by Chip Walter
The bewildering mystery of how this happens is what Douglas Hofstadter hoped to resolve when he penned his landmark book Gödel, Escher, Bach. He wanted to figure out, he wrote, “What is a self, and how can a self come out of stuff that is as selfless as a stone or a puddle? What is an ‘I’?” It’s an easy question to ask, and we’ve been asking it for as long as we have been around. The answer, though, is just a bit tougher to come by than the asking.
But we can try.
You may have noticed from time to time that when you think, you find that you are talking to yourself; not necessarily out loud, but in your mind. Nearly every waking moment we describe what is going on in our minds to ourselves, like a sports announcer calling a game, remarking on what we see, commenting on our own insights, planning our lives, polling ourselves on what we feel, wondering why this and how that? “God, I’m edgy this morning. This coffee is delicious! Hmmm, rain, better grab the umbrella. That’s an interesting choice in a hat, if you are insane! Don’t forget to get the oil changed and pick up the milk. You know you really have to get better at remembering people’s names.” Our reflections run the gamut from the mundane to the ethereal, occasionally the sublime, but they almost never stop, until we fall asleep.
When you think about how you think about yourself, you are experiencing what psychologists call metaconsciousness, the ability to be aware that you are aware. Though we take it for granted, this capability requires language, and the interesting thing about that is that language requires using symbols that we sound out in our minds so we can understand what we are saying to ourselves. That we can do this is remarkable enough, but doesn’t conversing as we do with ourselves make you wonder, if you are doing the talking, then whom are you talking to? Or, if you are listening, then who, precisely, is talking? Are we one person or two? Or many? Where does that voice we call “thought” come from? Who is the voice in your head, and how did it get there?
In the 1970s the Princeton psychologist and philosopher Julian Jaynes wrote a fascinating, bestselling book with the rather opaque title The Origin of Consciousness in the Breakdown of the Bicameral Mind. Jaynes’s insights as a philosopher and a psychologist remain respected (he passed away in 1997), but the book was highly controversial. He speculated that consciousness of the kind I just described is an extremely recent evolutionary development. Between 10,000 B.C. and 1000 B.C., he argued, modern humans thought that the voice they heard in their head wasn’t their own, but the voice of a chieftain or demon or god, some very real being outside their own minds. In other words, they didn’t talk to themselves the way we do, they instead believed they were listening to another all–knowing being who was observing them and their thoughts. Jaynes called this kind of mind bicameral, or two–chambered, one chamber that listened and one that spoke, but neither of which was aware that they were part of the same brain. “[For bicameral humans], volition came as a voice that was in the nature of a neurological command,” he wrote, “in which the command and the action were not separated, in which to hear was to obey.” As evidence he points to the statues and idols that ancient cultures in Egypt, Sumeria, and Mesoamerica created as the physical symbols of the gods and chiefs that spoke to bicameral humans; the voices. This is the only possible explanation, he argued, for why they were built, and, having been built, why they exerted the enormous influence they clearly did over their cultures.
Jaynes also maintained that the people who lived in these ancient societies did not in any way believe others truly died. They only moved on to another world and then, having arrived, spoke from that world directly to those left behind. And from that world the gods also spoke, commanding the creation of great temples and the elaborate rituals for their benefit because, after all, they were running the show. The first laws, Jaynes explains, such as Hammurabi’s Code and the Ten Commandments, were, as Hammurabi and Moses both said, rules passed directly to them by God.
Is Jaynes right? Were we once incapable of thinking for ourselves? Or more precisely, was there a time in our evolution when we were unaware that we were thinking for ourselves? It’s impossible to know, absolutely, what was going on in the minds of humans living in ancient Sumeria, Egypt, or the Yucatán thousands of years ago, but we do know that the human brain, even today, sometimes struggles to identify the speaker within us as our “self.” Schizophrenics hear a voice, or sometimes multiple voices, that they do not recognize as belonging to them. They come from “others” speaking from the outside. Yet brain scan studies illustrate that the voices are, in truth, being generated inside their own heads.1
The experience of schizophrenics, and Jaynes’s theory, raises the question that somehow the human brain came up with a trick that helped it talk to, and control, itself. At some point it found a way to transform those external voices into internal ones. If that’s true, though, then how did we manage it?
The answer begins with our brain’s unique ability to create symbols and weave them together in outrageously complex ways. Other animals can’t invoke symbols, but they can associate a single symbol or event with an experience they hold in their brain. Your dog, Fido, for example, may recognize the sound (but not the meaning) of the word walk and associate it with something he likes to do at the end of a leash with you each evening, but that’s the extent of the connection between the two. The sound you make when you say “walk” brings to Fido’s mind a specific experience, and so when he hears you make that noise, he runs for the door and waits. Scientists call this an iconic relationship between an experience and its external representation.
If you move a little farther along the evolutionary chain, you will find that primates possess more sophisticated symbolic capabilities. Take the case of two remarkable chimpanzees, named Sherman and Austin, at the Language Research Center at Georgia State University. Both were trained to associate specific symbolic pictures, or lexigrams, with certain events. The lexigrams were imprinted on a series of buttons in the laboratory where they trained. When they pressed a lexigram, they might find themselves rewarded with a goody, like a banana, or banana juice. Pretty quickly the lexigram came to stand for the reward in the minds of Austin and Sherman, not unlike the way the sound walk became linked to a good time outside for Fido.
Once the two chimps figured out the one–to–one relationship between a specific image and its reward, the researchers decided to present them with a new challenge. This time they were required to use two buttons in combination, like a verb and a noun, to receive a treat. One kind of lexigram represented “give” or “deliver,” and the others represented a specific kind of food. So to receive a banana, Sherman and Austin had to hit the lexigram for “give” and then the correct one for “banana.” This took some doing because there were multiple combinations of foods and commands, or verbs and nouns. But after some intense training, the chimps got the hang of the new system.
However, researchers had still more in store for the two hard–working chimps. Once they had absorbed the rules of the two–icon system, they were next provided a different alphabet of rewards and verb symbols that they had to use to receive treats. The question now became, could Sherman and Austin transfer the command–reward system they had learned to entirely new lexigrams on their own? After some trial and error, they again rose intrepidly to the challenge and learned the new system.
The big insight here is that they had comprehended an underlying organizing principle that made it easier to master the new lexigram vocabulary, an ability called an indexical symbolic relationship, one in which an animal can transfer a particular way of thinking to different situations. It represents a huge leap from iconic thinking because iconic symbolic relationships merely require one–on–one memorization.
If chimpanzees can manage this today, it’s likely our direct ancestors going back millions of years could master something like it as well. So what makes us different from them? Our special ability is that we cannot only make iconic and indexical connections between meaning and experience like Fido and
Austin and Sherman, we can weave indexes of symbols into much more intricate systems of entirely new symbols, and we can do it in an almost infinite variety of ways.
For example, you not only see the letter e in the words you are reading on this page and associate a sound with that letter (an iconic relationship), but your mind effortlessly combines many e’s and other letters into words, each of which has greater meaning than the individual sounds of the letters. And then you can pile together the words into sentences that have greater meaning than any one word, and so on. You also simultaneously understand the context of the letters. An e in one place can represent one sound, in another it can be silent (in English, anyhow).
Words can also change their meaning depending on the context, just as letters do. The interrogative “Turn right, right here, right?” uses the same word in one sentence three times, yet each time it has a different meaning because its context is different. Your mind understands this because it also grasps the underlying rules of the English language. It grasps them even if it can’t entirely explain them.
Chimps may eventually and painstakingly be able to comprehend a simple system of language like “(You) throw ball”—subject, verb, and object. But there will never be the ghost of a chance that any chimp, even if he is the Shakespeare of Pan troglodytes, will comprehend the intricacies of the apparently simple question “Turn right, right here, right?”
One of the reasons he can’t is that all human language is recursive, which is another way of saying that it can embed concepts within concepts. Just as letters are embedded inside words, strings of words can be embedded within sentences to make them more meaningful. Take the sentence “John, devilishly handsome as he was, refused the title King of the Prom, even though he secretly believed he deserved it.” The ideas that John was devilishly handsome, that he secretly believed he deserved the title, and that the title was King of the Prom all invest the basic sentence “John refused the title” with much deeper meaning and a lot of useful information. They tell us about John, his motives, why he has the feelings he has, and provide some insight not only into how he looks, but how he looks at himself. Yet all of these additional nuggets of thought sit nicely nested, one inside the other. Unlike iconic and indexical meaning, this ability is richly symbolic, and unique to our kind of brain.
Language isn’t the only symbolic system we have created that taps our special ability to recursively weave bundles of symbols into mosaics of meaning. We do it in mathematics with numbers and variables that can be built into proofs and theorems and formulas. In music we stack notes elaborately to construct melodies, then fashion melodies into themes and songs and symphonies, and even thread in harmonies with still other melodies, and finally, for good measure, add words to the melodies to create everything from pop hits to operas and Broadway shows.
Paintings are collaborations of symbols, too. Different colors of paint are situated together to represent real objects or feelings or ideas so that together they fashion works that have broader meaning than each of the dabs or drops of color themselves. Georges Seurat’s pointillist paintings are a perfect example—hundreds of thousands of individual dots of different colors that together bring an image alive. Without our ability to gather together an alphabet of symbols and connect them in elaborate, nested patterns, there would be no Hamlet or Faust or Moby–Dick, no laws of thermodynamics, no science, music, architecture, no Kabuki theater, sculpture, Renaissance art, or anything else that has made the great, expansive construction project that we call human culture possible. Every iota of it is built on the unique and powerful talent of your Homo sapiens brain to mysteriously direct the molecular machinations of its own neurons to manufacture symbols and then share them with the other symbol–recognizing creatures around you. This enables minds to meld, hearts to bond, and ideas to be shared and bent and shaped by many other minds. And the brain does this without really comprehending how it does it, something like the way a basketball player drives to a basket and deftly deposits the ball into it without a moment’s reflection.
We are able to embed symbols within symbols this way, and create intricate and outrageously complex thought structures, because our brain can take a concept, idea, or goal and set it aside temporarily while we shift our attention and work on something else. Scientists use two symbols to describe this earth–shattering capability: working memory.
In its simplest form, working memory is something like taking a call on your cell phone, starting a conversation, and then asking the caller to hang on while you take a second call. You can then begin that conversation without losing sight that you have another conversation–in–waiting because you have filed away the first phone call as a symbol, a kind of “object” the brain can retrieve from a folder or drawer. It’s almost as though it is a physical thing. This holds true for nearly anything we can think or imagine, from goals to concepts to worries.
What’s more, these “objects” we put on hold can have multiple concepts that live within them. We don’t have to remember each of the individual pieces of information that reside within what we have set aside, we just have to be able to recall the big idea. And when we do, everything else comes along for the ride. If you are envisioning the Taj Mahal, you don’t have to log and file away every detail while you shift your attention to preparing lunch. You simply prepare your meal, then reach back into your mind and pluck up the concept “Taj Mahal,” and all of your thinking related to it returns like a nicely nested matryoshka doll, a file folder of the mind.
It’s not clear where, precisely, this talent lies in our brain. Like nearly everything else cerebral, it almost certainly doesn’t reside in one place. The brain, like recursion itself, is nested and networked, woven together. But fMRI studies have found that humans and other primates have regions called the frontal operculum, which activates when they process indexical kinds of information, but only humans have the much more recently evolved Broca’s area, which handles language and grammar and syntax, recursive symbolic systems.
Broca’s area is part of the human prefrontal cortex, a sector of the brain that sits directly behind our foreheads. It is one of the reasons we find ourselves with the large, childlike heads we have, and why our foreheads don’t slope back as much as our cousin primates. The human prefrontal cortex (PFC) has, in evolutionary terms, sprinted toward its present state compared with other advances in brainware. While our brains have tripled in size over the past six to seven million years, our PFC has increased sixfold. When it comes to symbolic thinking, this is where the action is.
The action is here because the PFC has evolved as the brain’s chief executive officer. It polices, as much as they can be policed, the primal, impulsive activities of our minds. The PFC inhibits anger, fear, hunger, sexual attraction, and other strong, but ancient drives. Many of these capabilities emerged as our ancestors became increasingly social and more reliant on one another for survival. Evolution would have favored individuals with brains that were better at controlling purely selfish impulses and favored those that took a longer–term view of situations. When you live in groups, after all, it may not pay to act solely in your short–term self–interest since you may need the help of others in the future. So today perhaps you share your food so that another day food may be shared with you.
Not only does the prefrontal cortex act as an executive this way, it allows us to think ahead by taking symbolized ideas, concepts, and memories and cobbling them together into scenarios that are completely nonexistent except in the brain itself, or, put another way, imagined. By recalling information held in long–term memory, packaging that with new information, setting these newly organized symbols aside in working memory so they can percolate while we move forward with other goals and ideas, we advance through the day, prioritizing, organizing, imagining, worrying, creating. Sometimes the work is mundane, like figuring out how to get showered, make phone calls, answer e–mail, and catch the subway in more or less the correct order so we don’t show up at an
appointment late, unbathed and misinformed. Sometimes the work is profound and results in the special theory of relativity. You never know.
It doesn’t take much to imagine that the special abilities of the prefrontal cortex, whenever they finally came together, made our species dramatically different, outrageous really. Able to represent our thinking symbolically, then to embed these symbols inside one another, we became beings able to efficiently create, organize, and recall enormous amounts of complex information for still more revision.
In many ways it produced in us the same sort of ability that digital compression algorithms make possible. JPEG images and MP3 sound files, which make showing off the family photos on your iPad or listening to your favorite music on your phone possible, are the results of compression algorithms. What makes them useful is that they are not perfect reproductions of the images and sounds they represent, but the formulas that create the algorithms excel at extracting just enough of the right information to re–create a close facsimile that requires far less information and memory than the original. The copy is similar enough that most of us can’t tell the difference, yet it’s far less information intensive, and therefore more efficient. Symbols do the same thing. We don’t remember everything. Just what we need to remember.
As helpful as all of this symbolization is in the business of expressing what you feel and think to others, it also makes it possible for you to do something else rather amazing: explain what you feel and think to yourself. In fact, it makes your “self” possible, and that may be the most stunning illusion our brains have pulled off yet over the past fifty thousand years—maybe the most stunning ever. The ability to create the ultimate symbol: you. Which takes us back to the question we asked at the beginning of this chapter: Who are “you,” anyway?