How Did Reading and Writing Evolve?

Derek Hodgson writes for The Conversation about his theories on the development of writing:

The part of the brain that processes visual information, the visual cortex, evolved over the course of millions of years in a world where reading and writing didn’t exist. So it’s long been a mystery how these skills could appear some 5,000 years ago, with our brains suddenly acquiring the specific ability to make sense of letters. Some researchers believe that the key to understanding this transition is determining how and why humans first began to make repetitive marks.

Recent extensive brain imaging of the visual cortex as people read text has provided important insights into how the brain perceives simple patterns. In my new paper, published in the Journal of Archaeological Science Reports, I analyse such research to argue that the earliest human-made patterns were aesthetic rather than symbolic, and describe what that means for the evolution of reading and writing. […]

In 2000 I first suggested that the way the “early visual cortex” – the location where visual information from the eye first impacts the cortex – processes information gave rise to the ability to engrave simple patterns. We know that this area has neurons coding for edges, lines and “T” junctions. As distilled forms, these shapes preferentially activate the visual cortex. […]

At some point from around 700,000 years ago, this sensitivity to geometry and pattern perception enabled humans to start making refined “Acheulean tools”, which exhibit a certain symmetry. […] The tool making then further promoted an enhanced sensitivity and bias towards patterns in the natural environment, which our ancestors projected onto materials other than the actual tools. For example, they started accidentally making marks on rocks, shells and materials such as ochre. At some point, these unintentional patterns were intentionally copied on such materials – developing into engraved designs and later on into writing.

But how was this possible? Neuroscientific research has shown that writing text involves the premotor cortex of the brain, which drives manual skills. My theory therefore suggests that reading and writing evolved when our passive perception for discerning things started to interact with manual dexterity.

It’s interesting stuff, but I have no way of judging its plausibility, and I invite your thoughts on the matter.


  1. As far as I can tell from the blog and the linked article, he’s saying that the neural skill which first enabled humans to discern and produce certain geometric patterns wasn’t available before some hundreds of thousands of years ago. The history of writing is a small sliver of that time, and of literate societies an even smaller one. From the point of view of the history of writing, I don’t see a significant advance here.

  2. David Eddyshaw says

    It seems to me to be a non-question, in every way analogous to “How did humanity evolve the capacity to ride a bicycle?”

    “Our brains suddenly acquiring the specific ability to make sense of letters” is just as valid a statement of the relevant facts as “Mysteriously, two centuries ago, our brains suddenly acquired the specific ability to control bicycles without falling off.”

    It’s only the much greater cultural importance of literacy than velomania that prevents this from being obvious.

    The neural pathways were obviously there all along, and there for quite other reasons: if they had been different, humanity would have developed a different way of recording language that depended on whatever those different pathways happened to be. Ideally this would have involved juggling in some way, or perhaps line dancing. Perhaps it’s not too late, even now.

    It takes some remarkable contortions of thought to take what is surely one of the best possible examples of how human cultural evolution has broken free of human biological evolution and is advancing independently of it (and much faster), and to perversely try to concoct a biological basis for it.

  3. You said it better than I did.

  4. Ideally this would have involved juggling in some way

    Isn’t that sign language?

  5. David Eddyshaw says

    I would like to see a sign language that involved actual juggling. That would give the Universal Grammar people something to chew on.

    (On reflection, there may be pragmatic reasons why this does not seem to have happened so far.)

    Talking of Universal Grammar, I’ve just started Culicover and Jackendoff’s Simpler Syntax, and am impressed (and educated) so far. A book by two of Chomsky’s former students, with an enthusiastic endorsement by Geoffrey Pullum! (I suspect everybody with any expertise in this area is familiar with it, but it was new to me.) It does the exact opposite of the cult-Chomskyite move of attempting to bury inconvenient facts by relegating them by fiat to the realm of the peripheral, instead looking the peripheral to shed light on the supposed core.

  6. David Marjanović says

    The neural pathways were obviously there all along, and there for quite other reasons

    Bingo. Writing is developed when there’s a need for it, no sooner.

  7. The ability to perceive and read patterns is of vital importance to hunter gatherers. Eg. Children in hunter gatherer societies are taught to read animal tracks. Could it be that the same ability enables us to read writing?

  8. Stu Clayton says

    That must be it. Much of what is on offer to read in the media resembles the large heap of dinosaur shit in Jurassic Park that concealed a cellphone. In a neat twist, the cellphone led to the spoor, instead of the other way around.

  9. I believe there’s a well-known brain area that’s involved in reading. In people who never learned to read, it has a different function. This is the fusiform gyrus. See the brain section of the Wikipedia article on word recognition.

  10. There are indeed pragmatic reasons that language does not involve juggling: some people are just not capable of it. I personally have unusually slow reflexes. This manifests itself in all sorts of ways—from being unable to succeed at many video games (in spite of practice), to being unable to juggle balls. From a fairly young age, I wanted to learn to juggle, but despite quite a bit of practicing, I simply cannot make the necessary adjustments at the required speed to perform regular juggling maneuvers. I can juggle scarves with no problems, since they move more slowly than the standard balls. I can also start off a juggling pattern with three balls, but unless I loft each one perfectly, their trajectories start to wander around a bit, and I do not have to react quickly enough to make the adjustments needed to keep them in the air.

    So if we communicated by juggling, I would probably be limited to either speaking extremely slowly, or to monosyllables.

  11. @Brett As a long time juggler, from your description I can tell you know how to juggle and have juggled, although perhaps not as well as you would like. Your concern about the trajectories is shared by the best jugglers, if the throws are good the catches are easy. If you want to improve I suggest beanbags, they don’t roll away, and practice ten minutes a day.

    I agree with you juggling is unlikely to replace talking as a communication medium except to show others juggling, which is easier to do than to talk about. Most non-jugglers don’t see what the juggler is doing, it seems impossible or magical. I can tell you have passed that stage, at least for juggling three objects.

  12. John Cowan says

    As a programmer, talking about things that are hard to talk about is my profession. So it’s not surprising that I learned to juggle from “Juggling for Programmers”, or rather its original version in Seymour Papert’s book Mindstorms. It’s a pair of concurrent algorithms that execute a cascade juggle of three balls (or other objects). Here’s how it works:

    Actions: TOSSRIGHT is the action of tossing the ball from the right hand to the left in a high arc (the higher, the slower and the more time to think). TOSSLEFT is the same action, but from the left hand to the right. The eyes need to stay at the top of the arc and not follow the ball. This is perhaps the hardest part of the whole thing.

    Events: CROSSRIGHT is the event of a ball crossing the top of its arc when moving toward the right. CROSSLEFT is the event of a ball crossing the top of its arc when moving toward the left.

    Now suppose that we start with two balls in the left hand and one in the left. The first action to do is TOSSLEFT, which will shortly cause a CROSSRIGHT event to occur and would (if nothing else happened) leave us with two balls in the right hand and one in the left. But after this TOSSLEFT, we program ourselves to act as follows:

    1) Whenever CROSSRIGHT happens, TOSSRIGHT. This makes sure that there is room in the right hand for the incoming ball.

    2) Whenever CROSSLEFT happens, TOSSLEFT. This makes sure that there is room in the left hand for the incoming ball.

    Keep doing this, and we keep juggling!

  13. @david: I appreciate your encouragement and suggestions, but unfortunately, I really have tried hard to learn—including using a routine that was is pretty much what you suggest—and while I can mange minor improvements with effort, my reaction times really are too slow for me to do it consistently. (I have measured my reaction times more scientifically using other methods, and they pretty consistently are over 2σ worse than average.)

  14. For an earlier version of the argument that rhythmic pattern-making is a precursor of writing, see André Leroi-Gourhan, _Le geste et la parole (I: Technique et langage. II: La mémoire et les rythmes)_. Paris: Albin Michel, 1964–65.

Speak Your Mind