Skip to main content
Loading Explore guides…
Sociolinguistic Typology · Acedra
Home
Explore
Linguistics
Sociolinguistic Typology
Sign in to save
Chat
If YouTube asks you to sign in, open the video on YouTube.
Watch on YouTube
Reload player
Transcript
69 segments
Copy
Hide video
0:00
JMc: Hi, I’m James McElvenny, and you’re listening to the History and Philosophy of the Language Sciences Podcast, online at hiphilangsci.net. There you can find links and references to all the literature we discuss. Our exciting mini-series on contact linguistics reaches a dramatic climax in this episode, where we talk to Peter Trudgill about his work on sociolinguistic typology and
0:34
language complexity. Over his long career, Peter Trudgill has made numerous important contributions to English dialectology and to sociolinguistics more broadly. He has worked at a number of different universities but is now Emeritus Professor of English Linguistics at the University of Fribourg in Switzerland. So, Peter, in several of your publications, although most extensively in your 2011 book Sociolinguistic Typology, you’ve made the case that different kinds of social
1:07
structures are conducive to producing different kinds of linguistic structures. Most notably, you’ve argued that societies with small and very dense social networks, so so-called societies of intimates, will tend to have languages that are more complex than those spoken in large-scale, very diffuse communities. Could you briefly unpack this idea for us and explain how it works? PT: Yes. Thank you. I’ll give it a go. I’m glad you said ‘tend’,
1:38
because ‘tend’ is important. The idea is that some communities are more likely to see the development of linguistic complexity than others, more likely, and my suggestion is this. Some people have been kind enough to refer to this as the Trudgill conjecture,
1:58
so maybe I should say my conjecture is that small, isolated, tightly-knit communities with dense social network structures are the ones which are most hospitable to linguistic complexity development. That’s a term, by the way, I owe to Jim Blevins. I was giving a talk in Cambridge once and I was struggling to say that these communities don’t necessarily produce more complex languages, but they’re most likely to. So he said, ‘Aha, you mean they’re most hospitable to linguistic
2:28
complexity development,’ and I think that’s a very good way of putting it. So the point is that these communities, they don’t necessarily produce linguistic complexity, but they’re the communities which are most likely to do so. I guess the question, then, is why? Why would this be so? Well, I reckon that languages have a natural tendency to get more and more complex over time. Perhaps that’s not too surprising a claim. After all, we all agree, I think, that languages
2:59
have a natural tendency to change over time, and when our students ask why that is, ‘Why do languages change? What’s the point of that? How does that come about?’ I think we might be inclined to reply, ‘Well, they just do.’ May seem like a rather weak argument. Here we are, we’re the experts on language and all we can say is, ‘Well, languages are like that,’ but I think that’s the right answer. It’s inherent in the nature of human languages that they change. What I
3:28
want to suggest is that they also have an equally natural tendency to get more and more complex over time. Typologists seem agreed that any particular complex linguistic feature you point to, these features are all developed out of earlier, less complex states. So if you take German umlaut, for example, which is a complexity in the Germanic languages, then if you go back far enough into Proto-Germanic, there was no umlaut. Umlaut developed out of a situation where there was no
3:59
umlaut. So my point is, from the sociolinguistic point of view, crucially, this is what naturally happens and always used to happen, or nearly always used to happen, unless certain social factors intervene. And these days, social factors very often intervene because the most important social factor, I would say, is contact with other languages — specifically, of course,
4:27
contact with speakers of other languages. And this is specifically contact leading to the learning of the language by significant numbers of adult non-native learners. There’s a sort of history of linguistic science story to tell here, which I think is rather chastening. I’ve actually pointed out over the last 10 to 15 years — and again this has been given the name by some colleagues, very kindly, it’s called the Trudgill insight — but truly, I’m a bit embarrassed about it,
4:56
because it’s one of these things which is really obvious when you think about it. I’d been spending decades attending conferences about dialectology and sociolinguistics and talking to creolists and specialists on pidginization, and everybody agreed that language contact leads to simplification.
5:16
But then I started finding myself at conferences where there were typologists, and typologists took for granted the opposite point of view. They were all saying, ‘Well, we all know that language contact leads to complexification.’ So Johanna Nichols was very clear that in the areas which she’s investigated, contact, like in the Caucasus, contact between languages led to complexification because speakers would borrow, for example, grammatical categories from other
5:48
languages and gradually over time add to the grammatical complexity of their own languages. So on the one hand, language contact leads to simplification; on the other, language contact leads to complexification. We can’t both be right. Well, actually, of course, we are. It depends on the nature of the contact, and it’s become clear, I think, that the sort of contact that Johanna Nichols was talking about was long-term contact in geographical proximity between speakers of
6:19
different languages and different language groups where large numbers of people become bilingual and where it’s mostly small children who are doing the initial language learning. So we’re thinking about bilingual, trilingual villages where everybody speaks two or three languages and the children become very fluent in most of them or all of them. On the other hand, the sort of contact we, that us sociolinguists were thinking about, was rather short-term,
6:44
not very intimate adult language learning. And I think the point there is that small children are brilliant language learners, and they seem to be genetically programmed to learn perfectly any language they’re sufficiently exposed to, and they learn them very quickly as well. On the other hand, adolescents and adults are mostly — there are exceptions — but I think the correct technical
7:09
term is probably ‘lousy’ language learners. Adults are lousy language learners; I know I am. And this means that adults, wherever possible, will make things easy for themselves, and this is where the simplification comes in. So societies of intimates with low levels of external contacts, I would suggest, tend to have more complex languages because nothing has happened to impede the natural development of complexity, what I claim to be the natural development of
7:37
complexity. And actually, time is rather vital here. If you think about a language as complex as Ancient Greek, for example, then we’re surely talking about complexity development stretching back over hundreds of years, possibly even thousands of years. So if you ask, you know, if you ask, ‘What does it take for a language to develop a system where all the nouns, adjectives, pronouns, articles, numerals, and verbs are inflected,
8:06
and with three main different major declensions, which also has massive allomorphy — you know, what does it take to develop not only three numbers and three genders, but also five morphologically marked cases, with the definite article in Ancient Greek having 36 different forms? And I think one answer must be time. It takes a very long time for this to start with a situation where you didn’t have these features and end up with a situation where you do. I like something that
8:36
Roger Lass said about this. He said that there has been — and I’m reading here, James — ‘a traditional intuition of evolutionary direction that morphological complexity decreases’. And I think that was true. Amongst linguists, we’re all familiar with the fact that in some sense
8:53
Modern German is less complex than Old High German, that Italian is less complex than Latin, that Modern Greek is less complex than Ancient Greek and so on. So we can all see a direction in linguistic change, and that direction is in the direction of simplification. But in fact, this sociolinguistic typological perspective which I’m taking suggests that earlier linguists who thought that may have been right when they reckoned there was a kind of evolutionary trend in linguistic
9:23
change, but this was for demographic reasons and sociolinguistic reasons rather than linguistic reasons. OK, so linguistic simplification has been going on for, what, the last 2,000 years, but this makes it a very new phenomenon in the scale of the whole of human linguistic history.
9:41
So how old is human language? Well, linguistic scientists seem entirely agreed about this. Some people say 50,000 years, some say 100,000 years, some say 200,000 years, some even say 500,000 years, but whatever it was, it’s a very long time, and 2,000 years, the last 2,000 years, is a very short period of time, comparatively speaking. And I would suggest that this sudden rise in simplification is due to a dramatic increase in the world population,
10:11
and the resulting increase, therefore, you know, the more people there are, the more adult language contact there’s going to be. And Johanna Nichols said something very sensible about this. She said that language contact has been responsible — I’m reading here now — ‘language contact has been responsible for much reduction in morphology in Europe over the last two millennia’ but that this was probably ‘rare in prehistory’. And I entirely agree about that.
10:36
JMc: So you’re suggesting that there’s a difference in kind between societies in prehistory and modern-day societies, essentially, and you’ve also suggested that the increase in language complexity is a natural phenomenon. Is the flip side of that, then, that simplification through sociolinguistic contact is somehow unnatural, and is a further consequence of that that small communities that are around today that have a small amount of contact, that they somehow
11:08
are more natural, perhaps even living in a state of nature? And is this maybe
11:14
just a sort of a reiteration of old Romantic tropes, you know, about the difference between civilization and barbarity? PT: Well, I hope not, but, I mean, I see why you could put that twist on it. When I say ‘natural’, I’m not saying desirable or more admirable in any way. All I’m saying is simply that languages get more complex unless something else happens, to repeat myself there, and that
11:47
something else is language learning by adults who willy-nilly, whether they want to or not, end up simplifying languages because they can’t do anything else. It’s important to realize that the starting point for this point of view that I’m adopting is the assumption of our common humanity. The structural characteristics of all human languages, I would say, past and present are due to the nature of the common human language faculty. We’ve all shared the same human
12:17
language faculty ever since humans became fully human and acquired language whenever that was,
12:23
50,000 years ago, half a million years ago, who knows? And so the puzzle is, we all share this common language faculty, so where does the great diversity of human languages come from? It doesn’t seem to be produced directly by the human brain or the human mind or the human language faculty. Other things must have come into play, and those other things, I’m suggesting, are sociolinguistic things to do with societies. So I don’t want to romanticize small, tightly knit communities,
12:55
or to say that they’re more natural, but if you think about it over the period of, let’s say, 200,000 years, they have been… Well, let’s avoid the word ‘natural’, but they’ve been the norm until 2,000, 3,000 years ago. Nearly all communities were very small, and they were communities where everybody knew everybody else. We can say that if you look at the history of human languages, this has been typical. Typologists are very careful to avoid
13:22
bias when they’re making claims about the nature of human language and what are the limits of grammatical structures in human language and so on, and the bias they’re worried about is
13:34
geographical bias. If you’re going to make claims of that particular type of typological sort, you want to make sure you’re talking about languages from all continents of the world, because something which is true of European languages may very well not be at all true of Australian languages, say, and they also want to avoid genetic bias, so you don’t want to make your claims on the basis of languages from just a few language families — you want to include all
14:02
language families. Now, that’s right and proper, and I think we’re doing a good job of doing that these days, but one thing they can’t avoid is chronological bias. If languages are spoken today, and perhaps for the last 2,000 years, are not typical of human languages as they typically have been spoken for most of human history, then our generalizations apply to the modern situation and not to human languages generally. So I don’t
14:32
want… We’re not saying more natural or more romantically attractive in some way — we’re just saying more typical if you look at it from a historical perspective. Does that make sense? JMc: Yeah, that makes sense, but I wonder, what do we know about what has been normal in the history of humanity? So one thing that comes to mind is — at the sort of popular end of anthropology — David Graeber and David Wengrow’s book from 2021 with the
14:59
grandiose title The Dawn of Everything. Well, one of the many arguments they make is that this image that we have of Indigenous communities and colonial societies, of the division between modern and pre-modern, is actually a caricature, and that in so-called pre-modern societies there are many different kinds of structures and many different ways that people live. What do we actually know about what was normal in terms of the history of the earth?
15:25
PT: Yeah, I agree with you. One thing we do know, of course, is, we do have figures about population. We know that if you go back into the Neolithic Age or the Mesolithic or the Palaeolithic, we’re talking about very, very much smaller population size on the world scale, and we really do have to imagine not idyllic, natural communities living in a state of natural bliss, but we have to imagine the situation where in many parts of the world people were wandering around
15:56
in relatively small groups, only rarely bumping into other groups because there weren’t very many other people to bump into. And that’s real. This is not an artificial distinct… I’m not trying to make a big thing about Indigenous communities and colonial communities, about urban societies and hunter-gatherer societies. I don’t think that’s the issue, but I think the issue has to do simply with demography. The more people there are, the more contact you’re going to get. You asked,
16:25
‘What do we know?’ Well, we do know there were many fewer people in previous millennia. JMc: So, I mean, language complexity is itself a complex notion, isn’t it? So you’ve said that they’re features, the grammatical features that are difficult for adult learners to master, but would it be fair to say that complexity is in the eye of the beholder? Could it be that this conception of complexity is merely a product of the grammarian’s gaze,
16:50
if I can put it like that? So there are perhaps other aspects of language use that might mark someone as a non-native speaker, as someone who has learned the language as an adult, but which would not be included in the average reference grammar. PT: Yes, I think you’re right there, but it’s obviously correct that linguistic complexity and complexification are a complex notion, and so are linguistic simplicity and simplification. I don’t
17:16
really see that complexity is in the eye of the beholder, not the way I’m using the term, anyway. I’ve been really rather specific about what I mean by ‘complex’. But certainly within the linguistics world, there seems to be agreement here, James, so that, you know, polysynthetic languages, to take one example, have a degree of complexity which linguistic scientists seem agreed about. So Fortescue said they have a ‘daunting’ level of complexity. Mark Aronoff said ‘startling’,
17:45
startling complexity. Keren Rice said ‘legendary’ complexity. Nick Evans has used the word ‘exuberant’ complexity, and other people have said ‘spectacular’, ‘baroque’, baroque complexity, and ‘rich’ complexity, ‘unusual’ complexity, and so on. I think that what I felt I needed to do was answer the question, what is complexity, exactly? And here I’ve been guided by, you know, previous work. One of the greatest pieces of work on this topic is the marvellous book
18:17
by Östen Dahl, The Growth and Maintenance of Linguistic Complexity, and he’s discussed in some detail there what complexity actually is in language. My starting point was, again, really rather simple-minded because starting with the dialectology, the pidgin, the creole literature, it seemed to me that we pretty much agree what simplification is, and then I just turn that on its head and say, well, complexification, complexity,
18:44
is the opposite of that. But Peter Mühlhäuser showed us a long time ago now, many decades ago, through his study of the development of pidgin languages — you’ve talked about these on earlier podcasts with Felicity Meakins, for example — Peter Mühlhäuser showed us what simplification involves, and I think it’s worthwhile listing the features he talked about. What I took away from that work is that pidginization crucially consists of three factors:
19:11
regularization, loss of redundancy, and increase in transparency. I think if I spell out what I mean by complexification being the precise opposite of the simplification we see in pidginization, then, you know, we know where we are and we can continue the discussion. JMc: Those categories that you’ve presented are a product of the work of the linguists, though, aren’t they? They’re products of analysis. Are those categories real for the speakers?
19:40
PT: They’re not things which native speakers are aware of — not consciously aware of — of course, but they are aware of them at some level. Otherwise, they wouldn’t be able to employ them in a native way. They may not know what third-person singular is or first-person singular, but they know how to use and how to interpret /e/ as opposed to /a/. JMc: Sure, but I mean, the precise categories that are arrived at at the end are a product
20:05
of the work of the linguist, and the language could be analysed in a different way. So the particular analysis that the linguist makes is, in a sense, normative. Linguists like to say that they’re descriptivists, that they’re not prescriptivists, so they’re just describing a language as it is used and not imposing some sort of structure onto it, but the work of the linguist is perhaps inevitably normative because the
20:30
linguist has to decide, first of all, what belongs in the grammar, what is relevant, and then they have to decide how they’re going to chop it up, how they’re going to analyse it. And the way that they make various decisions is not given from the language immediately but is, of course, also dependent on the particular tradition that the linguist was trained in. PT: Well, I… Yes, I understand what you mean, but I don’t think I follow it altogether,
20:56
because after all, we haven’t just invented these categories arbitrarily. I mean, third-person singular means something, and we all know it means something, and it gets us places, we can make progress with this. And, of course, if you want to teach a foreign language, it’s very good to be able to say, ‘Well, this is to explain what the third-person singular of the present indicative means,’ and to teach in
21:16
that sense. I do not think this is a question of linguists imposing totally arbitrary categories on the data they’re presented with. I think there is a kind of reality to what we do in this work. JMc: They might not be totally arbitrary, but they’re not entirely natural either, are they? PT: Well, I think you can say that they were. I mean, ‘third person’ means, in the real world, it means that somebody who’s not speaking
21:41
and someone who’s not listening but somebody who’s being spoken about.
21:45
I think that’s real world stuff, isn’t it? ‘Third-person’ actually has a physical meaning. JMc: It’s perhaps a commonplace of the philosophy of science that all observation is theory-laden, maybe I could put it like that. PT: I’m not very happy about the philosophy of science. I’m a simple dialectologist, James. JMc: OK. PT: I tend to look at the data, and I see what I see and I try and work out what it tells me.
22:11
I don’t think I have a particular theory. Obviously, everybody has to have a theory. You have to have an idea about what third-person means and indicative, and so on, yes, of course, I accept that totally, but I don’t think I’m going to plead guilty to imposing irrelevant categories
22:27
on languages that we’re — say, Australian languages, just to give a very good example of the languages we’re investigating. JMc: Your argument about how the most widespread languages today are perhaps not representative of how human language has been typically throughout the history of humanity segues into an argument for documenting endangered languages. This returns to my earlier question about,
22:52
is it right to characterize the growth of complexity as natural? Do you think that saying that endangered languages are perhaps a repository of some, you know, of some special qualities of human language that are not seen elsewhere runs the risk of othering Indigenous communities? PT: Well, what I would say was, if we start from the assumption that we all have this human language faculty in common, you know, as I said before, we’re talking about our common humanity,
23:23
so we’re not othering those communities except in the sense, of course, that
23:28
small isolated communities are Other. I mean, they are generally Other in many ways from large urban communities. They’re not better or worse or… They just, they are different in significant and numerically measurable demographic ways. And I think it’s interesting to… You know, nobody’s making any value judgments here. I mean, of course, there is a tendency for us in linguistic typology to be much more excited by languages we don’t know and which have grammatical
23:58
categories we’ve never come across before, yes, OK, but that doesn’t mean to say we’re romanticizing them. We’re excited about it because what we want to know is, what are human languages like? What are the limits to human languages? What things can happen? What things can’t happen? And I think one of the most interesting books that has been published in recent years is the book edited by Jan Wohlgemuth and Michael Cysouw, the Rara & Rarissima, a collection of papers about
24:28
things which appear to be very rare in the world’s languages and are therefore very exciting to us and interesting to us. I remember when I was starting to study linguistics in the 1960s, we were told that there were several different possibilities of word order in grammatical sentence structure. You know, you could start with the verb, you could start with the subject, you could put the, you know, you could have VSO, VOS, and so on, but we were also told that
24:57
there were no languages in the world which were object-initial languages: it just so happened that there were no languages which were OVS or OSV. And furthermore, it wasn’t ‘it just so happened’ — the conclusion was drawn that this was significant in some way, that, you know, the human language faculty didn’t produce such languages or that human interaction didn’t encourage such languages. And then, of course, we made the very exciting discovery that that in Hixkaryana and
25:26
other related languages in South America, you did get object-initial languages, and therefore, well, the human language faculty can produce those. Now, that was exciting and interesting. We weren’t othering those people except that we were noticing that, ‘Oh, yes, this is very different.’ It wasn’t a romantic thing. It was a hard-nosed scientific appreciation of this particular linguistic fact.
25:50
It pushed back the frontiers of knowledge; it increased our knowledge of what human languages can be like. And in studies of rara and rarissima, we’re discovering all sorts of, for us, exciting things, and the interesting thing is that many of these rarissima, these very rare things, very rare features, appear to be found in — I mean, they’re rare in the demographic sense in that there may only be one or two languages which have these features, and they tend to be languages we
26:18
don’t know about. And of course they’re languages we don’t know about, because if we’d known about them, we wouldn’t have regarded them as being rare. The practical implications of this are that we should study the languages of small communities before they die out. I mean, that’s the urgency of it all. We shouldn’t worry too much about things to do with the universal grammar or philosophy of
26:39
science. We need to go out and just discover these languages and describe these languages before they’re gone — because many of them, as you know better than me, sadly, are going — so that we can make sure that or do our best to make sure that we know what the limits of human languages are. JMc: If I can just defend the philosophy of science very briefly, couldn’t it be argued that if we don’t have some sort of appreciation for the philosophy of science and our potential unexamined
27:09
preconceptions, then we might unintentionally run out into the field naively and do things without thinking about the possible implications, you know, such as these potential problems of whether the categories we’re describing are real, you know, whether our observations are theory-laden, and whether we are in fact, you know, othering these communities that we’re describing? PT: Oh, yes, I mean, you’re absolutely right about that, so I wouldn’t deny that at all.
27:37
But I’m just saying that I think in linguistics we have a practical problem which is urgent. So many of the world’s languages, as I don’t have to tell you, are disappearing. These languages are providing us with information which, if we want to know about the limits of the human mind and the limits of the human language faculty, we need to investigate as soon as possible, so we shouldn’t spend too much time navel-gazing. Sorry, I don’t mean to suggest that philosophers
28:00
of science are doing that; we need them very badly. But we better get on with describing, with doing some field linguistics, with more field linguistics, while we still can. JMc: But, of course, languages, too, aren’t natural species. If anything, language is an activity that the speakers of the language participate in, so could there possibly be a question of why people — you know, why language shift is taking place,
28:26
why people are changing the language they’re speaking? Like maybe that’s a more important question than trying to go, you know, butterfly hunting and find the most exotic species. PT: Yeah, well, of course, I would object to the term ‘butterfly hunting’. It was something which is an accusation which was levelled at dialectologists. JMc: Yeah, yeah, I just wanted to make it maximally polemical. PT: Of course, I understand. I’m reacting in kind. People who do dialectology for example
28:56
do not — these days, anyway — do not, these days, just go around collecting data. They don’t say, ‘Oh, here’s our data. We’ll put it in the butterfly display case and leave it where it is.’ No. Of course, the data is gathered for a purpose. Of course, you don’t always know what the purpose is. A lot of these sciences are, you know, empirical. You start with the data, you see what it tells you, and you see what you get, and then you try and interpret it.
29:23
JMc: So that might be a good spot to leave it. PT: It’s a very interesting discussion. I thank you very much. JMc: Yeah, yeah, thank you very much. PT: Things to think about, and as you can tell, things which I haven’t ever thought about before.
Here to help you learn
Sign in to keep a private conversation with your course tutor.
Ask me anything about the material…
Sign in to chat