an human beings with no common language still communicate? The phenomenon of glossolalia, or "speaking in tongues," has been documented since Biblical times. The plot of Neal Stephenson's bestseller Snow Crash even turns on the point, and anyone who's ridden a train through some foreign land has probably experienced something similar, oohing and aahing over the scenery with fellow travelers from, well, who knows where? They can't tell you, and you can't ask, but it hardly seems to matter.
It's tempting to speculate that humans have some innate communication lurking beneath the complex grammar and vocabulary of our 4,000-odd spoken languages. Telepathy in at least the literal sense: perception at a distance. And indeed, studies of human language have revealed striking evidence that this is so.
Babies provide one important clue. Consonant sounds, whether spoken or whispered, are generally formed when air is forced across mouth parts that are briefly pressed together. Vowel sounds, by contrast, are formed when the mouth is open. (In between there are whole classes of sounds that we won't worry about. "Sometimes Y," yeah. Sometimes H, and W, and R....) Anyway, "baby talk" consists primarily of repeated consonant-vowel pairs--the familiar "ga ga, goo goo" of cartoons and, surprisingly often, of actual babies. The interesting thing here is that, as University of Texas psycholinguists have discovered, 7-to-12-month-old babies all over the world "speak" in exactly the same way, regardless of cultural or parental influence.
Transition from animal calls
Specialized baby words like "mama" and "dada" and "baba" (bottle) occur in nearly every language; clearly, this isn't something we learn, just something we do, like seeing. And from a pure communications standpoint, repeating syllables twice is a great way to boost recognition. This may well have been important in our ancestors' transition from animal calls to actual, symbolic words.
Of course, baby talk eventually gives way to more complex babbling, which strings together the specific phonemes of the child's soon-to-be-native language. Soon afterward, actual vocabulary words begin to appear, and then a rudimentary grammar, and finally all the fiddly little exceptions and subtleties that are the hallmark of mature languages everywhere. Humans seem to require a certain amount of complexity and ambiguity in speech, although exactly where and how this occurs (e.g., inflection vs. conjugation vs. tone) is a matter of regional style.
Oddly enough, though, when speakers of different languages collide (as happens in trade, migration, and various forms of warfare) the need for communication rapidly spawns simple "pidgin" languages that borrow and merge important words from the parent languages. If the pidgin persists, later generations will recomplexify it into a "creole," and finally a whole new language. But as linguist Noam Chomsky discovered in the 1960s, the grammar of pidgins is not only simple and forgiving, it's almost always the same. That is, certain rules, such as repetition of words to express plurals ("Tiger, tiger, tiger") or emphasis (big, big tiger!), appear to be natural expressions of some "deep wiring" our brains are born with.
And really, this is no surprise; the brain regions responsible for language are as discrete and unambiguous as the centers of vision, and have been known--though poorly understood--since the mid-19th century. Broca's area, just over and in front of the left ear, controls speech production and vocabulary. Wernicke's area, located a few inches further aft, is where speech recognition occurs. There are other areas crucial for language as well, such as the auditory cortex (hearing), angular gyrus (reading) and motor cortex (vocalization and gestures). Because long-term memory consists simply of interconnections between brain cells, and instinct consists of genetically coded "memories," it's quite reasonable to suppose these areas, like others in the brain, are formed in a state of less-than-total naiveté.
Of course, humans speak in gestures as well as words. Many of us find it difficult to speak coherently when our hands are occupied, and the very existence of complex sign languages, to say nothing of their ready adoption by preverbal children, implies some natural ability to communicate this way. There even appears to be a "baby talk" of repeated hand gestures! Sign language specifics vary considerably from one culture to the next; but here, too, we find universals, such as pointing, nodding and head-shaking. Actually, there have been a few isolated cultures where the meaning of nod and shake were reversed, but the linking of head movements to the abstract concepts of "yes" and "no" is constant even here. So again, we see the hallmarks of genetic coding beneath the assorted complexities of culture.
The 1929 Sapir-Whorf hypothesis speculates that our thoughts are shaped, and in fact sharply restricted, by the rules and vocabulary of our native languages. While there is undoubtedly some truth to this, rare "feral children," raised by animals or indifferent human parents, point to a more complex answer. Feral children who are found and socialized after their toddler years are not generally able to learn whole languages in the adult sense. Their vocabularies tend to be limited to a few hundred unsubtle words, and their grasp of grammar is particularly poor--pidginlike, one suspects.
However, the ability of feral children to describe events and feelings from before their acquisition of language proves that Sapir-Whorf is not the dominant force in human thought. Instead, recent linguistic theory points to a kind of "mentalese," or preverbal symbolic language, which our cerebral cortex employs instinctively for manipulating abstractions, and later translates into learned forms of speech, writing or gesture. Mentalese, like pidgin and feral languages, is probably rather simple, lacking the richness and subtlety of "real" language. This may be why "talking out" complex ideas and feelings helps to clarify them--because the internal symbols to represent them simply don't exist.
Proto-Indo-European spoken here
The study of dead languages does nothing but strengthen these theories. Modern European languages are mostly derivatives of classical Latin, Greek and Germanic; and these languages, along with the Sanskrit that preceded them, express the same essential concepts, in roughly the same ways, as any modern language. They also show signs of having been creolized and recomplexified many times. Languages older than Sanskrit are harder to analyze, since they predate the invention of writing, but the science of comparative linguistics, using statistical methods adapted from genetics, now permits researchers to identify similarities between related languages, and thereby infer details of the mother tongues that preceded them.
By these methods, the ancient tongue of Proto-Indo-European--grandmommy of Sanskrit, and of languages spanning all parts of the modern world--has been fairly well reconstructed, despite the absence of living speakers for 5,000 years or more. For example, we can say, with reasonable confidence, that nearly every word in the language consisted of a conceptual root, plus a suffix to modify its meaning and usage ("-age" is an example of an English suffix, modifying the word "use"), and finally an ending to specify details such as singular/plural and masculine/feminine.
We can also cite specific words, such as "ano" for ring or circle, "ekwo" for horse and "nobh" for knob or protuberance. Some details may be a little off from the original, and undoubtedly many words have been lost or have yet to be rediscovered, but the high-confidence vocabulary now includes about 2,000 words--probably a substantial fraction of the total. Interestingly, while we don't know who the Proto-Indo-Europeans were, or exactly where or when they lived, we can deduce a great deal about their culture by the presence of words for "wheel," "copper," "trade" and various forms of grain, and the apparent absence of words like "empire," "capital investment" and "spaceman." It's hard to say just how far back these methods can reach, but already there are researchers attempting to link Proto-Indo-European with reconstructed languages from Asia and even Australia and the Americas. The Holy Grail of these investigations is Ürsprach, the language of a hypothetical parent tribe from which all modern cultures are descended. Whether or not comparative linguistics can find that signal across the noise of tens or hundreds of thousands of years, there almost certainly is an Ürsprach buried deep in our past.
And it may not be so hard, after all, to know what sort of language it was. Probably a very simple one, combining a few dozen familiar gestures with a small vocabulary of words built up from consonant-vowel pairs. It probably had a forgiving grammar, with rules any 21st-century parent or pidgin speaker would recognize. In fact, humanity's first language may bear a striking resemblance to every human's first language.
Ga ga, goo goo, little ones--the past and future are yours.
Wil McCarthy is a rocket guidance engineer, robot designer, science fiction author and occasional aquanaut. He has contributed to three interplanetary spacecraft, five communication and weather satellites, a line of landmine-clearing robots, and some other "really cool stuff" he can't tell us about. His short fiction has graced the pages of Analog,
Asimov's, SF Age and other major publications, and his
novel-length works include Aggressor Six, the New York Times Notable
Bloom, and upcoming The Collapsium.
