Writing and speech as chicken and egg


This is a guest post by Tim Gorichanaz

Evolutionary linguistics seeks to explain the origins and evolution of spoken language, but it tends not to consider written language. Perhaps rightly so: Writing is different from speech, and trying to consider both at once might only cloud things up. Still, given that writing is a symbolic representation of human thought—just as speech is—I believe analyzing the development of written language can be helpful in fleshing out the holistic story of the evolution of language.

We might, at first blush, think of writing as dependent on spoken language. Charles Hockett, for example, famously described writing as a *reduction* of spoken language. Indeed, our alphabetic writing system seems to correspond (granted, in some ways more than others) to the sounds of our language, minus paralinguistic and pragmatic features like speed, tone and volume.

But taking a historical and global perspective suggests that written language and spoken language may not be as intertwined as it seems at first: Classical written Latin had little to do with the Medieval Romance languages, though it was considered a visual representation of the same; moreover, the Chinese and Arabic writing systems are, even today, used to represent disparate dialects that are, in speech, not mutually intelligible.

Is it possible, then, that writing did not arise as a way to overcome the limitations of speech, but rather something that emerged independently and alongside spoken language? Let us consider the possibility that both writing and speech sprang up at the same time, both fruit of the same evolutionary boom that gave rise to our human brain with the capacity for symbolic thought. Writing, then, can be seen as a co-evolutionary force that developed alongside spoken language. Just as writing and speech influence each other today, perhaps this bidirectional influence was of some importance during the evolution of both.

After all, in the 100,000-year-old smears of red ochre on some rocks in Israel, we can see the roots of written language just as we can see the basis of spoken language: It is the oldest element of symbolic human thought, a necessary component of both. Our ancestors went from smears to hand tracings to images of humans and ruminant animals, and from there continue the history of art and that of writing. And also that of language.

What linguistic lessons can we learn from the evolution of pictograms to cuneiform wedges to veritable alphabets? And what of the different languages and cultures that collectively decided that their needs were best served by syllabaries as opposed to alphabets, or logographic systems as opposed to something more phonetic? I think we’ve barely scratched the surface.

N.B. If you’re interested in learning more about the history of written language, I highly recommend the book The Writing Revolution: Cuneiform to the Internet by Amalia Gnanadesikan.

Bio: After working a few years in digital advertising, Tim Gorichanaz returned to academia to earn a master’s degree in applied Hispanic linguistics. He wrote his thesis on the linguistic issues of written language on the Internet, and he plans to pursue similar investigations as a PhD student in the near future. In the meantime, he records his thoughts and research about written language on his blog, ScratchTap. When he’s not studying, Tim can be found running (ultra)marathons and drawing. 

5 thoughts on “Writing and speech as chicken and egg”

  1. I think people don’t consider written language in language evolution because there’s a consensus that, compared to language, writing has been around for an incredibly short amount of time.

    I’m not sure what your point is here though, are you trying to say that written languages was implicated in the emergence of language from the beginning? I think it’s fair to say writing’s probably affected language change since its invention, but I don’t think it was implicated in the initial emergence of language, and it almost certainly didn’t emerge independently. Even if writing is not representing the phonetic properties of language, but the semantic or other properties, it’s still an invented system to represent the language, and not an independent entity which emerged independently of language.

    Also, I’ve no idea what you mean by statements such as: “Classical written Latin had little to do with the Medieval Romance languages, though it was considered a visual representation of the same.” ???

  2. Thank you for taking the time to read and comment.

    I’m not so sure that writing emerged as a way to encode language (but it has, of course, proved handy for this). Though linguistic writing is only a few thousand years old, we have written records of numerical calculations that go back even further. In other words, the first “writing” was not meant to record speech, but rather tax information and business transactions.

    If we take cuneiform as an example, written language seems to have sprung from written arithmetic, which in turn may have sprung from other visual symbolic representations—think of cave paintings and burial sites. Going back in this way, we get to the first records of symbolic human thought, about 100,000 years ago. Certainly this expanse of time cannot be considered inconsequential when we are discussing a species that’s less than 200,000 years old.

    Is it so outlandish to suggest that language and writing both resulted from the same source? Writing seems to enjoy a long history quite separate from that of language; it is only in the past few thousand years that they seem intertwined. (Notably, writing has probably changed more than language during this time; we have seen tremendous changes in writing in just the past few thousand years.) I’m not suggesting that I have all the answers—I’m only offering some food for thought.

    Regarding the Latin versus Romance language distinction, we must consider the spread and development of Latin. In the writings of philosophy, history and science, Latin remained quite stable over the centuries. But in the mouths of people scattered over thousands of miles, it did anything but. As a result, somebody in the western Iberian Peninsula in the year 700 probably couldn’t understand someone from Venice very well. Even so, both the year-700-Lusitanic Latin and the year-700-Milanese Latin were both written in the same way. (Granted, writing was something not many did.) In other words, it was a case where two different languages were encoded via a single writing system. There was not yet the conception that Latin was a different language from year-700-Lusitanic or year-700-Milanese. In subsequent centuries, the new languages were given names and honored with their own writing systems (in other words, standardized), but for a long time they were simply considered spoken versions of Latin, disparate though they were. If you’re interested in the topic, you might like the book “Latin and the Romance Languages in the Middle Ages.”

  3. Even if writing systems originally emerged from arithmetic, this is just another example of written forms representing semantic information, rather than phonetic features. They’re still there to represent language.

    As for the Latin story, whether they considered it a different language or not, doesn’t mean that educated, high-society people (who did all the writing) in 700 learned to write in Latin in a certain register. Even then, it has it’s origins in representing how latin was once spoken. I don’t understand why this says anything about writing being independent of language?? Yes, they change and develop at different rates, but language has an almost* one way effect on writing (as you say, eventually the new languages born from latin invented their own writing systems to represent those languages).

    What do you mean by “both resulted from the same source”? I guess I just don’t really understand whether you’re trying to suggest any story other than that spoken language came before writing? Your title suggests you might be… I can’t think of a plausible scenario for this, and if you’re not suggesting that then it’s possible you’re conflating language emergence with language change. The latin stuff is definitely a question of change rather than anything that can tell us about writing origins, and all it’s showing us about change is that the linguistic register used for writing changes more slowly than in speech, which is already a well-known phenomenon in linguistics.

    *Though writing practices can arguably affect linguistic structure too http://www.lel.ed.ac.uk/~monica/Tamariz_Brown_Murray.pdf

  4. Tim,

    Thank you for this. I think you’ve raised a very interesting topic. I would however urge caution in your use of popular ideas about the relationship between Chinese script and the so-called dialects. It’s at best debatable to say that the Chinese script represents mutually unintelligible spoken dialects. Actually, the parallel with the use of Latin by speakers of a range of European languages, is quite strong. As Hannah has pointed out, this doesn’t seem to support your case. If your mother tongue is, say, Spanish, and mine Chinese, our English communication is surely not a representation of our mother tongues. We’re just communicating in a mutually intelligible language.

    The idea that language and speech sprang up at the same time is obviously something that doesn’t get aired too often. This is slightly disappointing, and I think Hannah’s initial comment about ‘consensus’ is pretty typical.

    @Hannah Why should we believe that writing has been around for an incredibly short time compared to speech? And why must we call speech ‘language’, suggesting that only speech, and not writing, can be thought of as language?

    I find it hard to imagine that early man formed distinct sounds for the numbers one to three before he learnt to make, and ‘read’, three scratches in the dirt. Why should the spoken and written forms not have emerged together?

  5. I never said speech = language, and i never would, I’m hyperaware of the difference because i’m doing my PhD on speech. I was using language to refer to the linguistic information in our brains and speech and writing are both externalisations of that. I was preferring to say that writing represents language, rather than speech, because this covers instances where there isn’t a one to one mapping of phonetic to orthographic information. If anything, Tim’s persistent use of the word speech was clouding his arguments about that. Also, speech doesn’t cover other forms of signal that humans can make to externally represent their language, e.g. gesture, and it might be easier to conceive people might have used points, or 3 fingers, to represent 3 before creating numerals in the dirt. Obviously, we can’t know, but there’s also a huge leap between iconically representing things in the dirt and the fully fledged writing systems we have today, which were getting a bit conflated in the above blog post too. And when I said writing has been about for an incredibly short amount of time, I was referring more to fully fledged, arbitrary systems, which we do know haven’t been around that long precisely because they’re forms of externalisation that aren’t ephemeral like speech is. As for speech, informed guesses now assume that this goes back as far as Neanderthals and sooner. As usual, the problems here are do with the correct use of terminology or assuming incredibly broad definitions whilst conflating them with narrower concepts.

Leave a Reply