A Chat With Guy Deutscher

Guy Deutscher is the author of The Unfolding of Language: An Evolutionary Tour of Mankind's Greatest Invention. He earned his Ph.D. in linguistics at the University of Cambridge and now teaches at the University of Leiden in Holland. The Unfolding of Language is his second book. He has also written for The Times and the New York Times.

Transcript of the chat

Joining us from Leiden, Netherlands, is our guest Guy Deutscher, the author of The Unfolding of Language: An Evolutionary Tour of Mankind's Greatest Invention. He teaches at the University of Leiden.

Welcome, Guy!

Guy Deutscher
Great to be here! My first online chat.

Faldage - USA
When I look at language change I see both change from regular to irregular and irregular to regular. Why do you seem to emphasize the former?

Guy Deutscher
I don't emphasize just regular to irregular. What I say is that the regular to irregular direction is much easier to spot, because you can reconstruct from the irregular present a regular past. But it's more difficult to go the other way around. Since once things have become regular, they usually leave no traces of their former misbehaviour.

Jeff - United States
Sorry, I haven't read your book. Can you give an example?

Guy Deutscher
One example I give is the Latin word for flower: it's flos in one case, but flor-is (with an r) in another case. This variation of s and r immediately alerts us that something went 'wrong' at some stage, and then we indeed find out that originally it was flos and flos-is, without any irregularity.

But on the other, when you have a form that is now regular, say the English word 'chose', you'd never imagine (unless you have records) that it was originally irregular. In fact, in Old English there was exactly the same s-r irregularity there, but it's disappeared without trace.

Faldage - USA
But if a word exhibits irregularities in one language and is regular in another how do you know which is the original?

Guy Deutscher
Good question. I don't think there is a rule which would always work here, since changes can go either way. You'd have to look at the context, and try to work out from there what's a more plausible scenario.

Brazilian dude - Brazil
By chose you mean the past of choose, right? Do you mean the other way around, irregular and regular?

Faldage - USA
Most of those verbs we now call irregular were classified as strong in Old English. do you believe that there was an earlier form that would look regular to us?

Brazilian dude - Brazil
Talking about present and past, what do you think of past tenses such as snuck, dove, squoze, and slinged?

Guy Deutscher
'Chose' as the past of 'choose' yes. Originally it was irregular: I 'ceas' but they 'curon' . But this irregularity has been wiped out... OK, trying to take one thing at a time: I love snuck and squoze etc. It's nice that things don't *always* just go the more regular direction. It would be boring otherwise. So yes, these are examples where the earlier form was 'regular' and it became 'strong'. If that answers the second question.

Brazilian dude - Brazil
But it's still irregular, isn't it? Otherwise we'd have choosed, the form curon was lost, but a new one emerged, indubitably similar to other irregular forms that probably already existed.

Guy Deutscher
OK - but we're talking about two different types of irregularity. Choose remains a verb with a 'strong' past, rather than the regular 'choosed'. But the irregularity I was talking about is within the past tense, the variation between s and r. That's completely disappeared.

zmjezhd - Germany
Why do you think that historical linguistics has had an easier time with reconstructing a proto-language's phonology than its morphology (e.g., case) or syntax? Is it just an historical accident or is historical morphology just tougher in and of itself to reconstruct?

Guy Deutscher
Ooh - really tough ones. Part of the reason may just be that much more effort went into phonology from the start. But I think by now there has been a lot of progress in morphology as well. Syntax is genuinely tougher than either morphology or phonology, because it's more abstract. We're not talking about actual forms mostly, but things like word order which leave fewer tangible traces.

indigotulip - United States
Are there particular time periods you find yourself studying more than others? In other words, do you find changes concentrating during particular periods?

Guy Deutscher
Well, it's not a great secret that certain social or political conditions can contribute to rapid changes - you just have to think of England in the eleventh century, with the Danes, Normans and so on. At this time we see great upheavals in English. But the reverse doesn't necessarily hold - social and political stability doesn't mean a language won't change.

gomek - usa
I hope this isn't out of context. I realize this is a broad subject, but at the beginning of human speech, what was the governing body that agreed what sounds would be applied to an object of person, etc. How did grunts and such evolve in to words?

Guy Deutscher
The million-dollar question... The honest answer is that no one has a clue, although thousands of people have speculated. In my book, I concentrate on the period when there were already some words for simple objects (like tree or stone) and actions (go, kick), and try to take it from there, and see how our sophisticated languages could have developed *without* any governing body devising grammar.

indigotulip - United States
What do you think are the effects of increased international media and technologies on linguistics/morphology, particularly regarding changes of regularities and irregularities ?

Guy Deutscher
Well, there are reasons to think that in a more diverse community (and the global village is as diverse as you can get), there would be stronger motivation for simpler regular forms to establish themselves. And maybe on the whole this will happen. But you can see it's not straightforward with the example of 'dove' which was just mentioned a moment ago. It's an American English innovation, but because of the prestige of American, it's likely to spread.

Well, concerning the more most vs. -er -est: there are signs that more words now are joining the more most camp. But I believe a hard core of -er -est will always remain, such as older, and other monosyllabic words.

zmjezhd - Germany
What do you make of a case like the grammaticalization of the comparative/superlative in English. Currently there are two partially overlapping ways to form them: tall, taller, tallest, or irregular, more irregular, most irregular. It's hard to foresee if the latter (newer) will succeed over the former (older) method. Or if these forms will live on in an irregular way.

indigotulip - United States
For us non-professionals, what are the topics and questions of hot debate and study among linguists today - among professionals/in academia?

Guy Deutscher
Well, linguistics is a very wide field, and what's very hot for some is pretty irrelevant for others. But I guess one debate/controversy that's been hot for a long time and isn't showing signs of cooling is the 'innateness' question: how much of language is coded in the genes, and how much is a product of cultural evolution. It's very controversial, and splits linguists into two (rudely) opposing camps.

indigotulip - United States
So it's revealed again: linguistics is a science, down to the question of nature vs nurture?!

Syd Perry - USA
Which side of the camp do you live it, language coded in the genes or language as a product of cultural evolution?

Guy Deutscher
Hmmm. I suppose I'm more on the nurture side. I accept of course that we have an innate ability and propensity to learn language. But I don't believe there is a need to invoke innateness to explain the more detailed rules of grammar. These, I believe, can evolve through the pressures of communication, and in my book I've tried to show exactly how.

Faldage - USA
What other languages have you studied?

Guy Deutscher
Other than what?... My main area of specialization is the Semitic languages, and in particular, Akkadian, which was the language of ancient Babylon and Assyria, and was spoken in what is today Iraq, some 4500 years ago up to 2500 years ago.

indigotulip - United States
Do you think that maths were a helpful preparation to linguistics, and if so, how do you apply that knowledge in your current work?

Guy Deutscher
I think maths was a great preparation, in that it teaches one how to think logically. That's more important than any details. So I'm extremely happy I did maths to start with. But I don't use much of it directly now, except some statistics sometimes.

Brazilian dude - Brazil
Why is it that English seems to evolve much faster than other languages? Maybe because English doesn't have an organized body that "controls" language such as the Real Academia Española or the Académie Française or those rapid changes are due to the indisputable flexibility of the English tongue? Just to illustrate, our verbs have to end in -ar, -er, or -ir, anything else doesn't qualify.

zmjezhd - Germany
LOL, Brazilian dude. All English infinitive verbs have to start with "to ".

Guy Deutscher
English changed very fast over a few centuries between around 1000-1400, partly because of the socio-political context, as I mentioned. Other than that, I am not sure it has changed so much more radically than French or other languages. What may be the case is that in France, say, the standard written language has been artificially preserved more than written English has. But the result is that there's a much wider gap there between the spoken language and the written language. In other words, spoken French has changed just as much as spoken English recently, only that you are not allowed to notice that from the writing.

indigotulip - United States
Are there tangible legacies of Akkadian? (scrolls, later-written documentation) What does one study or refer to when studying a language (dead, I assume?) over 4000 years old?

On a more personal note, how did you learn about Akkadian and what drew you to this speciality?

Guy Deutscher
Akkadian was written on clay tablets (a wedge was used to make marks in the wet clay). Once these clay tablets have dried, and especially when they have been burnt, they are almost indestructible. So hundreds of thousands of tablets have been found in Iraq and neighbouring countries, and there are hundreds of thousands more undoubtedly still waiting to be found. We have an enormous amount of texts of all types and genres: from shopping lists to epic poetry. I have a picture of one of these tablets and its nice text in my book, and also on my website: unfoldingoflanguage.com.

And on the personal note: I wanted to study a language with a long history, and which starts very early. Akkadian starts at 2500 BC, and goes on for 2000 years. (English only has 1000 years of written history). So it couldn't easily be beaten.

Faldage - USA
Thanks for the web site and we'll be looking for your book. Some irregularities we can reconstruct, e.g., umlaut plurals coming from following vowels and I have a vague feeling that we've explained the strong verb ablaut series in a similar manner. Do we know as much about other irregularities, such as the s/r alteration?

Guy Deutscher
The s-r alteration comes from a very simple and very common type of sound change: s changes to r between vowels, because it's easier to pronounce. In Latin, the original flos-is changed to flor-is because the s was between two vowels, but in flos, the s remained unchanged because it wasn't between two vowels. This has caused the irregularity we see in the historical period.

Faldage - USA
What is your take on the great descriptivist/prescriptivist debate?

Guy Deutscher
Well that's one issue where at least among professional linguists, there isn't much debate. We're all descriptivists. That doesn't mean I believe it doesn't matter how people speak. Of course it does, because people are being judged by others by how they speak. So if you speak a non-standard variety in a job interview, say, you might be disadvantaged. But what is plain silly is to say that non-standard varieties are objectively or inherently less good than standard ones.

zmjezhd - Germany
With such a large corpus of texts, is there any evidence of non-Assyrian speakers writing in Assyrian? I seem to remember that some of the materials preserved are of a pedagogical nature (e.g., scribes learning their cuneiform). Examples of Assyrian changing and the resistance to those changes?

Guy Deutscher
Yes, people learnt Akkadian all over the ancient near east. Egyptian pharaohs wrote to Hittite kings in Anatolia in... Akkadian. It was the lingua franca of the day, and scribes from all over had to learn it - sometime they make really bad mistakes.

Brazilian dude - Brazil
On a personal note, isn't it frustrating to study a language that has no one to converse with? Coming from a guy who studied Latin and never felt that frustration because Latin lives on.

Guy Deutscher
I suppose it is sometimes quite frustrating. But some of the texts I work with, e.g. simple everyday letters between people, are so 'alive' - so you're reading a text from 4000 years ago, say a letter from a nanny to her employer talking about the children, and it's almost as if it was spoken today. Also, these people dictated their letters just as they spoke, so the style is very simple and conversational sometimes. It's as close as we get to chatting in Akkadian.

Faldage - USA
Were Akkadian and Assyrian the same language?

Guy Deutscher
Assyrian and Babylonian are the two main dialects of Akkadian. Assyrian was spoken in the north (north of Iraq today) and Babylonian in the south.

indigotulip - United States
Did Akkadian progress through vowel changes as the Germanic languages have (assuming Akkadian used vowels and consonants - was it based "alpha betically"?

indigotulip - United States
Is the change between Akkadian/Assyrian related in any way to the c & k pronunciation question in Latin?

Guy Deutscher
Yes, there were many changes in both vowels and consonants through the 2000 years of Akkadian history. In fact, an Akkadian speaker around 500 BC would almost certainly not understand very much of what a speaker around 2000 BC was saying. But not much that's equivalent to the k&c issue in Latin that I can think of.

indigotulip - United States
Has anyone translated some of these tablets - especially, say, the letters?

Guy Deutscher
Yes, there are various translations. I have a small selection in my previous (academic) book: Syntactic Change in Akkadian (Oxford University Press). Also, in my current book, in the notes at the end, there are some references to books on the subject.

Anu Garg (Moderator)
That was our last question for today. Thanks to all who participated even if we couldn't field your questions due to limited time. For more, please see unfoldingoflanguage.com.

Anu Garg (Moderator)
Thank you, Guy, for being here today and thanks to all the participants.

Guy Deutscher
Thanks, Anu, for inviting me, and thanks for all those sharp questions.

indigotulip - United States
Thank you, Guy!

Brazilian dude - Brazil
Thank you, Mr. Deutscher.

Rayob - USA
Thank you.

indigotulip - United States
And thank you Anu!

zmjezhd - Germany
Thanks and goodbye, Professor Deutscher.

Guy Deutscher
Thanks again, everyone.

Faldage - USA
Thank you.

Anu Garg (Moderator)
We look forward to your comments on today's chat. Please send your feedback to (garg AT wordsmith.org) on how you enjoyed the chat and how we can make it better. Thank you.

Our next Wordsmith Chat guest is Dr. Madan Kataria, the founder of laughter yoga movement. This event will take place on September 6, 2005, 9 PM Pacific (Sep 7, 4 AM GMT). We hope to see you then.