The Riddle, the Mystery, and the Enigma

London - 1912. By age 35, Winston Churchill had already arrived.

Winston Churchill said Russia was a “riddle wr/apped in a mystery inside an enigma.” The same can be said of the origins of language. It is not just one puzzle but several and they are tangled together. Language itself rests on a unique social structure that provides a unique function. It is build up from a system of sounds, symbols, and syntax that have nothing else like it, perhaps in the world, certainly not among primates. Thus, instead of having to explain the origin of one or even two novelties, we need to understand the appearance of a series of remarkable developments, sorting out the riddle, the mystery, and the enigma as we go.

In the part of the quotation that usually goes ignored, Churchill ended his observation by saying, “but perhaps there is a key.” This blog has long persisted on the assumption that there are two keys to unlocking the tangle—the theory of evolution and the speech triangle. Last week’s post (here) described a series of innovations proposed by Ray Jackendoff and analyzed in a paper by Willem Zuidema and Arie Verhagen. I wonder what would happen if we combined that paper’s logic with my possible keys.

One thing that happens right away is that we abandon talk of a continuum from ape cries to modern language. The Zuidema/Verhagen paper concludes that language contains ‘a number of ‘design features’ that distinguish us from nonhuman primates” [p. 62]. In this statement the authors ignore both keys to the riddle/mystery/enigma.

It is a little dangerous to write of “design features” and evolution in the same breath. I know the phrase here was used metaphorically, but it points in the wrong way, away from structure and function and toward progress. It encourages Lamarckian sentences like, “The combinatorial nature of speech follows from the difficulty of production and the large repertoire of words in human languages [55].” A more Darwinian sentence would have “[overcoming] the difficulty of production and the large repertoire of words” follows from “the combinatorial nature of speech.” The evolutionary key is helpful in suggesting which way the causes flow.

The other key, the speech triangle (a speaker and a listener focus their joint attention on a topic) helps sort out speech’s communal structure. (By the way, the speech triangle works for any medium-sounds, signs, or even writing.) It is invaluable at suggesting the function of some aspect of language.

Last week’s post looked at the things that had to change to get from ape cries to human language, regardless of the functions they hold. It listed eight necessary innovations and lumped them into three categories: conceptual innovations, phonological innovations, and innovations associating concepts with sounds. (The authors limited themselves to considering phonology, but the same innovations are possible with hand signs, so their theory appears to work for either modality.)

By taking the two keys and the innovations in last week’s post we can move along in addressing the puzzle.

The riddle: How did language get to be so flexible?

Linguists long ago found the basic solution to this question. Language becomes flexible by combining elements. Elemental sounds/signs get combined into words. Words get combined into sentences. Sentences combine into paragraphs or conversational exchanges. And on and on. From the perspective of language origins, the mystery is why and when these combinatorial elements appeared. Jackendoff’s list puts the use of a generative, combinatorial phonology at innovation three, but did it really have to come after innovation one (using symbols) and number two (using words)?

The evolution key requires that the combinatorial elements came from somewhere. Sounds or signs that existed for non-linguistic reasons could have been broken down into parts or built up into words. If language began with signs, a breakdown process seems more likely. Gesture was already millions of years old, so perhaps early sign words were broken down and by combining into something new. With this scenario, Jackendoff’s innovations 1, 2, and 3 all seem to occur at once, giving us a symbol, a word, and a combinatorial structure all at once.

If sounds were the first words, the opposite process strikes me as more likely. Today we see infants vocalizing, developing syllables, and then combining them into words. If language began that way, there was a period of vocalization preceding speech and then at some point the sounds took on new duty as words. In this scenario, innovation three came before the first two on Jackendoff’s list.

The mystery: What makes language so different?

Many answers have been proposed, including the use of symbols, a learnt vocabulary, combinatorial phonology, recursive phrase structure, or Jackendoff’s full list of innovations. They all sound reasonable and picking one might seem arbitrary. Fortunately the theory of evolution provides a key: species don’t evolve complex novelties unless they provide some functional adaptation to the environment,

With that rule in mind we can omit most of these novelties. A learnt vocabulary, for example, is obviously necessary for language, but learning the vocabulary is not the function of language. The same objection stands for more technical sounding novelties like recursive phrase structure. What is its function? I often read that without recursion we would not be able to generate an infinite list of sentences, but that result cannot be why we evolved recursion. No purposes allowed.

Most accounts of language gloss over the issue of functional novelty because they use the word communication to describe the function of both animal cries and language. I have contended many times that animal communication is used to control or manipulate another animal. (For the most recent discussion of that point see The Laws of Evolving Animal Communications) Is that the function of human communication too? The second key, the speech triangle, helps out here by defining the relationships that evolved.

So what is the function of this relationship? The speech triangle allows consideration of a topic. We consider topics all the time; no other species considers them even a little. This function organizes those other novelties: we have large vocabularies because we can converse on many topics; we use recursion to get at a topic’s details; language is flexible because it has to be adaptable to any topic, etc.

The enigma: Why do only people speak?

Explanations for this one are particularly unsatisfactory. Jackendoff’s list would seem to suggest that a number of lucky innovations were required, but we don’t see any species that got part way up the list. Why not? Chomsky argues that we got a single mutation that enabled us to think recursively. Okay, but why did no other species ever get that mutation as well? These explanations boil down to saying that humans alone speak because humans alone speak.

The speech triangle key offers a clue. Speakers need to trust one another before they will risk sharing information, but trust is rare in the animal kingdom so the opportunity to evolve language is also rare. But rare is not the same as unknown, so this key does not solve the whole problem of why humans alone talk.

Turning to the other key, evolution, we notice something else. Evolution is chiefly concerned with what anthropologist Chris Knight calls the brute facts of existence: surviving predators, finding food, and reproducing. But the function of language is much broader than that. Most of the topics considered are about softer facts of existence. It is true that ultimately many of these softer facts point to ways of engaging the brute facts, but they require a flanking approach. All animals need food, but we seem to be the only mammals that come at it indirectly by learning about agriculture. It seems to require a special imagination to see the relevance between some neutral topic and a more pressing need.

That observation leads us to the conclusion that only humans are imaginative enough to benefit from considering neutral topics, but that comment just leads us back to the problem: why are we the only ones who are so smart?

A solution comes from a paper published by Karin Isler and Carel van Schank,  “Why are there so few smart mammals (but so many smart birds)?” (abstract here) . They report a “gray ceiling,” a point where the expense of adding any more brain matter forces a trade off that puts the whole species at risk of extinction. But they found that the ceiling can be penetrated if a species becomes cooperative.

This is the point where the riddle, the mystery, and the enigma come together. To develop language, a species has to have some kind of social output (vocalization, gesture, etc.) that can be transformed into the combinatorial elements that support expressive flexibility. Then they have to develop the trust and cooperation that would make sharing information acceptable. And finally they have to be imaginative enough to see the benefit of considering topics only indirectly related to the brute facts of existence.

There cannot be many lineages that went through all those hoops. Interestingly, only 50 thousand years ago there were several species of talkers. Somehow we have outlasted them all, but that’s another story.

Comments

[gregorylent]

you might benefit from a bit of study of sanskrit ...

guys have been thinking about your thoughts about for centuries, and relating them to consciousness, reality, being, vibration, etc ... huge literature on this, a bit smarter than western efforts, imo

in sanskrit, start with the word vak
--------------------
BLOGGER: Yes, Sanskrit is important. I've even thought of studying it, more or less for the heck of it, but it still remains a few tasks down on my todo list.

[J. Goard]

That observation leads us to the conclusion that only humans are imaginative enough to benefit from considering neutral topics, but that comment just leads us back to the problem: why are we the only ones who are so smart?

Are you familiar with Frost's poem "Out- Out-"? Basically, a kid is gazing at the mountains, imagining remote stuff, while cutting firewood with a big bandsaw. He gets his hand cut off and dies. The final lines describe the other workers:

No more to build on there. And they, since they
Were not the one dead, turned to their afairs.

The "gray ceiling" hypothesis focuses on the costs of growing and maintaining thinking power, but we should also consider the potential costs of thinking too much when not adequately constrained.

It's easy to look at ourselves now and bask in the glory of our speech triangles, with their diverse topics often far removed from the Here and Now and Me. But add a little bit of that to most animals in their natural environments, and it's likely to get them killed off as swiftly as Frost's young dreamer. If that kid was ever gonna become a professional poet, he was gonna have to survive enough "buzzsaw" situations to earn enough for college, while also maintaining and nurturing his reverie. In the early stages of evolving symbolic communication, there may very well be such a deep and long trough.

[J. Goard]

I should add that this basically reverses the question. Why should hominids have survived the early stages of distraction from the self and the immediate context?
-----------------------------
BLOGGER: Although I doubt that early talkers did much philosophizing, they wouldn't have survived if they hadn't already been a cooperative community able to use the input from an extra pair of eyes,

[Ben Sano]

Childhood, a period between weaning and the competence to live by your own efforts, seems to have been invented by Homo habilis. This seems to be the beginning of group trust. How did it happen and why?
-----------------------------------
BLOGGER: It does look like the Homo genus was distinguished, from the get go, by cooperation. The enlarged brain, Oldowan tool kit, and sped up fertility rates all point to it. I've discussed Homo habilis on this point a few times. See: Symbolic Thinking Began 2 Million Years Ago, The Persistent Burden, Scylla and Charybdis.

[J. Goard]

An animal distancing its attention from the immediate context is just the beginning which might eventually lead to philosophizing. Early humans wouldn't have forced everybody around to look at the beautiful sunset? How about arguing for an hour over who threw a stone the farthest yesterday?