Harnessed: How Language and Music Mimicked Nature and Transformed Ape to Man (3 page)

Why should we believe what we can’t consciously perceive—that language and music mimic nature at all but the highest hierarchical level? Why not go all the way and make language and music
completely
like nature?

Let’s not forget that language and music are not
merely
trying to mimic nature. They have
jobs
to do: writing is for putting thoughts on the record, speech is for transmitting thoughts to others, and music is perhaps for something like evoking feelings in others. Language and music want to capture as much of the structure of nature as they can so that they have an easy ride into our brains, but they must serve their purpose, and will have to sacrifice nature-mimicry when it is necessary to do so.

So one can see how sacrifices of nature-mimicry may sometimes be part of doing business. But why should the sacrifices be up near the top, where we have greater conscious access? The principal reason for this is that if the earlier regions of the hierarchy receive stimuli that
they
can’t make any sense of, then they will output garbage to the next higher level, and so all levels above the unhappy level will be unhappy. Breaking nature-mimicry at one level will break it at all higher levels

For example, I have argued in earlier research and in
The Vision Revolution
that writing looks like nature. In particular, I have suggested that written words look like visual objects. But words do not necessarily look natural at
all
levels up the hierarchy. Strokes look like contours, and letters look like object junctions; and thus the lower and middle levels of your visual hierarchy are happy. But because in alphabetic writing systems the letters in a word depend on how it is spoken, there is no effective way to make entire words look like objects. (For example, the junction-like letters in the words you are currently reading are simply placed side by side, which is not the way junctions in scenes are spatially related.) Your highest-level regions, of which you are most directly aware, only notice the nonnatural look of written words. And when visual signs do more closely match the visual structure of objects at the highest levels, people
do
see the resemblance to nature—this is why trademark logos and logographic writing systems like Chinese look (to your conscious self) much more object-like than the words you’re reading here.

My claim in this book that language and music mimic nature must be understood in this light. I claim that they mimic nature, indeed, but not necessarily “all the way up.” The reason why writing, speech, and music don’t
obviously
seem like nature is that nature is
not
being injected at the higher levels, perhaps—as we’ve seen with writing—in order to better accomplish the functions they are designed to carry out.

We see, then, why it is that the nature-mimicry in language and music has remained a secret for so many millennia. If only your lower-level visual and auditory areas could speak! They’d have long ago let you know that language and music are built like nature. Because those lower homunculi are
part
of you, there is a sense in which you have known about this ancient, deep secret code all along. Pieces of meat inside you knew the secret, but weren’t telling. In this light, one can view this book as a kind of psychoanalysis—if you’re into that—digging up the homunculus-knowledge you already have deep inside you, and working through the ways it shaped who you are today.

Nature’s Hard Core

“Language and music mimic nature.”

I
will
try to convince you of that over the course of this book. But we humans are very bad judges of whether a stimulus is natural or not, as we just finished discussing. How, then, are we to have any idea whether language and music do or do not mimic nature? And isn’t the phrase “mimicking nature” awfully imprecise?

Indeed, it
is
imprecise. Part of the book’s point is to make this more precise—to say
specifically
which
aspects of nature are mimicked by language and music. But can this really be done? How can we possibly know the details of the natural world our ancestors experienced? It is, after all,
that
version of nature that our brains evolved to be good at processing, and that was a long time ago. In order to show that language and music mimic the primordial natural habitats that shaped our brains, it would seem that we have no choice but to start familiarizing ourselves with the sights and sounds of the savanna, and wherever else our ancestors hung out.

A field trip is in order. Like all natural habitats, savannas feature a hodgepodge of characters and settings, and we must take a visual and auditory inventory. We will need images from all over the savanna, including acacia trees, tall grass, sunsets, rocks, giraffes, lions’ manes, and termite mounds. And we will need to record the sounds within those habitats, including wind, rustling leaves, bird calls, insect buzzes, and rhino grunts. Once we have built an encyclopedia of how our ancestral world looked and sounded, we can
then
ask whether language and music mimic nature.

Hold it right there! Sorry, but I’m pulling your leg. That is definitely
not
what we’re going to do. Trekking across the Serengeti with a camera and tape recorder sounds grand, but it isn’t the route to a compelling scientific explanation. We want
generalizations
about what visual nature looked and sounded like for our ancestors, not merely an itemized list of all the furniture in the savanna. We need to grasp the fundamental “grammar” of nature. We need to pick apart nature, carve it at its joints, and elegantly summarize its structure. With such a “grammar” in hand, we will be able to make powerful predictions about what a nature-mimicker should be like . . . and thus, what language and music should be like.

Might there be fundamental regularities that cut across a wide swath of terrestrial habitats? Could it be that, although there are large salient differences across habitats, there are nevertheless deeper respects in which they are all similar? Although the savanna and any other specific habitats shaping our ancestors have hosts of peculiar features, might the sights and sounds of these habitats nevertheless have the same fundamental “core grammar”?

If there
is
a “core grammar” to nature, then this core would have been a highly steady invariant over our evolutionary history, and would thus have been a strong shaper of our visual and auditory systems. Sensory structure specific to the savanna or other particular habitats would, on the other hand, have been highly variable and intermittent over evolutionary time, and consequently less important for understanding what our visual and auditory systems became good at.

It’s this hard core of nature that we want. But is there one? Yes, indeed. There
are
solid core grammars underlying the structure of visual and auditory nature—there are “universals” in the structure of nature. That’s what I’ll endeavor to show you in this book; and
then
I’ll show you that language and music mimic these cores. I’ll give you a preview of these hard cores in the next section, when I introduce the central tenets of the theory—that is, when I first reveal the big secrets.

The Reveal

At this point, we have discussed the possibility that language’s and music’s mimicry of nature could be what enabled us humans to acquire language and music. We also took up why, if this were so, it would not be obvious to us. And in the section just above, I made myself clearer about the role that “nature” will play in understanding the origins of language and music: the goal is to find fundamental principles underlying the structure of nature (as opposed to a catalog of savanna paraphernalia), so that we’ll be in a strong position to ask whether these fundamental principles underlie language and music.

What I have
not
yet done is give you any specifics on what the hard cores in nature
are
that are being mimicked by language and music. That is, I haven’t revealed what any of the “ancient secrets” might actually be. Let’s rectify that, and simultaneously summarize the book’s two main theses, concerning speech and music.

Chapter 2 of the book is about the secret code underlying speech. Here’s the secret:
human speech sounds like solid-object physical events
. Notice that this secret makes no mention of rustling leaves or rhino grunts. Instead, the “nature” that speech mimics encompasses a very broad class of events, namely those among solid objects. The main observation is that events involving solid objects bumping and crashing about have a signature core auditory structure, and I will provide evidence that human speech has this signature structure. Speech can thereby get into our brains by harnessing auditory-recognition mechanisms we have long possessed for processing the “pinball” sounds of nature. This secret code gives us hominids the power to recognize speech without having speech-recognition mechanisms.

And here is the deep secret underlying music, which is the topic of Chapters 3, 4, and 6:
music sounds like humans moving and behaving (usually expressively)
. Notice how general the notion of “nature” is here. It isn’t the sounds of people’s heartbeats, or heavy breathing, or missionary-style sex, or skipping—it is the sounds of humans behaving. When we carry out behaviors we tend to make noise, and our auditory system can infer a lot about each other’s behavior from the noise. Music has the signature auditory structure of humans doing stuff, and can thus get into our brains by tapping into our auditory recognition mechanisms for identifying the actions of other people. No music-processing mechanisms are required. This secret code of sounding like expressive human movers is what allows music to flow into our auditory system and be understood by it, even though we possess no special brain gears for processing music.

Although it is the ancient secret codes lying beneath speech and music that I’ll be whispering about in this book, this is not the first time I have written about deep secrets of this kind. In my previous book,
The Vision Revolution
, I wrote about (among other topics) another ancient secret, the one explaining how we hominids came to have a
written
history. And the secret is this:
writing looks like three-dimensional scenes with opaque objects
. Writing looks like nature, but as before, nature of a very general kind—no images of acacia trees or termite mounds are needed. Writing has come to mimic the contour combinations occurring in natural scenes with opaque objects, and in such a way that written words mimic the structure of visual objects. Writing gets into our brains by harnessing our visual object-recognition mechanisms. The secret code of looking like nature is what allows writing to be read by us hominids without any reading mechanisms in our brains.

The “nature” stories of the origins of speech, music, and writing are, then, not in the least about acacia trees or the other particulars in the rummage shop of our ancestors, but rather about solid-object physical events, human movement sounds, and opaque objects in a three-dimensional world. “Nature” is a highly general notion, just what is needed to make theoretical headway and empirical testing possible and practical.

And although the notions of “nature” I will rely on are very general, they are not so general that they include everything. For example, “solid-object physical events” covers a wide swath, but it doesn’t cover sounds made by air and water. And “opaque objects in a three-dimensional environment” is fundamental, but a habitat with semitransparent objects (like clouds at high altitudes) would not be included.

Distinguishing between the surface features of habitats (which vary wildly from habitat to habitat) and the core features (found in most or all habitats) is helpful in understanding why, even if writing, speech, and music have underlying core similarities, they nevertheless come in such tremendous variety. If nature were all core—if it had little or no variability across habitats—then our visual and auditory systems would have evolved to be competent at processing just the very specific kinds of stimuli in the world. Language and music that harnessed such a brain would be expected to have a very specific and consistent surface structure, something they do not, in fact, have. If, instead, as is the case, there is a small core of invariant structure to nature, yet loads of variability across habitats, one would expect us to end up with brains that are more open-minded about what they’re willing to accept. Our brains would be most competent at processing stimuli that have the core signature, but in other respects, our brains would be open to many variants. Language and music that harnessed this kind of open-minded brain would be expected to take widely varying shapes across cultures, but to share certain similarities. This is a much more accurate description of language and music as found on Earth: subject to large differences across cultures, but sharing certain core structural characteristics across cultures.

You have now gotten a peek at the ancient secret codes hidden inside speech and music. Hopefully you can appreciate their generality, and appreciate why it might be that the natural structure in speech and music has stayed hidden from us. In the next and final section of this chapter, I will be as clear as I can about how my nature-harnessing theory differs from other stances on the origins of language and music, and I will justify why we should
expect
language and music to have nature instincts (i.e., designed to mimic nature) rather than just brain instincts (i.e., designed to be well shaped for the brain).

Purps vs. Quirks

In the Introduction, I touched upon two standard, contrasting viewpoints on origins, the first being that we evolved brains specialized for language and music (i.e., we have instincts for these things), and the second that, on the contrary, we evolved to be general-purpose, universal learning machines that handle these artifacts because we can learn
lots
of unusual stuff. I suggested that language and music seem unlikely to be instincts because writing, too, reeks of instinct, but is definitely not an instinct. But I also intimated that there is a wealth of data and argument—summarized and argued convincingly in Pinker’s books, for example—that we do not possess blank-slate brains. How, then, are
predisposed
brains like ours able to learn any human language and comprehend music—among the most complex and sophisticated computational tasks on Earth—if we’re neither designed specifically for it nor particularly impressive general learners?

The answer is that once culture got up and running, there was a new blind watchmaker in town. Cultural evolution could, over comparatively short periods of time, fit language and music into the shapes our grooved (non-blank-slate) brains are able to learn. It is not so much that our brains learn language and music, but rather that culture learned how to package language and music so that they fit right into our brains. Culture learned how to harness us.

Despite the title of this book, there is nothing new about the idea that we are harnessed by culture, that cultural artifacts may have been selected to be structured well for our brains. What
is
new here is that I am putting forth specific proposals for how culture actually goes about harnessing us. Saying that language and music might be shaped for the brain doesn’t take us very far in understanding the shape of language and music, because we don’t have a good understanding of the brain. What we need is a
general
theory
of harnessing. And
nature-harnessing
is the theory I am proposing.

Earlier I said that language and music have evolved to possess a brain instinct, rather than the brain having evolved to possess language and music instincts. But in a sense, in this book I am arguing that language and music have, not a brain instinct, but a
nature
instinct. Language and music carry in them the structure not of the brain so much as of nature, which of course is just right for the brain—because the brain is just right for nature (see Figure 1).