Descriptivism, prescriptivism and the evolution of language

General discussion about learning languages
galaxyrocker
Blue Belt
Posts: 516
Joined: Mon Jul 20, 2015 12:44 am
Languages: English (N), Irish (Teastas Eorpach na Gaeilge B2)
Language Log: viewtopic.php?f=15&t=757
x 881

Re: Descriptivism, prescriptivism and the evolution of language

Postby galaxyrocker » Sun Dec 18, 2016 3:34 am

Serpent wrote:Modern linguistics yes, but previously there have certainly been linguists who considered Latin or Sanskrit to be superior to other IE languages.


Finny wrote:...not to mention the sordid history of bilingualism research, and how until only a few decades ago, (primarily monolingual) researchers categorized it as a deficit in children and invented an endless array of restrictive qualifications to be truly "bilingual" (many of which persist to this day in language learning communities online).


That is true, but I definitely don't see it as a switch to embracing "moral relativism"; working in a descriptive framework is no more morally relativistic than working in one in any other science. I just see it as a shift to a more science-based discipline, rooted in the fact of how people actually use language and not our preconceived notations of it. Part of me feels it wouldn't be too far off to credit Chomsky for this, though I don't know enough about the early history of modern linguistics to speak definitively (the 50s were well before my time!)
1 x

User avatar
Iversen
Brown Belt
Posts: 1168
Joined: Sun Jul 19, 2015 7:36 pm
Location: Denmark
Languages: Monolingual travels in Danish, English, German, Dutch, Swedish, French, Portuguese, Spanish, Catalan, Italian, Romanian,
Ahem, not yet: Esperanto, Norwegian, Afrikaans, Platt, Scots, Russian, Serbian, Bulgarian, Albanian, Greek, Latin, Indonesian ...
Language Log: viewtopic.php?f=15&t=1027
x 2171

Re: Descriptivism, prescriptivism and the evolution of language

Postby Iversen » Sun Dec 18, 2016 10:57 am

Actually the generation before Chomsky (Bloomfield & co.) were more purebred descriptivists - and also more interested in other languages than the one they could speak natively themselves. Chomsky invented the criterion for a grammatical utterance that it is OK if a native speakers says it is OK - and that is definitely less prescriptivist than a criterion that says only members of an official academy or those who write grammars or your teacher is qualified to judge whether and utterance is OK or not - but Chomsky's criterion still says that the opinion of any native speaker is more valid than the opinion of any second language learner, including those who have progressed beyond C2.

In Danish we have constructions with "fordi at" (roughly 'because that'). Everybody regularly use them (including those who firmly deny it), they have been used since Old Norse, a similar construction is still accepted in Icelandic, and the construction with "at" is totally logical if you break the construction down into its constituent parts: the -"di" is actually an old demonstrative, so the construction actually means "for the (reason) that" - and in such constructions the "that" is the norm, while the omission is preciusely that - and omission. Nevertheless generation of teachers in Danish schools have being chasing users of this perfectly legitimate construction simply because somebody once have established an erroneous pedagogical tradition. Similar things have of course happened in many languages (like the interdiction of 'hanging' prepositions or split infinitives in English).

We can't totally skip prescriptivism - parents and other savvy persons need to teach a language 'as it is' because the alternative would be total and utter chaos. But when a pedagogical tradition starts living its own life in spite of the linguistic (i.e. descriptive) facts then it has become a repression mechanism rather than simply teaching somebody how to use the language. But there is still a case to be made for moulding learners against the realities of a language, namely that it makes them more suited to live in an evil and unforgiving world where some speech forms are considered 'good' and others 'bad'. In its extreme form this line of thought will lead to wholesale killing of other languages and dialects within an territory - like France, where a bunch of mouldy old former presidents have decided that the country is entitled to ignore the EU regulations that protect minority languages. And for some that repression is a good thing because they see linguistic uniformity as the optimal state (and want to protect the citizens from running into trouble with people like themselves), but for those of us who really like linguistic diversity it is just one step down from the KZ camps. Well, both sides may have a point...

The simple fact today is that minority languages and dialects are being wiped out an alarming rate - not so much because of the academies and school teachers as a result of the influence from the mass media - but on the micro level our individual ways of speaking have probably become less easy to control because the hardcore purists have become less revered and respected.
2 x

Cainntear
Blue Belt
Posts: 698
Joined: Thu Jul 30, 2015 11:04 am
Location: Scotland
Languages: English(N)
Advanced: French,Spanish, Scottish Gaelic
Intermediate: Italian, Catalan, Corsican
Basic: Welsh
Dabbling: Polish, Russian etc
x 1317
Contact:

Re: Descriptivism, prescriptivism and the evolution of language

Postby Cainntear » Sun Dec 18, 2016 12:27 pm

Speakeasy wrote:
Cainntear wrote: ... The western association of the term "ayatollah" with authoritarian government via a mere two of the many scholars referred to by the term is really rather ignorant of the meaning of the term within Shia society.
You are beating a dead horse ... and you know it.

Nevertheless, despite the purported ignorance of Western Society vis-à-vis the true significance of the term "Ayatollah", as you are well aware, in Western Popular Culture, it has become synonymous with an unrelentingly harsh, authoritarian and despotic regime.

So what? It's OK to spread ignorance simply because everyone else is doing it? At some point someone's got to be the better person.

Besides, you still went out of your way to insult me and Serpent, and then proceeded to object to my "woah there" as being "uncalled for" and showing "disrespect".

Then again, perhaps you really would have preferred that my (offensive to you) post read "Fascists of Right Thought." Whatever, my point concerning the inherent danger of "right thought" stands.

That would be a ridiculous comparison. You do know that Mussolini's party was extraordinarily rigid in their views of correct speech, right? A fundamental pillar of fascism is national unity and uniformity, and the fascists were hugely prescriptivist in their views on language, even going so far as to declare Maltese as a mere "dialect", which is just utterly ridiculous.

Tomás wrote:This thread reminds me of the strong version of moral relativism that took over cultural anthropology several decades ago, with only minor pushback. Have linguists fought this out amongst themselves, or is the relativistic position really that hegemonic?

This has nothing in common with moral relativism.
Language changes all the time, we can see that.
Prescriptivism is an arbitrary choice by a minority of people -- it's authoritarianism and can only be equated to absolute morality if your point is that absolute morality is arbitrary and oppressive.
If you believe in an absolute morality given by a deity, then it is entirely different from language, because we know that the language we speak today is massively different from the language of the garden of Eden, or the language of Elijah, or Jesus, or Mohammed.
6 x
A year of Tatoeba recordings: 40 / 365 One donated recording every day in 2017.

galaxyrocker
Blue Belt
Posts: 516
Joined: Mon Jul 20, 2015 12:44 am
Languages: English (N), Irish (Teastas Eorpach na Gaeilge B2)
Language Log: viewtopic.php?f=15&t=757
x 881

Re: Descriptivism, prescriptivism and the evolution of language

Postby galaxyrocker » Sun Dec 18, 2016 2:48 pm

Iversen wrote:Actually the generation before Chomsky (Bloomfield & co.) were more purebred descriptivists - and also more interested in other languages than the one they could speak natively themselves. Chomsky invented the criterion for a grammatical utterance that it is OK if a native speakers says it is OK - and that is definitely less prescriptivist than a criterion that says only members of an official academy or those who write grammars or your teacher is qualified to judge whether and utterance is OK or not - but Chomsky's criterion still says that the opinion of any native speaker is more valid than the opinion of any second language learner, including those who have progressed beyond C2.


I agree with Chomsky on that point. I don't know why one would use a non-native speaker to study what is natural in a different language, regardless of their level. They haven't fully been immersed in the language, and don't necessarily have the ingrained grammar that native speakers do. It'd be like studying tigers to explain the behavior of lions. Not saying they can't be studied at all; there's several ways I could think of studying highly fluent non-native speakers, but I wouldn't study them to say what's valid in their non-native language, especially since they generally learn it in a completely different way, and often focus solely on the standard.


We can't totally skip prescriptivism - parents and other savvy persons need to teach a language 'as it is' because the alternative would be total and utter chaos. But when a pedagogical tradition starts living its own life in spite of the linguistic (i.e. descriptive) facts then it has become a repression mechanism rather than simply teaching somebody how to use the language. But there is still a case to be made for moulding learners against the realities of a language, namely that it makes them more suited to live in an evil and unforgiving world where some speech forms are considered 'good' and others 'bad'. In its extreme form this line of thought will lead to wholesale killing of other languages and dialects within an territory - like France, where a bunch of mouldy old former presidents have decided that the country is entitled to ignore the EU regulations that protect minority languages. And for some that repression is a good thing because they see linguistic uniformity as the optimal state (and want to protect the citizens from running into trouble with people like themselves), but for those of us who really like linguistic diversity it is just one step down from the KZ camps. Well, both sides may have a point...


I think what needs to be taught is the concept of register. Certain dialects/languages are acceptable in certain situations, but you need to use these other ones in other situations. This, along with the fact that anyone can be billingual, I think would really go a long way to getting over linguistic prejudice.
3 x

User avatar
Iversen
Brown Belt
Posts: 1168
Joined: Sun Jul 19, 2015 7:36 pm
Location: Denmark
Languages: Monolingual travels in Danish, English, German, Dutch, Swedish, French, Portuguese, Spanish, Catalan, Italian, Romanian,
Ahem, not yet: Esperanto, Norwegian, Afrikaans, Platt, Scots, Russian, Serbian, Bulgarian, Albanian, Greek, Latin, Indonesian ...
Language Log: viewtopic.php?f=15&t=1027
x 2171

Re: Descriptivism, prescriptivism and the evolution of language

Postby Iversen » Tue Dec 20, 2016 1:15 am

I don't say that Chomsky's criterion is bad, but it doesn't solve all problems. Native speakers don't always agree (we have seen examples of that in discussions both here and at HTLAL), and some native speakers flatly deny that they ever say certain things which they demonstrqably do say - their stated judgment could actually be a result of things they have been taught in school rathr than an ideal chomskyan intution. And on the other hand there is a alternative to ask native speakers, namely collecting an insane amount of examples (preferably both written and oral, but admittedly most collections are based on the written sources) - with large digitized corpora this has become easier than it was just a few generations ago.

If a certain expression is used in a number of texts by native speakers or writers then I would accept it even if one specific native speaker claimed it was wrong. And since some of the best grammars have been written by wellinformed foreigners who have studied the actual use of their target language in detail for years on end. I would not simply disregard the conclusions (and intuitions) of such persons even if a certain native speaker disagreed. However in such cases it ought be possible to find out when and where the expression or construction under scrutiny actually has been used in practice - and the examples may show that it only is used in older literature or in a certain area or whatever.

In short, Chomsky's reliance on native speakers is a good and efficient (and fast!) way of checking grammaticality, but it shouldn't be seen as infallible or the only allowed test method in the world.
1 x

galaxyrocker
Blue Belt
Posts: 516
Joined: Mon Jul 20, 2015 12:44 am
Languages: English (N), Irish (Teastas Eorpach na Gaeilge B2)
Language Log: viewtopic.php?f=15&t=757
x 881

Re: Descriptivism, prescriptivism and the evolution of language

Postby galaxyrocker » Tue Dec 20, 2016 2:41 am

Iversen wrote:I don't say that Chomsky's criterion is bad, but it doesn't solve all problems. Native speakers don't always agree (we have seen examples of that in discussions both here and at HTLAL), and some native speakers flatly deny that they ever say certain things which they demonstrqably do say - their stated judgment could actually be a result of things they have been taught in school rathr than an ideal chomskyan intution. And on the other hand there is a alternative to ask native speakers, namely collecting an insane amount of examples (preferable bot written and oral, but admittedly most collections are based on the written sources) - with large digitzed corpora this has become easier than it was just a few generations ago.


But the thing is why should we expect all native speakers to agree on the same thing? Each individual person has their own individual idiolect, and many idiolects form together to make a dialect region. Technically, no two speakers have the same internal grammar. So the fact that native speakers disagree isn't a flaw of descriptivism, it's a feature of language change and how normal language works.

I also disagree with the necessity of a written corpus, because writing tends to follow certain standardized rules that are rarely used in speech Especially formal writing.


If a certain expression is used in a number of texts by antive speakers or writers then I would accept it even if one specific native speaker claimed it was wrong. And since some of the best grammars have been written by wellinformed foreigners who have studied the actual use of their target language in detail for years on end. I would not simply disregard the conclusions (and intuitions) of such persons even if a certain native speaker disagreed. However in such cases it ought be possible to find out when and where the expression or construction under scrutiny actually has been used in practice - and the examples may show that it only is used in older literature or in a certain area or whatever.

In short, Chomsky's reliance on native speakers is a good and efficient (and fast!) way of checking grammaticality, but it shouldn't be seen as infallible or the only allowed test method in the world.



Again, you seem to assume that all native speakers should speak the same, and find the same things as being grammatical, when this is just not true. Native speakers find things in their own dialect grammatical, but not necessarily in others. There's speakers in America who find it perfectly acceptable to say things like "The car needs washed", but that doesn't work in my dialect. It doesn't mean either of us are "wrong", just that we have different grammatical rules -- something which should point out the lunacy of saying there is one "true" English.

And, I would disregard those persons if they were trying to speak of a dialect over someone who speaks that dialect. Or call someone wrong because they speak a dialect. That's like a non-native speaker telling me I'm wrong for saying "I don't need no x" because it's part of my dialect, but not standardized. In reality, when it comes to determining what is correct and acceptable in a language, we need to rely on the native speakers, not on learners.
2 x

User avatar
aokoye
Blue Belt
Posts: 901
Joined: Sat Jul 18, 2015 6:14 pm
Location: Portland, OR
Languages: English (N), German (B2), Swedish (beginner), Dutch (beginner), French (beginner)
Language Log: viewtopic.php?f=15&t=2935
x 1270
Contact:

Re: Descriptivism, prescriptivism and the evolution of language

Postby aokoye » Tue Dec 20, 2016 5:26 am

One of the many useful things about corpora - they show the multitude of ways people actually use language. While I haven't taken a corpus linguistics class, I have taken a class taught by someone who teaches corpus ling and the grammar book we used was corpus based (the one that I referenced earlier in this thread).

Iverson - what do you mean by "preferable bot written and oral"? I can't imagine there are good enough speech to text programs that would allow an accurate corpus based on spoken language that wasn't seriously gone over with a fine tooth comb by human editors. I might be misunderstanding you though.

That said it looks like COCA (the Corpus of Contemporary American English) was complied completely from transcripts. Meanwhile in regards to the spoken data in Longman Grammar of Spoken and Written English, if it came from transcripts it looks like at least some of the transcripts were hand written/typed as opposed to automatically created from a speech to text generator. I do think corpora based on written sources are useful because, at least in English, different genres and registers of writing do tend to be different in terms of types of specific grammatical features. Writing a linguistics paper is different than writing an ecology paper, is different than writing a young adult book, is different than...

I also agree with Galaxyrocker - not every native speaker is going to agree that a specific sentence is grammatical (grammatically correct). What is correct in my dialect might not be in another person's dialect. That difference doesn't mean that either dialect is wrong. I think that is one of the many useful things about corpora.
1 x
Prefered gender pronouns: Masculine

User avatar
sfuqua
Blue Belt
Posts: 742
Joined: Sun Jul 19, 2015 5:05 am
Location: san jose, california
Languages: English(N),
Samoan(FSI 4+, rusty),
Tagalog(talk at home, read newspapers, watch TV),
Spanish(chat online, read newspapers and books, watch TV)
French(study)
Language Log: http://how-to-learn-any-language.org/vi ... f=15&t=772
x 1515

Re: Descriptivism, prescriptivism and the evolution of language

Postby sfuqua » Tue Dec 20, 2016 5:51 am

It's always interesting to see how quickly people fall into the idea that there are some forms of a language that arre "superior" to others.
When you look at such distinctions, it always almost goes back to some evaluation that some people are superior to others. This isn't a completely racist (or whateverist) idea. It is a common idea that if you want to be accepted by a certain group, it will help you if you talk like them. There are the classic exceptions to this rule, English split infinitives, dangling prepositions and other things.

I wonder if anywhere in England, after a couple of beers, there wasn't somebody who didn't think that a preposition wasn't a good thing to end a sentence with...
1 x
10K: 8226 / 10000 cards
ES 10000: 5469 / 10000 pages
ES 250: 2500 / 15000 hours
FR 10000: 427 / 10000 pages
FR 250: 294 / 15000 hours

Cainntear
Blue Belt
Posts: 698
Joined: Thu Jul 30, 2015 11:04 am
Location: Scotland
Languages: English(N)
Advanced: French,Spanish, Scottish Gaelic
Intermediate: Italian, Catalan, Corsican
Basic: Welsh
Dabbling: Polish, Russian etc
x 1317
Contact:

Re: Descriptivism, prescriptivism and the evolution of language

Postby Cainntear » Tue Dec 20, 2016 10:53 am

Iversen wrote:I don't say that Chomsky's criterion is bad, but it doesn't solve all problems. Native speakers don't always agree (we have seen examples of that in discussions both here and at HTLAL), and some native speakers flatly deny that they ever say certain things which they demonstrqably do say - their stated judgment could actually be a result of things they have been taught in school rathr than an ideal chomskyan intution. And on the other hand there is a alternative to ask native speakers, namely collecting an insane amount of examples (preferable bot written and oral, but admittedly most collections are based on the written sources) - with large digitzed corpora this has become easier than it was just a few generations ago.

If a certain expression is used in a number of texts by antive speakers or writers then I would accept it even if one specific native speaker claimed it was wrong. And since some of the best grammars have been written by wellinformed foreigners who have studied the actual use of their target language in detail for years on end. I would not simply disregard the conclusions (and intuitions) of such persons even if a certain native speaker disagreed.

Which is all pretty much the point of descriptivism.

But Chomsky was correct in saying that native speakers have an intuitive recognition of grammatical vs ungrammatical utterances. One part of the Chomskian way of thinking that I think you're glossing over here is the idea that people recognise grammaticality without consciously knowing why -- the effective result of which is that anyone who can give you a reason why a sentence is correct isn't really a useful informant, because they are relying on conscious learned knowledge, not intuition.

If corpus linguistics had never been invented, we would now be living in a golden age of Chomskian grammaticality investigations of English, where eyetrackers and MEG and EEG headsets would be used to log native English speakers' unconscious reactions and reaction times to a vast stream of sentences on computer screens, and we'd build up our descriptivist grammars by harnessing the minds of countless natives.

As it is, corpus linguistics has given us a much more efficient alternative.

Your point about non-native-written grammars is perfectly valid -- historically, it may well be that foreigners are more likely to be descriptivist in their outlook as they are consciously attempting to gain access to another culture. With a corpus-based grammar, the compiler's native language is almost completely inconsequential.

However, I think the point earlier in the thread was that everything has to be built on native informants and their unconscious speech. Whether the author of a grammar is native or non-native, accurate rules can only be determined by studying the speech of other native speakers.
1 x
A year of Tatoeba recordings: 40 / 365 One donated recording every day in 2017.

User avatar
Iversen
Brown Belt
Posts: 1168
Joined: Sun Jul 19, 2015 7:36 pm
Location: Denmark
Languages: Monolingual travels in Danish, English, German, Dutch, Swedish, French, Portuguese, Spanish, Catalan, Italian, Romanian,
Ahem, not yet: Esperanto, Norwegian, Afrikaans, Platt, Scots, Russian, Serbian, Bulgarian, Albanian, Greek, Latin, Indonesian ...
Language Log: viewtopic.php?f=15&t=1027
x 2171

Re: Descriptivism, prescriptivism and the evolution of language

Postby Iversen » Tue Dec 20, 2016 2:26 pm

galaxyrocker wrote:But the thing is why should we expect all native speakers to agree on the same thing? (...) I also disagree with the necessity of a written corpus, because writing tends to follow certain standardized rules that are rarely used in speech Especially formal writing.


The good thing about corpora is exactly that they almost certainly contain things which any one native speaker wouldn't have expected - and still may not accept after having seen that others use the language in other ways than the person does him/herself. The flaw in accepting the native intution of any native speaker as your one and only guide is precisely that you'll miss some of the things that actually are in use.

Demanding that the grammarians use corpora based both on written and spoken sources has the same purpose: since the written language probably differs from the spoken one (more or less depending on the language) insisting on having spoken corpora too is necessary if you want to learn about all sides of the language. Unfortunately it is more difficult to get good and large spoken corpora because you need a human being with a solid knowledge (and almost certainly a native speaker) to go through the recorded utterances one by one, but that only means that we should cherish the collections that exist and laud their creators even more.

Please also note that I wrote: "If a certain expression is used in a number of texts by native speakers or writers then I would accept it even if one specific native speaker claimed it was wrong." So the corpora should be based exclusively on utterances by native speakers/writers, but this does not mean that only native speakers/writers are capable of analyzing the collections, and when I consider all the grammars and monographs I have read in my life I can't really see that those written by native speakers are better than those written by good foreign learners.

Those written by and for native speakers may actually be less relevant for foreign learners because they tend to skip over phenomena which the writer finds totally natural (and therefore not worthy of consideration), but which may be extremely relevant for learners. On top of that, a lifetime spent at looking at concrete utterances by native speakers also gives you an intuition of sorts about a language. And possibly an intuition which is more relevant for writing grammars since it has been developed by looking at and analyzing utterances within a grammatical framework, whereas the intuitions of a native user without any linguistic training typically just will tell you whether a certain formulation is OK or not.

So the use of corpora should definitely not be ditched -not even by grammarians who also happen to be native speakers themselves. Actually I would be more suspicious of a native linguist who chose only to rely on his/her own intuitions about a language. After all, that's what the truly prescriptivist ones do - they believe in their own intuitions as the gospel..
3 x


Return to “General Language Discussion”

Who is online

Users browsing this forum: Amerykanka and 6 guests