rather than Chinese is a cultural artifact, the learning of language itself is "a distinct piece of the biological makeup of our brains . . . [much as] spiders spin webs because they have spider brains, which give them the urge to spin and the competence to succeed" (Pinker, 1994: 18).

          Usually around the first birthday we notice the babbling has begun to resolve itself into recognizable words. Another six months passes before two or three words can be strung together into phrases; and a real capacity to form sentences does not usually develop much before a child's third birthday, when it begins rapidly to acquire a huge vocabulary, and seemingly without effort, the grammar rules that make sentence construction possible. The whole maturational time-table requires "minimal levels of brain size, long-distance connections [between the neurons], and extra synapses, particularly in the language centers of the brain" (Ibid., 289).

          Although it takes a while for a child to eliminate a few characteristic childish mistakes ("me" instead of "I" or "are" instead of "am"), the basic grammar rules of a language are picked up quickly and effortlessly by every child in every language. In contrast, no other primate has ever learned grammar. Certainly some monkeys have specialized calls -- very much like words -- to indicate danger from above (an eagle), below (a leopard), or deep in the grass (a snake). And when a single monkey vocalizes one of them, the entire troop will take the appropriate evasive behavior (Dunbar, 1996: 48). But such word-like sounds are never strung together. Monkeys have no grammar. In fact, no chimpanzee or dolphin that has been taught a kind of sign language ever grasps the most elementary notion of grammar or syntax (the arrangement of words into sentences). [13] Indeed, Pinker points out that chimps "just don't get it": (1) never seem to feel in their bones what language is, (2) do not take turns, (3) seldom sign spontaneously, (4) only imitate directly what their trainer has just signed, (5) fail to understand that a sign might refer to an object (an apple) and not the giving or the eating of the apple, and (6) turn every sign into a demand for something like food or tickling (Pinker, 1994: 340).

          That an intensively trained chimpanzee is no match for a three year-old child reveals the complexity, detail, and fluidity that the language archetype accomplishes in every one of us without any extra effort on our part. The archetype guides our intuitive grasp of grammar and easy accumulation of thousands of vocabulary words. Furthermore, the archetype is indifferent as to which specific language it is exposed. It does as well with Navaho, Irish, and Greek as with Arabic and English. Pinker shows us, however, that the archetype is even more amazing than that; for it does not simply imitate what we hear spoken around us, it enables each of us to recreate language anew: "Complex language is universal because children actually reinvent it, generation after generation -- not because they are taught . . . but because they cannot help it" (Ibid., 32). The evidence for this last claim lies in the difference between a pidgin language and a creole.

Pidgins and creoles. A pidgin is a rudimentary language system that develops when people who have already acquired separate languages learn to communicate simple notions -- as when a colonizing community of English traders establishes a base in a far corner of the world. Tok Pisin ("talk pidgin"), for example, is the English-related pidgin of Papua New Guinea. A tiny vocabulary is stretched to cover a variety of situations, virtually without grammar. Thus pik man ("pig man") is a male pig, pik meri a sow, and pikinini pik ("pickaninny pig") a piglet. Pul bilong kanu ("pull belong canoe") is a canoe paddle and pul bilong pisin is a bird's (pidgeon's) wing, while pul bilong pis is the "pull belonging to a fish," namely its fin (Aitchison, 2000: 11).

          In contrast with the limited, halting, ungrammatical pidgin, a creole is a real language with a large vocabulary and its own complete grammar. Haitian Creole, for instance, is a French-based language spoken by an entire nation. Originally, French colonists and the native speakers generated a pidgin, but as soon as that language was learned by a generation of children as their mother tongue, the language archetype in those children created a brand new language. "A pidgin can be transmuted into a full complex language in one fell swoop: all it takes is for a group of children to be exposed to the pidgin at the age when they acquire their mother tongue" (Pinker, 1994: 33).

          This process of "creolization" is, furthermore, not limited to spoken languages. Pinker reports the same transition from pidgin to creole in the use of sign language by the deaf in Nicaragua. An initial sign language developed spontaneously in the playgrounds and school busses of Nicaragua, beginning in 1979, when the Sandinista government reformed the educational system of the country and tried to teach deaf children lip reading and speech. Like any spoken pidgin, it was used differently by every signer and required elaborate circumlocutions to compensate for what it lacked in grammar. But in the next generation, it became a real language, a creole, that is as fluid, compact, and grammatical as any. Hence, the language archetype applies to communication of all types, is not limited to sound and speech; and although it requires an ambient means of communication on which to operate, it is always creative -- almost as though we are born knowing what interpersonal communication is and needing only a few stimulating hints to begin the reinvention process anew in every child.

Universal Grammar? As a student of Noam Chomsky, Pinker believes evolution has equipped us "with a plan common to all grammars of all languages, a Universal Grammar, that tells us how to distinguish the syntactical patterns out of the speech of our parents" (1994: 22). "Children do not have to learn a long list of [grammar] rules because they were born knowing the super-rules. Huge chunks of grammar . . . [become] available . . . as if the child were merely flipping a switch to one of two possible positions (Ibid., 112). This is a claim very similar to Jung's idea that the archetype itself cannot be known but that it structures everything that we do come to know. It would explain why a Korean patient who dreams of the Buddha and a European patient who dreams of the Christ may be dreaming the same "message," albeit in separate cultural imagery. It implies that there may universal human themes that always appear in distinct cultural garb. But the Universal Grammar hypothesis also seems to embrace the weaker postulates of "primordial image" and "collective representation" that Jung later qualified by saying that such concepts do not imply an image, really, but the possibility for imagining and responding to images.

          Pinker goes even further than the notion of Universal Grammar, however, and speculates that we do not think in English or Chinese, as we believe we do, but rather in "mentalese," the "language of thought" which "must be richer in some ways and simpler in others" than any of the known languages (Ibid., 81). Knowing English, then, would mean "knowing how to translate mentalese into strings of words" (Ibid., 82). Pinker creates an unnecessary problem, here. Do we inherit a "collective language" like "mentalese," or do we inherit the possibility to recreate language? These are two separate hypotheses, and Pinker seems to want to embrace them both, thereby putting himself out on the limb Jung tried unsuccessfully to saw off.

          I suspect that "mentalese" is an instance of what the philosopher A. N. Whitehead called "misplaced concreteness," a compelling but misleading fiction. [14] Since neither Chomsky nor any of his followers has yet determined the contents of "mentalese," it is probably sufficient to think that our brains are built to recognize linguistic structures, that originally there is no structure but rather a propensity to respond to structure, and that these (archetypal) potentials have no shape at all until they become firmly set through learning the vocabulary and grammar of a mother tongue. [15] Once set, what formerly was an open potential has been given a definite form, say, that of English. After this point, the determined shape of English interferes with the language archetype's ability to adapt to the rules of a second language. This would be why we have to laboriously memorize the rules for verb forms in French or Russian although we picked up English without effort. The language archetype (like any archetype) is an "open program" [16] or an "empty form" that has to be "filled" with what culture provides. This is pretty much what Jung said in 1945/54:

The unconscious supplies as it were the archetypal form, which in itself is empty and irrepresentable. Consciousness immediately fills it with related or similar representational material so that it can be perceived. For this reason archetypal ideas are locally, temporally, and individually conditioned (CW13: 476).

Language and the brain. The anatomy of the language archetype is, in principle, fairly simple to establish: "If there is a language instinct, it has to be embodied somewhere in the brain, and those circuits must have been prepared for their role by the genes that built them" (Pinker, 1994: 299). Fortunately, our present knowledge of the brain allows us to specify precisely where most of the "language circuits" are located. In a left-side view of the human brain, the most prominent characteristic is the Sylvian fissure that diagonally divides the frontal lobe from the temporal, running upward and toward the back of the head. Just in front of that fissure in the lower portion of the left frontal lobe is a bump known as Broca's area (identified in 1861 by Paul Broca) that is involved in the motor output of communication and the stringing of sounds together to form words. Broca's area also handles visual information about language, from reading and signing. A little higher up the path of the Sylvian fissure, but behind it on the temporal lobe is Wernicke's area (identified by Karl Wernicke in 1874) which is concerned with analyzing the auditory input to the brain. [17]

          We can imagine removing the cerebral cortex from the surface of the brain and flattening it out into a single sheet of tissue, six ultra-thin layers deep. If we do so, the diagonal strip running part way across that cerebral table cloth and marking the location of the deep fold of the left Sylvian fissure contains the central circuits that govern language. On the front half, with Broca's area, would be the motor strip that controls the jaws, lips, and tongue. Occupying the rest of the front half of the perisylvian region would be the grammar processing portion of the brain. On the back half of the strip, Wernicke's area sits at the crossroads of three lobes of the brain (frontal, temporal, and parietal) where it can integrate information regarding visual shapes, sounds, bodily sensations, spatial relations, and some aspects of meaning (Ibid., 307-13).

          The whole left perisylvian region of the cortex "can be considered the language organ," says Pinker in one place (1994: 307), although elsewhere he says, "No one has yet located a language organ or a grammar gene" (Ibid., 46). This confusing state of affairs results from the fact that it is not quite true that the brain is "hard wired" for anything: injuries to parts of the brain outside the "language organ" can result in the loss of certain types of language mastery (Ibid., 313ff), and injuries to parts of the left perisylvian are sometimes followed by other parts of the cortex taking over the temporarily lost language function (Ibid., 309f). Indeed, there are so many exceptions to what has seemed to be established about the brain, that one theory claims, "[M]ental processes are patterns of neuronal activity that are widely distributed, hologram style, all over the brain (Ibid., 314). Pinker shows his distain for this model by calling it the "meatloaf theory" rather than the "hologram theory"; but it is clearly not a case of one theory excluding the other. The situation seems comparable to theories of light: some experiments justify seeing light as streams of particles, while others make no sense unless light is propagated in waves. [18]

          Nature has chosen the convenient centrality of the left perisylvian region to locate most of the language circuits, but can apparently shift some of them when necessary. An inherited pattern provides the "pre-set preferences" (Aichison, 2000: 46) according to which the brain constructs itself, and these strongly favor the perisylvian. Meanwhile practice at language tasks strengthens and elaborates the connections between neuronal groups, leading to some minor differences from individual to individual; and this "practice factor" can also result in new neuronal pathways being built in instances where the brain has suffered limited damage to some of its circuits. We are left to conclude that the archetype itself cannot be located in the brain in any absolute and once-and-for-all manner. In some sense it looks as though the pattern comes first, and the elaboration of neuronal circuits is the result. If so, the archetype is not quite "in" the brain, it uses the brain. Perhaps we have to look deeper, into the genes, to find the archetype.

Language and genes. According to modern biological theory, the building up of the entire body, with all its tissues and organ systems is controlled by the genes, which are stretches of DNA, i.e., strings of four different "bases" (conventionally abbreviated as A, C, G, and T). [19] Contiguous "letters" spell three-letter "words" (AAA, CTG, etc.), each one constituting the code and the template for the manufacture of a single amino acid. A string of such three-letter words on the DNA strand becomes the template for a chain of amino acids, that is a protein, one of the building blocks of the body. [20] This is the simple part of the theory that no one doubts. How such simple beginnings can lead to what might some day be identified as the genetic foundation of the language archetype, however, still escapes scientific understanding. Popular accounts of genetics sometimes make it seem that there is one gene for each human trait (brown eyes, curly hair) or each human disease (breast cancer, Alzheimer's disease). If such a one-to-one correspondence were the case, as unfortunately it is not, we might hope one day to locate the gene or group of genes that produces the proteins and shapes the cells of the brain's language circuits. We might then call those genes, the genetic foundation of the language archetype. For instance:

The grammar gene would be stretches of DNA that code for proteins or trigger the transcription of proteins, in certain times and places in the brain, that guide, attract, or glue neurons into networks that, in combination with the synaptic tuning that takes place during learning, are necessary to compute the solution to some grammatical problem (like choosing an affix or a word) (Pinker, 1994: 322).

          In order for the brain to "compute the solution" to a grammatical problem, it has to be built with the right sorts of circuits. It would have to be built differently from the brain of a chimpanzee or a woodchuck. Evidently, it would be the "language genes," whatever they might turn out to be, that would make this possible. Unfortunately, what we know about the way brains build themselves is too general to enable us to distinguish between the genetic factors specific to chimpanzees, woodchucks, or humans. We know general principles better than we know specific details, but even so the process is complex:

Neurons destined for particular cortical areas are born in specific areas along the wall of the ventricles, the fluid-filled cavities at the center of the cerebral hemispheres. They then creep outward toward the skull into their final resting place in the cortex along guy wires formed by the glial cells (the support cells that, together with the neurons, constitute the bulk of the brain). The connections between neurons in different regions of the cortex are often laid down when the intended target area releases some chemical, and the axons growing every which way from the source area "sniff out" that chemical and follow the direction in which its concentration increases, like plant roots growing toward sources of moisture and fertilizer (1994: 321).

Clearly the timed generation of new neurons at the right places along the ventricle walls, the building of the "guy wires" and choosing their direction, and the appropriate release of the right chemicals at the right times to enable the formation of specific neuronal connections rather than others -- all these things and many more are somehow under the control of the genes, even though no one has yet determined how. Once the entire process is known in minute detail, scientists may be in a better position to discern what subtle differences in brain construction provide for the distinctively human left perisylvian region.

          However, even after such processes have been established, they may not be clearly traceable to specific genes; for most developments in the body are not under the control of a single gene or even a simple series of genes. To begin with, there are far too few genes in the entire human genome to control every single tissue and every single process on a one-to-one basis. Most anatomical results come from highly complex cumulative effects in what is called an "emergent process," where large numbers of simple processes, each following its "local rules," results in complex structures without relying on any "higher-level instructions" (Johnson, 2001: 19). For example, the beehive. A large number of bees working simultaneously, each following only a few simple instinctual rules, enables the "emergence" of the architectural marvel of a beehive. Even the hexagonal shape of a hive's cell does not need to be carried in any bee's psyche; for the shape of the bee's head and the fact that hundreds of bees are working in hundreds of contiguous cells simultaneously result in the geometric simplicity of the hexagon. Thousands of bees following simple rules build a hive from the bottom up. No one of them has a blueprint in its mind -- not even the queen -- and yet the results are always remarkably the same. In contrast, when we build a house, we work from the top down, starting with an architect's blueprint. We work from an overall plan down to the specific details, while the bees work, each with its own specific details, and a marvelous beehive emerges. In similar manner, a complex brain with a dependable organization results in every human being, simply on the basis of thousands of genes operating at the simplicity of amino acid assembly.

          Not only beehives but human cities too, can be built from the bottom up -- and were always built that way before the advent of urban planning. When urban planning is kept to a minimum, people locate their homes and businesses in what is the most convenient and efficient place for each of them, and a complex and stable urban structure emerges. Brains, beehives, and cities have this bottom-up structuring in common; and so does genetics, where the process of emergence was named "epigenetics" in 1947 by C. H. Waddington. Molecular processes and cells interact with one another like bees in an emerging hive. Genes turn one another "on" and "off" under certain conditions. Proteins "teach" one another how to fold themselves into complex shapes, each shape being the basis for different sorts of bodily structure. The emergent process of epigenesis leaves the hope of finding the roots of the language archetype in a specific stretch of the DNA molecule rather doubtful. Yet language itself is hardly doubtful. It always appears in healthy humans, appears in an orderly timed sequence, and in every way is as dependable as the human body with its fingers and toes and its Sylvian fissure.

          Language, in short, is an emergent product of millions of simple biological processes that, at bottom, are controlled by the genes. If language is an innately guided behavior, the archetype that guides its acquisition and use does not likely lurk somewhere in a simple stretch of DNA, like a tiny oak tree inside an acorn. Rather the guiding pattern itself also emerges. The picture is more tricky and mysterious than we had expected. When we look for the archetype in the brain, it seems that the archetype is prior to the brain, the innate guidance system that shapes the cerebral neurons and their interconnections to its purposes as surely as it shapes our behavior. The brain is as much an expression of the language archetype as are the linguistic behaviors it guides. We are led to suspect the archetype is prior also to the brain, and that it exists somewhere outside the brain and simply uses it to bring about its effects.

          When we follow up this suspicion and look to the biological processes that precede the brain's assembly, we are led to the genes. But no single gene and no simple series of genes seems likely to determine the archetypal pattern. It seems evident, rather, that the archetype itself emerges along with the tissues and organ systems of the body. It exists in the genome -- not as a segment of direct instructions -- but unfailingly as an emergent process. Apparently it is not located in any one place but everywhere at the same time -- like a hologram, perhaps.

Language and Grooming

          The language archetype has biological roots in the sense that it is an inherited pattern of behavior, or better yet an inherited guiding principle by which, at the age of about three years, we can acquire any human language on earth, rapidly, accurately, and without special effort. When we try to follow those biological roots into the living matter of our body, we find a universal tool in the brain circuits around the left Sylvian fissure. The building of this "language organ," as it may loosely be called, is also guided by an inherited pattern, a deeper biological level of the language archetype where specialized cells are designed, built, and assembled into systems. We can go all the way down to the molecular level, however, to the point where proteins are assembled by the genes, without finding a blueprint for the brain or its language organ. It appears that no collection of proteins or chromosomes contains or constitutes a blueprint of any sort. Rather it seems probable that the archetype "emerges" like a beehive.

          While language is surely rooted in the biology of the body, we also need to know how rooted it is in our history: how it evolved, and what our ancestors did before they could talk. The answers to these questions, if they are to be satisfying, ought to involve evidence that resembles evolutionary change in anatomical structure. For example, the stiff flight-supporting feathers in a bird's wings and tail evidently started out in several species of flightless dinosaurs where they resembled hair or down and were used primarily for thermal insulation. Only later did some become specialized by developing rigid shafts and interlocking barbs and thereby retain their shape well enough to support flight (Prum and Brush, 2003). Flight has tremendous survival value. The ability to escape flightless predators and gain access to new sources of food opened up a new environmental "niche." Such studies show us the pathway through which evolution seems to have "selected" bird's wings. If language is an archetype like any other, we ought to be able to find its precursors and see how it, too, was favored by natural selection.

The evolutionary strategy of primates. Probably the most widely accepted evolutionary theory of the origin of language is Robin Dunbar's Grooming, Gossip, and the Evolution of Language (1996). His theory is not only simple and plausible, but it is backed up with some very interesting statistics.

          The order of mammals to which we humans belong is that of the primates, which includes lemurs, monkeys, and apes, as well as humans. Primates are characterized by a collar bone, a free-moving shoulder joint, an elbow that allows forearm rotation, five digits on hands and feet, usually an opposable thumb and toe, binocular and color vision, enlarged brain, exactly two mammary glands, and young that are usually born singly. Most of these features are adaptations to arboreal living (Thain & Hickman, 2000). Primates are among the oldest lineages of mammals, our earliest ancestors having been small, squirrel-like inhabitants of the tropical forests during the closing millennia of the Age of the Dinosaurs (Dunbar, 1996: 12). Never as powerful as the great predators, primates have had to live in groups to survive:

Indeed, sociality is at the very core of primate existence; it is their primary evolutionary strategy, the thing that marks them out as different from all other species. It is a very special kind of sociality, for it is based on intense bonds between group members, with kinship providing a platform for these relationships (Ibid., 18; emphasis added).

          Sociality, however, is a problematic refuge. On the one hand, the larger the group the better chance each member has of surviving an attack from a predator -- namely that someone else will be the victim. Thus fear of predation encourages an increase in group size until overcrowding and competition for a limited food supply render it unstable. There is always an ideal range of group size, where the forces of danger from within are in balance with those from without. Meanwhile, every individual animal must not only be on guard against predators but must form and maintain coalitions with a minimum number of troop mates. Each animal needs a dependable buffer of reliable relatives and friends to defend itself against relative strangers inside its own troop (Ibid., 19), and the primary means for cultivating friendship cliques is grooming.

Evolutionary crisis. Somewhere around five million years before the present (5m, B.P.), an ecological crisis forced an evolutionary change upon the apes of the African forests. An Ice Age had not only cooled the earth, but massive quantities of ice extending well down from the poles had locked up vast quantities of water, lowering the sea level and drying out the atmosphere. African forests shrank and fruit supply diminished, putting the apes at a distinct disadvantage. Monkeys were more nimble climbers and able to eat unripe fruit, whose tannins apes could not digest. Consequently, apes were forced to the dangerous edges of the forests and into the savannahs, where shade from the equatorial sun was hard to find. There in the tall grasses, upright walking became a distinct advantage: it helped them to see by elevating their eyes above the tops of the grasses, and it kept them cooler. Upright posture offered a smaller surface area of the body (head and shoulders) to direct sunlight and opened more of the body to the cooling breezes (Ibid., 106f). "A hairless, bipedal, sweating hominid could have doubled the distance it traveled on a pint of water" (Ibid., 108).

          Steven M. Stanley adds to this picture the likelihood that an upright posture had to have evolved at the same time as a large-brained, totally helpless infant; for carrying the infant would have required that the hands be freed from both climbing and locomotion (Stanley, 1996). A large part of the secret of humanity's evolutionary success has to do with our having a brain so large and an upright pelvis so narrow that childbirth has to take place before the brain is fully grown. As a result, we are born dangerously helpless; but by way of immense compensation, we are exposed to the world and actively learning while the brain is still under construction. This plays no small part in the acquisition of language. No doubt, a linguistic brain is patterned by inheritance, but the job is completed by interactive practice of baby and family members. Practice lays down, develops, and multiplies neuronal connections.

Larger groups, more grooming. Risks from predators and other groups of hominids became much greater once our ancestors had left the protection of the trees, and savannah living forced them to become nomads to survive. They had to follow the movement of food-supplying animals and gather fruits and other vegetables in the seasons when they were ripe and at the locations where they grew. Nomadic life is inherently more dangerous, as it takes a group into new territories where enemies and predators may lurk (Dunbar, 1996: 118f). Thus the pressure to increase the size of the hominid groups grew in order to deal with these environmental threats, and simultaneously the pressure grew on each individual to form more dependable alliances inside the groups.

          They did so by stepping up their grooming activities. Although grooming is essential to keep the skin and fur of the head and back clean and free of parasites, primates devote a fifth of every day to grooming duty. It relaxes them, reduces heart rate (Ibid., 36), and stimulates the pituitary gland to produce endorphins, the natural opiates that suppress pain, elevate mood, and enhance memory (Thain & Hickman, 2000). Opiate highs are no doubt the immediate reward that induces primates to groom one another, and regularly to spend about a fifth of every day doing so; but there is a long-term evolutionary reward as well. For grooming cultivates and cements friendships, and friendly alliances insure that one's day-to-day life will be relatively low in stress. "Only animals that have long-standing relationships groom each other regularly. It is an activity for friends, not for acquaintances. Grooming between acquaintances is often brief, perfunctory and decidedly lacking in enthusiasm" (Dunbar, 1996: 68). Lower status animals with fewer grooming alliances experience higher levels of stress, which in females can interfere with ovulation. Thus poor groomers are less likely to reproduce, with the result that natural selection tends to favor grooming aptitude.

Larger group, larger brain. The larger a group becomes, the more individuals there are to keep track of, and the more brain power is required. Thus becoming savannah nomads favored the reproductive success of the largest brained among our ancestors. Correlation between group size and neocortex size [21] first occurred to Dunbar through studies of bats. Those who live in stable groups, like the vampire bat, have a larger neocortex, spend more time grooming, and have special friends who are essential for their survival. Individuals who have failed to find a blood meal are regularly fed by their more lucky companions who regurgitate part of their own meal to help their less fortunate friends. The regurgitator not only preserves the life of a member of its own clique, it also improves the chances that it will profit at some future date when its own foraging is not successful (Ibid., 65).

[Each animal] needs to know who is in and who is out, who is friends with whom, and who is the best ally of the day. In the social turmoil, these things were in a permanent flux, changing almost day by day. The animal has to keep track of all this, constantly updating its social map with each day's new observations (Ibid., 66).

In addition, the ability to use subtle strategies and "exploit loopholes in the social context" depends on the brain's "computing power" and therefore the size of the neocortex.

          When he plots neocortex size against average group size for different species of primates, Dunbar finds a consistent pattern. In fact neocortex size correlates only with group size and not with size or complexity of diet, size of territory, or any of the obvious ecological factors (Ibid., 62). When these figures are extrapolated to include the human neocortex (which is four times the volume of the rest of the human brain), an average stable group size is predicted to be about 150 members. Dunbar finds this prediction amply supported by anthropological evidence. Contemporary hunter-gatherer clans, Near Eastern villages around 7k, B.P., [22] modern horticulturalist villages in the Philippines and South America, Hutterite communes, and the groups founded by Brigham Young all have about 150 members (Ibid., 69-72).

Taken together, these results suggest that human societies contain buried within them a natural grouping of around 150 people. . . . The figure of 150 seems to represent the maximum number of individuals with whom we can have a genuine social relationship, the kind of relationship that goes with knowing who they are and how they relate to us. Putting it another way, it's the number of people you would not feel embarrassed about joining uninvited for a drink if you happened to bump into them in a bar. . . .Psychologically speaking, we are Pleistocene hunter-gatherers locked into a twentieth-century political economy (Ibid., 77).

Or put the other way around, human groups tend to become unstable when there are too many members for each to feel comfortably related to each of the others. Our brain capacity is not able to handle the complexities of groups larger than 150 members.

Evolution of language. In an Ice Age hunter-gatherer clan of 150 members, Dunbar estimates that each member would have been forced to groom 40% of every day just to maintain a friendship clique sufficient to protect themselves from the inevitable harassment that comes of living in such a large group. This is twice the average of all other primates and simply impossible to sustain in the face of life's other necessities. In fact, such large cliques can only be maintained with language. Gossip is our human variation on the primate evolutionary strategy of sociality. Assuming that four persons would probably be the upper limit for conversation groups, Dunbar observes that gossiping can cultivate three personal alliances in the same time that grooming can cultivate only one (Ibid., 121).

          Since a primate group finds itself under severe social pressure as soon as grooming time rises above about 20% of the day, Dunbar estimates that language would have to have evolved when hominid groups became large enough to require 30% of the day in grooming. Calculating group size from neocortex size, as represented in fossil skulls, Dunbar puts the date of the appearance of language at about 500k, B.P., when group size had reached 115-120 members (Ibid., 110-112).

          This figure of a half million years ago is ten times as far back as conventional wisdom puts the date, 50k, B.P., when the quality and variety of stone tools increased markedly and cultural marvels such as the painted Ice Age caves followed soon after. A second popular candidate for the emergence of language is 250k, B.P., when a significantly larger left temporal area, suggestive of the language organ, becomes undeniable in fossil skulls. [23]

          Dunbar believes that language emerged slowly as grooming time increased. Perhaps the most rudimentary forms of vocal grooming came into use as early as 2m, B.P., in Homo erectus. Probably grunts of approval and disapproval together with fragments of ape-like chatter became formalized into archaic forms of "How are you?" and "Have a look at my left shoulder." Over the millennia it gradually developed into sufficient vocabulary and grammar to make gossip possible -- the form of human vocal interaction that still predominates today. It was at least 400k years between the first emergence of language and the creative explosion of the High Paleolithic (100k - 10k, B.P.). We will have more to say about this date and these cultural phenomena in the next chapter.

Further supporting evidence. A few suggestive details may be added in support of the thesis that language is a human specialization of an aptitude for communication that is evident in all warm-blooded animals, at least. [24] Birdsong provides an obvious parallel, especially since the brains of birds are also asymmetric in favor of the left side of the cortex. Young birds have to acquire their song during a critical period early in their lives, during which a sort of "sub-song" twittering resembles the babbling of human infants (Aitchison, 2000: 8). Baby birds deprived of the opportunity to imitate the songs of their conspecifics develop rudimentary versions of their species' song or radically abnormal songs (Bermúdez, 2003: 154). Not only do baby birds inherit an archetype for recognizing and responding to their own species' song, but they learn that song with regional variations, much as humans have the archetype to recognize and recreate language but need to hear and learn one specific human language.

          The brains of non-human primates are organized to allow them to discern the phonemes of speech, even though the animals themselves cannot produce the sounds of speech (Allman, 1999: 201). Thus, the transformation of the hominid brain in the direction of linguistic capabilities was underway long before we reached the grooming crisis Dunbar describes. Furthermore, chimpanzees in the wild communicate vocally in stereotyped ways. For example, the "pant-hoots" of males to other males appear to establish male-male bonds; and the style of such interaction differs from troop to troop (Aitchison, 2000: 67). Even more interestingly, chimpanzees who have been taught American Sign Language have been observed to teach it to other chimps and to employ it regularly to communicate certain kinds of simple request (Linden, 1974: 130-134).

          One of the most respected treatments of language as embedded in a much larger range of communicative sociality, is Harvey B. Sarles' Language and Human Nature (1985). [25] Sarles begins from the premise that humans are not the only social creatures. "[A]ll social animals do communicate; that is they have a shared repertoire of behavior, well developed cooperative abilities, a social structure, and the ability to pass on an adult standard to infants through parentally mediated experience" (1985: 64). Such abilities also cross species barriers, as in the case of a hunter and his dog (Ibid., 82). Such observations make it clear that "language, gesture, and movements [are] different actualizations of the same process" (Ibid., 200). Children learn to speak not merely by hearing the spoken word. They intensively study the faces of their parents while the parents play question and answer games about their relationship with one another and with the outer world. Seeing the face of the speaker is clearly as important to infants as hearing, which is why it takes much longer for a child to transfer its language skills to the telephone, where visual cues are missing (Ibid., 201).

          We are so impressed with the uniqueness and flexibility of language that we tend to overlook the reality that neither talk nor writing takes place in a behavioral vacuum. The entire interactive behavioral context gives each spoken sound essential elements of its meaning. Thus learning to read French in school turns out to be insufficient preparation for communicating on a Paris street corner. Each phoneme looks the same on paper but sounds different depending on the interactive context in which it is uttered. Sarles notes that if he cuts a film into pieces to compare similar gestures, the sound of the speech accompanying those gestures may sound quite strange because the context for interpreting it has been lost (Ibid., 106).

          All of these details and many others that could be gathered demonstrate that Dunbar's thesis must be on the right track. Human language evolved out of the most social of mammalian orders, where communication is a constant daily necessity. It takes place via sound, gesture, and other behaviors; and the brains of lower primates are designed so as to detect the sorts of nuance that became the foundation of language. The language archetype, we may say, is a refinement of social communication, the hallmark of the primate order, and no stranger to other warm-blooded animals, including birds and cetaceans, at least.

The Meaning of Archetype

          Using human language as the prototype, we can now offer a description of the notion of archetype that can be used later on, either (a) to decide whether an alleged archetype satisfies the criteria we have established through examining language, or (b) to rethink our definition.

          An archetype is a pattern of behavior, of recognition, and/or of imagination that is universal in a species. It is an aspect of instinct, namely the instinct's innate guidance system, whereby relevant objects, conditions, situations, relationships, and the like, are recognized as salient and are organized with an aptitude that is both striking and universal. The three-year-old child's swift grasp of vocabulary, grammar, and syntax and consequent ability to create its own original sentences illustrates the nature of the innate guidance.

          If a pattern is archetypal, it should in principle also be traceable in the neuronal networks of the brain. Modern imaging techniques (CAT scans, MRI's, etc.) can reveal the activation of specific brain tracts while the pattern of behavior is enacted. Alternatively, the release of certain hormones, synaptic transport molecules, and the like, may be identified as universal manifestations of the pattern. In this sense, then, the archetype's guiding function manifests not only in an observable behavior but in the building up of the brain itself. Ideally it might be hoped that the genetic roots of these organic structures can be identified in the form of stretches of DNA code. Failing that, perhaps elements of the emergent process of the archetype's realization may be isolated.

          As a species-universal pattern, an archetype's expression will resemble the behaviors seen in closely related species and allow us to trace a hypothetical line of inheritance back to some distant, possibly extinct, ancestor species. In every case the expression of the archetype and its "precursor" patterns in related species will evidently have survival value. Historically, behavior guided by the archetype will be seen to have surmounted some sort of ecological crisis for the species and enabled its offspring to have some advantage in the struggle to proliferate. That is what it means to be "naturally selected."

          The archetype is an "empty program" requiring cultural "content" to fill it. The archetype belongs to the nature of being human and applies to all persons and all races and cultures. But each culture has the freedom to find its own way of imagining and enacting the universal pattern. Hence, language: we are all innately equipped to become linguistic beings despite the fact that our linguistic capacities can be satisfied with any human language, regardless our racial or ethnic heritage. And once we learn that first language, our empty program has been filled, and the learning of other languages becomes difficult. We learn what is presented to us, but organize it in typically human ways.

          Strong emotional values are also attached to enacting archetypal patterns. There can be not doubt of this in regard to sex and nourishment, and we have established the release of endorphins in the case of grooming. Consider, too, the enthusiasm with which a child celebrates and extends its newly acquired ability to walk -- or the extraordinary attention and joy with which it tries out new words. We have little evidence for it at the moment, but I think we have to grant that Jung's ascribing "numinosity" (compelling emotion) [26] to the archetypal experience is probably correct. When we encounter universal patterns, we are whisked out of the "noisy ephemeral life of the present" and feel the same joyous or resigned compulsion that our ancestors felt -- all the way back, perhaps, to those squirrel-like primates who tried their best to stay clear of the dinosaurs.

 

---

§
1. In what follows, I have not attempted an historical account of how Jung's idea of the archetype developed over the course of his career. Two recent articles have addressed that material. Gary V. Hartman (2003) gives us a complete but sketchy overview, while George B. Hogenson (2001) discusses Jung's relationship to Larmarckian formulations. Far more thorough is Sonu Shamdasani (2003), who abandons an attempt to present a linear developmental scheme in favor of providing much historical depth for every element in Jung's always confusing and sometimes contradictory claims.
   
   §
2. I have discussed such experiences in Divine Madness (Haule, 1990).
   
   §
3. Cf. CW8: 254, 275f; CW9i: 5; CW11: 88.
   
   §
4. Cf. CW10: 14; CW11: 518.
   
   §
5. Gregor Mendel (1822-84), an Augustinian monk who discovered the principle of the dominant and recessive genes through research on pea plants. His results were published in 1866 and ignored until three separate investigators rediscovered them in 1900.
   
   §
6. Similar passages may be found at 10: 53 (1927/31); 18: 1228 (1935); and 18: 539 (1961). In one place (1958) Jung distinguishes his approach to the psyche from Freud's in that Freud "would not let himself be taught by the findings of ethologists and historians" (10: 659).
   
   §
7. I say "largely" repudiated, since something remains in those evolutionary theories that speak of "neoteny" whereby a species B that evolves from species A retains a more juvenile anatomy in the adult form -- especially displaying cranial and facial characteristics belonging to the juvenile form of the earlier species. For example, adult Cro-Magnons are said to have a neotenal relationship to Neanderthals, even though they are not direct descendants.
   
   §
8. The term "drive" (Trieb) has entered the language of psychanalysis as a rough equivalent for "human instinct," since both Freud and Jung were careful to follow the German tradition of ascribing "drives" to humans where "instincts" (Instinkte) would be ascribed to animals (Shamdasani, 2003: 191).
   
   §
9. Cf. CW8: 268, 277, 398; CW10: 547; CW18: 1260.
   
   §
10. Cf. Bermudez, J. L. Thinking Without Words. Oxford: University Press, 2003.
    
    §
11. Cf. CW5: 388; CW8: 405, 841, 590); CW10: 530.
    
    §
12. Aitchison (2000) comments on the significance of Pinker's work on language. In 1866, the Linguistic Society of Paris refused to accept any papers speculating on the origins of language or the invention of a universal language; for there had been too much unfounded nonsense proposed. In 1990, Pinker and Paul Bloom reopened the discussion and set it on a solid foundation in their paper "Natural Language and Natural Selection."
    
    §
13. Perhaps the strongest statement we can make is this one from Frans de Waal: Kanzi, "the world's most famous bonobo . . . can correctly identify a large number of English words heard over earphones to prevent unintentional cuing" (de Waal & Lanting, 1997: 44). Possibly more is to be learned by letting apes teach us instead of the other way around: "We do not know the meaning of the ape gestures . . . nor do we know very much about the mind of the ape . . . Our task now is to learn the language of the animal in order to understand its perceptions. Then we may communicate" (Candland, 1993: 351).
    
    §
14. "[T]he `fallacy of misplaced concreteness' . . . consists in neglecting the degree of abstraction involved when an actual entity is considered merely so far as it exemplifies certain categories of thought" (Whitehead, 1929: 10). In short, it means treating a concept as though it were an object.
    
    §
15. Evidently, more than one "mother tongue" can be learned by children whose parents speak different languages or who are otherwise exposed to the full range of a second or third language during the formative period of childhood. The example of Helen Keller's rapidly learning language and conceptual thought as an adult shows, too, that something that had been present from birth had become "activated."
    
    §
16. Ernst Mayr, German ethologist, cited in Stevens (1983: 51): The "open programme" of an instinct "enables the organism to adapt appropriately to environmental conditions."
    
    §
17. Broca's aphasics (those suffering damage to Broca's area) have severe difficulties with grammar, while Wernicke's "fluent aphasics" produce grammatical sentences with little meaning (Smith & Szathmáry, 1979: 150).
    
    §
18. We owe it to the physicist Niels Bohr to call such apparent contradictory descriptions of nature "complementary." Both descriptions are valid, but they cannot be observed or studied simultaneously (Freeman J. Dyson, "One in a Million," New York Review of Books ( March 25, 2004): 4-6.
    
    §
19. The initials stand for Adenine, Cytosine, Guanine, and Thymine. Each base is connected to a carbohydrate molecule bearing a phosphate ion; and the whole unit (base, sugar, and phosphate) is called a nucleotide. Linked nucleotides form each of the dual backbones of the DNA double helix.
    
    §
20. The smallest protein contains fifty amino acids.
    
    §
21. By "neocortex size" Dunbar means the ratio of the volume of the neocortex to the volume of the rest of the brain.
    
    §
22. 7k, B.P. means 7000 years Before the Present; it would be the same as 5000, B.C.E.
    
    §
23. Dunbar sees asymmetry in brain development as continuous from the age of the trilobites (250m, B.P.). All these animals have scars more predominantly on the side of the body corresponding to the modern right, or less dominant half, of the brain. Right-handed humans, for example, usually show their left side (under control of the right, non-dominant, non-linguistic hemisphere) to adversaries. (Note that the side is reversed for invertebrates.) (Dunbar, 1996: 137f.)
    
    §
24. "Warm-blooded, at least": There is clear evidence that amphibians, insects, and other cold-blooded animals also communicate a great deal -- sometimes primarily by pheromones, but also by sound, behavioral display, etc.
    
    §
25. Originally published in 1977 as After Metaphysics: Toward a Grammar of Interaction and Discourse. Lisse, Netherlands: Peter de Ridder Press.
    
    §
26. The "numinous" is a category Jung borrowed from Rudolf Otto (1923), The Idea of the Holy. The term derives from numen (Latin) "having the force of a divine command."