Communicating with the dead is a standard job requirement for a psychic such as the infamous medium John Edward of the television show Crossing Over who claims to be able to listen to what the deceased family members of his studio audience have to say.  Hearing the thoughts of the dead would appear to be one superpower we certainly do not possess. Surely this superpower must remain firmly in the realm of fiction (Edward included). However, a little thought reveals that we in fact do this all the time. …by simply reading. With the invention of writing, the ability for the dead to speak to the living suddenly became real. (Progress in communicating in the other direction has been slower going.) For all you know, I’m dead, and you’re exercising your spirit-reading skills right now. Good for you!

Before the advent of writing, in order to have our thoughts live on after we had gone we had to invent a great story or catchy tune and hope that they’re singing it by the fire for generations. Only a few would be lucky enough to have a song or story with such legs (e.g., Homer’s Illiad), and at any rate, if our ancestors were anything like us, their greatest hits probably tended to include “ooh-la-la” and “my baby left me” much more often than “here’s my unsolicited advice” and “beware of milk-colored berries.” Getting your children to be your audio tape in this fashion is probably futile (and aren’t they just as likely to purposely say the opposite?), but at least it relies on spoken words, something readily understandable by future generations.

The problem is getting your voice to last. Voices are just too light and insubstantial, like a quarterback finding an open receiver and throwing to him a marshmallow. Marshmallows are great to hold, but impossible to throw far. I suppose if you were to speak loudly enough during a heavy volcanic ash storm, ripples on the rapidly accumulating layers of ash might record your spoken words, one day to be recovered by clever archaeologist decoders. However, much of what you’re likely to say in such circumstances will be unrepeatable in polite company.

What prehistoric people did successfully leave behind for us to read tended to be solid and sturdy, like Stonehenge or the moai statues of Easter Island. These were quarterback passes that got to the receiver all right, except that now the quarterback is throwing something that is uncatchable, like porcupines or anvils. Massive monuments are great if your goal is to impress the neighboring tribes or to brag to posterity. But if your goal is to actually say something that can be understood, this tact is worse than writing abstruse poetry, and literally much heavier. The only thing we’re sure of about such communications is that they had too much free time on their hands. Not the most informative spirit-reading.

The invention of writing changed spirit-reading forever. It also changed the world. Reading now pervades every aspect of our daily lives, so much so that one would be hard pressed to find a room in a modern house without words written somewhere inside. Lots of them. Many of us now read more sentences in a day than we listen to. And when we read we must process thousands of tiny shapes in a short period of time. A typical book may have more than 300,000 strokes, and many long novels will have well over one million. Not only are we highly competent readers, but our brains even appear to have regions devoted to recognizing words.  

Considering all this, a Martian just beginning to study us humans might be excused for concluding that we had evolved to read. But, of course, we haven’t. Reading and writing is a recent human invention, going back only several thousand years, and much more recently for many parts of the world. We are reading using the eyes and brains of our illiterate ancestors. And this brings us to a deep mystery: Why are we so good at such an unnatural act? We read as if we were designed to read, but we have not been designed to read. How did we come to have this super power?

Reading as a super power? Isn’t this, you might ask, a bit of an exaggeration?

No, it really is super. To better appreciate it, when you next have the illiterate caveman neighbors over to the house—the ones who always bring the delicious cave-made bunt cake—wow them with how you can transmit information between you and your spouse without speaking to one another. ...by writing and reading. They’ll certainly be impressed. It’s not your use of symbols that will impress them, however, because they leave symbols for one another all the time, like a shrunken head in front of the cave to mean the other is at the witchdoctor’s. And they have spoken language, after all, and realize that the sounds they utter are symbols.

What will amaze them about your parlor trick is how freakishly efficient you are at it. How did your spouse read out the words from the page so fast? Although they appreciate that there’s nothing spooky in principle about leaving marks on a page that mean something, and someone reading them later, they conclude that you are just way too good at it, and that, despite your protestations, you must be magical shamans of some kind. They also don’t fail to notice that your special power would work even if the writer was far away. Or long dead. Their hairs stand on end, the conversation becomes forced, they skip dessert, and you notice that their cavekids don’t come around to throw spears at your kids any more. As the saying goes, one generation’s maelstrom is a later generation’s hot tub.  We’re just too experienced with writing to appreciate how super it is, but not so for your cave neighbors. 

We have the super power of reading not because we evolved to read—and certainly not because we’re magical in any way—but because culture evolved writing to be good for the eye. Just as Captain Kirk’s technology was sometimes interpreted as magic by some of the galaxy locals, your neighbors are falsely giving you credit for the power when the real credit should go to the technology. The technology of writing. And not simply some new untested technology, but one that has been honed over many centuries, even millenia, by cultural evolution. Writing systems and visual signs tended to change over time, the better variants surviving, the worse ones being given up. The resultant technology we have today allows meanings to flow almost effortlessly off the page and straight into our minds. Instead of seeing a morass of squiggles we see the thoughts of the writer, almost as if he or she is whispering directly into our ears.

The special trick behind the technology is that human visual signs have evolved to look like nature. Why? Because that is what we have evolved over millions of years to be good at seeing. We are amazingly good at reading the words of the dead (and, of course, the living) not because we evolved to be spirit-readers. Rather, it is because we evolved for millions of years to be good at quickly visually processing nature, and culture has evolved to tap into this ability by making letters look like nature. Our power to quickly process thousands of tiny shapes on paper is our greatest power of all, changing our lives more than our other powers. Literacy is power, and it’s all because our eyes evolved to see well the natural shapes around us and we, in turn, put those shapes to paper.

Good Listening


“How did your date go?” I asked.

“Great. Wow. What a guy!” she replied. “He listened so attentively the entire dinner, just nodding and never interrupting, and—”

“Never interrupting?” I interjected.

“That’s right. So supportive and interested. And so in tune with me, always getting me without even needing to ask me questions, and his—”

“He asked you no questions?” I interrupted, both eyebrows now raised.

“Yes! That’s how close the emotional connection was!”


It struck me that any emotional connection she felt was a misreading of his eyes glazing over, because her date was clearly not listening. At least not to her! I didn’t mention to her that the big game was last night during her dinner, and I wondered whether her date might have been wearing a tiny ear phone.

Good listeners don’t just sit back and listen. Instead, they are dynamically engaged in the conversation. I’m a good listener in the fictional conversation above. I’m interrupting, but in ways that show I’m hearing what she’s saying. I am also able to get greater details of the story where I might need them. In this case about her date’s conversational style. That’s what good listeners do. They rewind the story if needed, or forward it to parts they haven’t heard, or ask for greater detail about parts. And good communicators tend to be those who are able to be interacted with while talking. If you bulldoze past all attempts by your listener to interrupt you, your listener will probably soon not be listening.

Perhaps he’s heard that part before and is tuning out now. Or perhaps he was confused by something you said fifteen minutes earlier, and gave up trying to make sense of what you’re saying. Good listeners require good communicators. My fictional friend above appears to be a good communicator because she dynamically reacts to my queries midstream. The problem lies in her date, not her, and I politely suggest he may not be the right one for her.


Even though we evolved to speak and listen, but didn’t evolve to read, there is a sense in which writing has allowed us to be much better listeners than speech ever did. That’s because readers can easily interact with the writer, no matter how non-present the writer may be. Readers can pause the communication, skim ahead, rewind back to something not understood, and delve deeper into certain parts. We listeners can, when reading, manipulate the speaker’s stream of communication far beyond what the speaker would let us get away with in conversation—“Sorry, can you repeat the part that started with, ‘The first day of my trip around the world began without incident’?”—making us super-listeners, and making the writer a super-communicator. 

We don’t always prefer reading to listening. For example, we listen to books on tape, lectures, and talk radio, and in each case the speakers are difficult to interrupt. However, even these cases help illustrate our preference for reading. Although people do sometimes listen to books on tape, they tend to be used when reading is not possible, like when driving. When one’s eyes are free, people prefer to read stories rather than hear them on tape, and the market for books on tape is miniscule compared to that for hard copy books.

We humans have brains that may have evolved to comprehend speech, and yet we prefer to listen with our eyes, despite our eyes not having been designed for this!

Television and movies have an audio stream that is not easily interruptable, and we do like that, but now the visual modality helps keep our attention. And although students have been listening for centuries to the speech of their professors, until recently with relatively little visual component, anyone who has sat through years of these lectures knows how often one’s mind wanders. …how often one is not actually listening. Talk radio has some popularity, and tends to be more engaging than traditional lectures, but notice that such shows go to great lengths to be conversational, typically having conversations with callers, and often having a pair of hosts (or a sidekick), to elicit the helpful interruptions found in good listening. 

Canned speech, then, tends to be difficult to listen to, whereas genuine, dynamic, interactive conversation enables good listening. There is, however, one kind of audio stream our brains can’t get enough of, where interruption is not needed for good listening, and where we’re quite happy not seeing anything. Music.

Audio tapes that give up on communication and aim only for aesthetics are suddenly easy listening. The rarity of books on tape, and the difficulty with listening to canned speech more generally, is not due to some intrinsic difficulty with hearing per se. The problem is that speech requires comprehension— music doesn't — and comprehension can occur most easily when the listener is able to grab the conversation by the scruff of the neck and manipulate it as needed so he can fit it into his head. Good conversation with the speaker can go a long way toward this, but even better listening can be achieved by reading because then you can literally pick up the communication with your hands and interact with it to your heart’s content.

Working Hands


Having a conversation is not like passing notes in class. Although in each case two people are communicating back and forth in turn, when passing notes you tend to do little reading and lots of wiggling—either wiggling your hand in the act of writing a note, or twiddling your thumbs while waiting for your friend to write his. Note-writing takes time, so much time that passing notes back and forth is dominated by the writing, interspersed with short bouts of reading. All the work’s in the writing, not the reading. Conversation—i.e., people speaking to one another—is totally unlike this. Speaking flows out of us effortlessly, and comes out nearly at the speed of our internal thoughts. That is, whereas writing is much more difficult than reading, speaking is not much more difficult than listening.

The reason for this has to do with how many people we’re communicating with.  When we speak there are typically only a small number of people listening, and most often there’s just one person listening (and often less than that when I speak in my household). For this reason spoken language has evolved to be a compromise between the mouth and ear: somewhat easy for the speaker to utter, and somewhat easy for the listener to hear. In contrast, a single writer can have arbitrarily many readers, or “visual listeners.” If cultural evolution has shaped writing to minimize the overall efforts of the community, then it is the readers’ efforts that will drive the evolution of writing because there are so many of them. That’s why as amazing, as writing may be, it is a gift to the eye more than a gift to the hand. For example, a book may take six months to write, but it may take only six hours to read. That’s a good solution because there are usually many readers of any given book. 

Is writing really for the eye, at the expense of the hands?

One of the strongest arguments that this is the case is that writing has been culturally selected to look like nature, something we’ll see later. That’s a good thing for the eye, not the hand, because the eye has evolved to see nature—the hand has not evolved to draw it. Not only does writing tend to look like nature, but I have found that even visual symbols like trademark logos—which are typically never written by hand, and are selected to be easy on the eyes—have the fundamental structural shapes found in nature. And note that for some decades now much of human writing has not been done by hand, but instead has been done by keyboard.

If the structures of letters were for the hand, we might expect that now that our hands tend to be out of the picture, the structures of letters might change somewhat. However, although there are now hundreds of varying fonts available on computers, the fundamental structural shapes have stayed the same.

Shorthands, however, have been explicitly selected for the hand at the expense of the eye, and shorthands look radically different from normal writing, and I have shown that they have shapes that are not like nature. I have also taken data from children’s scribbles and shown how the fundamental structures occurring in scribbles are unlike that found in writing and in nature. Finally, one can estimate how easy a letter is to write by the number of distinct hand sweeps required to produce it (this counts sweeps resulting in strokes on the page, and also sweeps of the hand between touchings of the paper), and such estimates of “motor ease” do not help to explain the kinds of shapes we find in writing. 

Could culture really have given no thought whatsoever to the tribulations of the hand? Although selection would have favored the eye, it clearly would have done the eye no good to have writing be so difficult that no hands were willing to make the effort. Surely the hand must have been thrown a bone, and it probably was. The strokes in the letters you’re reading, and in line drawings more generally, are quite a bit like contours in being thin, but there is an important difference.

Real world contours occur when one surface stops and another starts, like the edge between two walls, or the edge of your table. Usually there is no line or stroke at all (although sometimes there can be), but a sudden change in the nature of the color or texture from one region to the next. The visual system would therefore probably prefer contours, not strokes. But strokes are still fairly easy to see by the visual system, and are much easier for the hand to produce. After all, to draw true contours rather than strokes would require drawing one color or texture on one side of the border of the intended contour and another color or texture on the other side.

I just tried to use my pen to create a vertical contour by coloring lightly to the right and more darkly to the left, but after a dozen tries I’ve given up. I won’t even bother trying to do this for an “S”! It’s just too hard, which is why when we try to draw realistic scenes we often start with lines as contours, and only later add color and texture between the lines. And that’s probably why writing tends to use strokes. That we use strokes and not true contours is for the benefit of the hand, but the shapes of our symbols are for the eye. 

And we're going to learn about what that means in the next installment, Harness The Wild Eye.

This is a modified excerpt from my book, The Vision Revolution.

Further reading:

Changizi MA (2009) THE VISION REVOLUTION: How the Latest Research Overturns Everything We Thought We Knew About Human Vision (BenBella Books, Dallas TX).

Changizi MA, Zhang Q, Ye H&Shimojo S (2006) The structures of letters and symbols throughout human history are selected to match those found in objects in natural scenes. The American Naturalist 167: E117-E139

Changizi MA&Shimojo S (2005) Character complexity and redundancy in writing systems over human history. Proc Roy Soc Lond B 272: 267-275