Two Longstanding Metaphysical Errors in Explaining the Origin Of Language

A post in which I find more ways to criticize Noam Chomsky and Steven Pinker

Peter Brueghel the Elder’s “Tower of Babel” (1563)

I’m a believer in the idea that the way language first began is an important question. And I don’t need to spend a whole lot of energy justifying the claim that it’s a universally interesting problem as well. Since man first came up with the ability to tell stories, he has almost certainly been fascinated by his own ability to speak, the foremost feature that separates him from the rest of the animal world. But the way we’ve approached the question hasn’t been particularly great for most of history. This is unfortunate, but it’s also understandable. The question of the origin of language might be the hardest problem in science, and we haven’t had an especially motivated reason to truly understand it for the majority of our existence on earth. “Just so” stories have been able to suffice.

For the purposes of this discussion, there are two key ideas on language’s origin that I’d like to examine, both of which I consider mistaken, and I’d like to remark on their surprising tenacity. The first idea is that there was a “first language” that was both unified and grammatically complex, which then separated off into a bunch of other languages over time. We can call this the myth of an ur-language. The second is that language began for the purpose of making declarative statements rooted in logic.

I. The Myth of an Ur-Language

Generally speaking, the main theories on the origins of language for most of pre-evolutionary western history have posited a monolingual origin. That is to say, language started off as the expression of just one system — often (though not always) seen as the best one — and then it broke off into a bunch of mutually unintelligible different systems, and thus a bunch of different languages.

Time and time, you see this basic pattern. This Wikipedia page makes it pretty apparent. For instance, here’s the African Wasania tribe:

The Wasania, a Bantu people of East African origin have a tale that in the beginning, the peoples of the earth knew only one language, but during a severe famine, a madness struck the people, causing them to wander in all directions, jabbering strange words, and this is how different languages came about.

Here’s the Salishan tribe of North America:

A Salishan myth tells how an argument led to the divergence of languages. Two people were arguing whether the high-pitched humming noise that accompanies ducks in flight is from air passing through the beak or from the flapping of wings. The argument is not settled by the chief, who then calls a council of all the leading people from nearby villages. This council breaks down in argument when nobody can agree, and eventually the dispute leads to a split where some people move far away. Over time they slowly began to speak differently, and eventually other languages were formed

Here’s the Ticuna tribe from South America:

The Ticuna people of the Upper Amazon tell that all the peoples were once a single tribe, speaking the same language until two hummingbird eggs were eaten, it is not told by whom. Subsequently, the tribe split into groups and dispersed far and wide.

And the examples go on and on.

You’ll notice that these theories don’t typically approach the question of why or how the ur-language itself first started, and these groups don’t seem to consider the possibility that man, at one point, couldn’t speak at all. The stories typically begin in medias res — there was a group of people saying stuff in a particular way, and then some calamity occurred that made people say the same stuff but in a different way.

This particular pattern of thinking suggests an ancient, preliterate epistemology. After all, it’s pretty normal to see yourself as the center of the universe when you don’t have a lot of information to tell you otherwise, and it’s also reasonable to identify other tribes — tribes who compete against yours — with strife, and then see language as the cause or at least a prominent factor in all of it. But this mentality persisted for many, many centuries in the west, probably longer than it should have. And it was passed on either directly or indirectly through Jewish and Greek metaphysics.

(Bear in mind that I’m partly cribbing from this blog post, which itself cribs from a YouTube video by a Master’s Degree student that is now evidently offline, and I can’t find his thesis, so you’ll have to pardon me if this section seems vague or unsubstantiated at times).

In the Bible, the text is pretty clear about it: someone built the Tower of Babel, and the hubris of such an undertaking was responsible for the linguistically diverse state of affairs that we’ve inherited (on the island of Hao in Polynesia, there’s a very similar story, so it doesn’t speak much to the uniqueness of the Israelites’ way of thinking). The Bible represented a comparatively advanced way of thinking compared to that of other pagan belief systems, but its account of how languages became diverse was pretty par-for-the-course.

However, I think that Platonic metaphysics bolstered the impulse to see language as originally monolingual as well, albeit with a more theoretical approach: in the (neo)platonic system, everything in its originary state of being was unified as pure Nous, or spirit. But then, things emerged out of it, which became increasingly corrupted as they grew disparate, weighty, corporeal, and telluric. The fall and spread of languages from a unified ur-language would seem to fit with this basic description of how all of matter first formed, since the description suggests that things begin in a pure state and grow tarnished as they assume diverse and widespread forms. And that principle also corresponds to the native Greek myth of Hermes spreading diverse languages to various peoples, causing discord and strife.

II. Language as Primarily a Tool for Logic

In any case, it’s clear that the Israelites’ story of the Tower of Babel was not a radical departure from other ancient myths on the origin of language. Where the Israelites were a bit more unique than the other groups, however, was in their claim that the first thing Adam did after being created was to name all the birds and beasts. Here, language is more or less determined by a human, and it’s given a primarily logical function. Language’s responsibility in this case is to say “this = that, that = this,” and so on, the nouns setting up the framework for syntax to fill in the blanks. This act of Adam is not so different from the way God creates the whole universe by saying “fiat lux,” or “Let there be light,” essentially speaking it into being (perhaps in a language that differs from the human speech that Adam uttered — though the subject is complicated, because God and Adam often talk to one another aloud in Genesis).

The act of peacefully speaking the world into being also amounts to a major departure from pagan creation stories, which often begin with a passionate act of violence, with some slain chthonic god’s corpse forming the earth. In those pagan stories, the world is created through the heat of battle — emotion; passion. In the Israelites’ account, it’s created through an act of speech, with a verb phrase and a noun phrase.

So for the majority of Western Civilization’s development, these two notions prevailed: first, that language started out as a unified system which eventually split off; and second, that language was primarily about naming things and (presumably) making declarative statements out of them. Regarding the first notion, it’s not as though other possibilities were inconceivable. Epicurus, for instance, argued that human language must have developed amid a diversity of people saying things in different ways. But this idea didn’t go very far — at least, not for a thousand years or so.

III. Language as a Tool for Emotive Expression

As for the second notion, there were other classical philosophers like Lucretius (an Epicurean from the Hellenistic period) who put it into question by arguing that emotional, passionate language likely preceded the desire to put names to things — but again, the ideas didn’t go far until the Romantic period. The bridge between Lucretius and the Romantics was Giambattista Vico (1668-1744), who was barely read during his lifetime but proved quite influential on the Romantics later on down the road. He argued that post-diluvian man’s first mode of thought was dominated by sense and imagination, not reason, and thus it was animistic, passionate, and mythical. Essentially, he felt that language was first expressed through poetry rather than prose.

Vico’s idea was to become common by the mid-18th century. As M.H. Abrams writes in his The Mirror and the Lamp (1953),

Theorizing about linguistic and poetic origins became a popular occupation among Scottish writers of the mid-eighteenth century — Blair, Duff, Ferguson, Monboddo, and others — who had in common an absorbing interest in the reconstruction of the genesis and prehistoric development of human arts and institutions. (Men “who love to talk of what they do not know,” is the way Dr. Johnson drily characterized such conjectural historians.)

[…]

What later on the German philologist Max Müller was jocosely to call the “pooh-pooh” theory of language and poetic origins achieved international currency in the course of the eighteenth century. As early as 1746, Condillac introduced these ideas into France, in his Essai sur l’origine des conaissances humaines. In his posthumous Essai sur l’origine des langues, Rousseau, whose general emphasis upon the primacy of instinct and feeling influenced the emotive theory of art, insisted that since man began not by reasoning, but by feeling, the first words were cries extorted by passion, and the first languages were song-like, passionate, figurative, and therefore the language of poets, not geometers. In Germany, Hamann combined a mystical view of the divine inspiration of language in the garden of Eden with the assumption that this pristine speech was musical and poetic. His young contemporary, Herder, said in 1772 that language had been both expressive and mimetic in origin, and therefore doubly poetic. […] And the concept that speech and poetry were co-original in the press of feeling, and that even in their developed forms they are analogous as expressions of spirit, became a commonplace in the linguistic speculations of the German romantic generation.

Even when the pendulum swung back towards a logic-based understanding of language, Nietzsche came along. And after he died, an assortment of post-Nietzschean vitalists continued to dismantle the idea of language as primarily a tool of logic. The most important of them was Ludwig Klages, whose ideas on language and its relationship to logic are worth reading elsewhere in full.

IV. Chomskyan Linguistics As a Revival of Judeo-Christian Metaphysics (with a tinge of the Neoplatonic)

Yet nevertheless, the pendulum did swing. In the 20th century, the romantic understanding of language had receded back into obscurity entirely by the time Chomskyan linguistics was born.

I don’t want to go over everything about Chomsky’s understanding of language, because it’s a huge topic. But there are a few elements I’ll briefly highlight: one is that that Chomsky first proposed the idea that humans have an instinct for language because syntax has a skeletal structure (he called it “universal grammar”) that is stored within our brains from birth. The other is that language is essentially a computational tool with productivity as its major feature. By “productivity,” he meant that syntax can produce statements about any combination of things, even though a sentence can be entirely “meaningless” (he coined the sentence “colorless green ideas sleep furiously” to support his view that a sentence can be perfectly grammatical yet devoid of all meaning). The corollary to this view is that grammar is, by nature, abstract, and thus its form is distinct from the meaning of what it says. The figure and the ground are wholly distinct from one another.

These two contributions from Chomsky, while they appeared to resemble the cutting edge of scientific reasoning at the time, are instead I think the extension of how western metaphysics had always understood language, that is, excluding exceptions like the Epicureans and Romantics.

V. Universal Grammar and Mentalese as Ur-Languages Designed For Logical Postulations

Consider the idea of universal grammar. On the one hand, Chomsky’s idea was an improvement over the ultra-empiricist Behaviorist conception of language acquisition, because the Behaviorists had no way to account for how rapidly children learn language. But on the other hand, the way Chomsky actually drew up the properties of universal grammar was flawed, and this is why he and his followers had to revise it over and over throughout the decades, something for which he’s been widely criticized. In its first form, it was very specific, and it really did hearken back, however unintentionally, to the archaic theories of a once-perfect, long-lost ur-language. In the blog post I linked above, the author makes this tentative remark (my emphasis):

Later [post-Platonic] Ancient theories saw linguistic diversity as a fundamental aspect of the early stages of language evolution. They also thought about the experience of individuals in a community. The Judeo-Christian approach to language undermined this position and re-introduced elements of the idea of a ‘perfect language’ and a monogenetic origin of language. To get even more Dan Brown about it, I might suggest that this set the stage for the idealism of Chomskyan Generativism. It is only relatively recently that the role of diversity in studies of the origin of language has been foregrounded again.

Honestly, I don’t think he needs to use such self-deprecating language here. There’s nothing “Dan Brown” about speculating on why the most popular language theory of the 20th century — the very thing that allowed the field of linguistics to present itself as a legitimate science — postulated a then-popular but now-(basically)-discredited model of syntax that supposedly every human on Earth is born with because it’s located somewhere within their brains. Of course, Chomsky’s whole idea was that no language is better than the other, since they all amount to variations on this merely skeletal universal grammar. So it isn’t as though he thought that universal grammar could be best realized within some kind of perfect language. But this skeleton he postulated also did far too much heavy lifting, and it placed far too heavy expectations on theories of how language came to be.

This is because Chomsky’s theory implied that grammar is some sort of evolutionary adaptation (why else would something so specific occur innately within the human brain?), yet it also would need to have shown up fairly quickly, since nothing like it is seen in man’s nearest evolutionary ancestors, the apes. Moreover, he would need to show how the brain could evolve to have a built-in grammatical structure that worked completely independently of what it could potentially think about, i.e. demonstrate how the form of man’s thinking could be altogether separate from its content. And finally, to make matters all the more frustrating, he was generally dismissive of any attempts to theorize about how humans evolved to have the “hardware” for universal grammar. He basically set up an evolutionary theory but for decades frowned upon any attempt at a follow-through.

To solve these problems, evolutionary psychologist Steven Pinker went ahead and ignored Chomsky’s dismissals of evolutionary explanations and argued that Chomsky’s theory was correct: man evolved to have an embedded structure of syntax located somewhere in the brain. To do it, he made heavy use of something called “mentalese,” an idea he borrowed from analytic philosopher Jerry Fodor. Mentalese is the idea that the mind can think — and indeed think rather complex thoughts — without a language that has distinct articulated signs. This idea of mentalese would allow him to argue that syntax can be entirely distinct from semantics because humans can think in a completely internal, silent “language” with no words. It also allowed him to justify one of the major implications of universal grammar, namely that the apes evolved the ability to think logically before they ever had any words. In other words, syntax (an abstract computational structure) came first, and then words came second. Chomsky was merely approaching the arcane notion of a purely abstract ur-language, but Pinker fully realized it. And, in case my tone hasn’t made it clear, the idea is nonsense.

The criticisms against “mentalese” are plentiful, but the neuroscientist Terrence Deacon in his Incomplete Nature (2011) has summed up the problem succinctly:

If thought equals language, then all that needs to be explained is the translation. But wait a minute! Exactly who or what is producing and comprehending mentalese? Modeling thought on language thus implicitly just pushes the problem back a level. […] For a language to be “spoken” in the head, an unacknowledged homunculus must be interpreting it. Does this internal speaker of mentalese have to translate mentalese into homunculese? Is mentalese self-interpreted? If so, it is not like a language, and it is unclear what this would mean anyway.

The problem is that all language is made up of symbolic signs (as defined by C.S. Peirce), in this case words. Words require an outside horizon of things to which they can refer, but they themselves also must be “outside” in the sense that they have to play an intermediary function. If you have a purely internal sign that suggests an outside referent but it doesn’t play an intermediary function (i.e., to be read/heard and understood potentially by others), then you’ve got maybe an icon or an index (again, see Peirce), which all animals have as well, but nothing like the building blocks of language.

Another reviewer puts it this way:

At the beginning of chapter 3, Pinker trots out Wittgenstein's remark that a dog cannot think "perhaps it will rain tomorrow" as an example of a philosophical claim that animals lack consciousness. He lumps this in with the Sapir-Whorf hypothesis, and later dismisses the whole package as "wrong, all wrong", styling cognitive science as the defender of common sense (that thought is independent of language). There is indeed something wrong here. Consciousness is a separate can of worms from the language-thought relationship. And common sense tells us that doggie thoughts are as complicated as doggie communication skills, no more and no less. Thought is not identical to language, but the two are closely related by their very definition. Ironically, Pinker shows us that cognitive scientists feel the same way: because they are unable to discuss raw thought, they immediately imagine it as being couched in a fictional language, mentalese.

With this in mind, it is interesting to look at the examples that Pinker uses to argue for the primacy of thought over language. For most thought (and certainly most interesting thought) the medium of communication is language. But more primitive behavior can communicate very simple thoughts. This is what is happening in the example of a baby looking surprised when it sees one object suddenly turn into two. Pinker seems to imply that the baby is not just thinking "Wow!", but something more sophisticated, such as "I assume that things don't split into two like that, but this sense data is inconsistent with my assumption. I must look more closely." These sentences may be related to what is in a good cognitive science model of the baby's behavior, but to say the baby is actually thinking this is silly. We could just as well claim that a falling stone is thinking about Newton's law of gravitation, or, in the case of a particularly clever stone, Einstein's.

And again, I don’t think I need to add to what this reviewer has understood so well: it’s silly to suggest that “mentalese” can do so much legwork while our actual articulated words are mere shadows of what the mind does. Mentalese takes on a mystical dimension. It becomes something like René Guénon’s Language of the Birds.

Should I stop quoting things that bash Pinker and his nonsense? I suppose I should, but the temptation is nearly too strong to resist. And given how wildly enthusiastic the reviews were for his work when he first published it in 1994, I feel like it’s my duty, thirty years later, to make sure that such an embarrassing theory of human communication is never produced again.

But my point here, in this discussion, is really to show perhaps why people felt so inclined to believe in the theory. The philosopher of language John Searle once remarked that Noam Chomsky’s ideas on language really do feel highly peculiar when you sit down and think about them for just a little while. I agree, but there is also something reassuringly familiar about them. Distant though they might be, they bear the traces of a kind of linguistic theology that remained with use for centuries. Chomsky and Pinker’s formulations were so tempting because they primed people to use evolutionary reasoning as an excuse to rehabilitate the old myths of a pristine and unfettered ur-language, perfect in its logical capabilities, and unmolested by the tangled, muddy world of exterior objects and things.

VI. Conclusion

I’ll end this discussion with a list of a few propositions I think are correct about the origins and nature of language, and of course you’ll understand that these are fairly tentative. You’ll indulge me anyway:

• Words came first, syntax came second.

• There was never an ur-language: different languages developed coextensively, and this co-extensive development contributed to the complexity of grammar.

• Language has always been sloppy and messy, and it always will be, but nevertheless we’re probably better off trying to sanitize ours as much as possible.

• Language came about alongside other various physical gestures such as hand signals.

• Language first accompanied various physical objects playing a mnemonic and/or symbolic role, and our use of such objects has never gone away (e.g. wedding rings).

• Words were first used to express emotions rather than name things, but the systems they were couched in were so primitive that they couldn’t properly be called “language” until much later, as they bore no distinction from normal animal communications systems.

• Once grammar had reached the point in which noun phrases interacted with verb phrases (to borrow from Chomsky’s generative terminology), language carried the roles of making declarative (logical) statements, expressing the passions, and aiding in mnemonics, all in roughly equal fashion.

• Deictic words like “this,” “that,” “here,” “now,” “then,” “there,” “I,” “you,” “he,” “they,” and so on, came before proper nouns.

• The human brain did evolve for language up to a point, but not with any clearly localized features, and with hardly anything so complex as a universal grammar. The brain doesn’t have the hardware to tell the difference between a preposition and an adverb.

• The locus of language is not in the human brain, nor in platonic forms, nor in any purely internal or otherworldly realm for that matter, but rather in the intersection between the mind and the physical world. As soon as language, or any symbolic system, has become unmoored from the worldly and telluric, it has taken on a parasitic quality (I’ll probably write about this point later).

• All linguistic statements carry the potential for meaning, since the receiving human mind creates meaning out of them. Meaning cannot exist in a vacuum. There is no such thing as a “syntactically correct but semantically meaningless” utterance.

• Now more than ever, we ignore the social function of language at our peril.

That’s all for now.