Talk:Linguistics/Archive 1

From Citizendium, the Citizens' Compendium
Jump to: navigation, search

First edits

I've now refreshed the page with the 16th December Wikipedia version and started hacking away at it. The old article is under my user page here if you want to examine it and revert things. Also, the page is now 'live'. John Stephenson 21:38, 15 December 2006 (CST)

For a non-expert bystander it reads fairly well David Tribe 05:45, 23 January 2007 (CST)
Thanks - though I've only edited the first few paragraphs. There really is a lot to do here, so anyone with an interest, dive in. John Stephenson 19:09, 23 January 2007 (CST)

Other Things

I know a few linguistics who might take issue with the statement that phonetics "complement linguistics rather than form a central component." I strongly urge putting phonetics out at the level of the other theoretical levels (as I did in my edit). --Joshua Tauberer 19:39, 3 February 2007 (CST)

First, well done for the recent edits - you're right to strip out the stuff about UG in the introduction, and also the prescription part is now in much better shape.
As for phonetics: in taking this line I'm following firstly the distinction between linguistics and phonetics (e.g. as Crystal's dictionary title does), and secondly the view that phonetics and phonology should not be conflated. As Aitchison argues in her (2003) introductory book, phonetics is about speech sounds, rather than the study of abstract patterns as most linguistics concerns itself with. Crystal's definition refers to phonetics as being about transcription, description and classification. This seems to be followed in much work:
  • The fact that you have books devoted to phonology rather than phonetics (e.g. Kenstowicz, 1994);
  • That there are books that treat them separately (e.g. Davenport and Hannahs, 2005);
  • The observation that phonology underlies many writing systems, sign languages (e.g. Brentari, 1999) and other communication systems in which phonetics plays no role;
  • That describing phonemic units in terms of articulation fails to capture essential behaviour and relationships (Davenport and Hannahs, 2005: 92-94);
  • That this distinction is taught to undergraduates (e.g. here);
  • That there are linguists who argue they're not related (e.g. David Odden and Jonathan Kaye);
  • That even the phonetics-based phonology approach (e.g. Hayes et al., 2004) leaves room for controversies (e.g. Harris and Lindsey in Burton-Roberts et al., 2000, who argue that phonology can't be described satisfactorily through either articulation or raw acoustics; and Hale and Reiss in the same volume, who see it as a branch of psychology).
To this we can add that phonological rules are largely independent of articulatory facts: for example, though I was taught by English teachers that assimilation occurs to make words easier to pronounce, in fact this is not universal (Russian lacks assimilation in words like bank) and applies to only some features and not others, e.g. place but not manner.
The late phonetician Peter Ladefoged certainly saw phonetic study as informing phonology, but in his well-known coursebook wrote that a phonetician's knowledge of language concerns how the vocal apparatus and auditory systems are used; their field is describing speech in terms of perception and production. Observing where a change in speech entails a change in meaning is helpful in doing this, but nowhere in his book does he actually get into a real, abstract phonological theory.
So for these reasons I can't see the justification for making phonetics as prominent as you would like; it deserves its own page, and relationships should be pointed out, but phonology does not really treat phonetics as its starting point.
  • Aitchison, J. (2003). Teach Yourself Linguistics. London: Hodder. 6th edition.
  • Brentari, D. (1999). A Prosodic Model of Sign Language Phonology. Cambridge, MA: MIT Press.
  • Crystal, D. (2002). A Dictionary of Linguistics and Phonetics. Oxford: Blackwell. 5th edition.
  • Davenport, M. & S.J. Hannahs (2005). Introducing Phonetics and Phonology. London: Hodder. 2nd edition.
  • Harris, J. & G. Lindsey (2000). Vowel patterns in mind and sound. In N. Burton-Roberts, P. Carr & G. Docherty (eds) Phonological Knowledge: Conceptual and Empirical Issues. Oxford: Oxford University Press. pp.185-205.
  • Hale, M. & C. Reiss (2000). Phonology as cognition. In N. Burton-Roberts, P. Carr & G. Docherty (eds) Phonological Knowledge: Conceptual and Empirical Issues. Oxford: Oxford University Press. pp.161-184.
  • Hayes, B., R. Kirchner & D. Steriade (eds) (2004). Phonetically-Based Phonology. Cambridge: Cambridge University Press.
  • Kenstowicz, M. (1994). Phonology in Generative Grammar. Oxford: Blackwell.
  • Ladefoged, P. (2006). A Course in Phonetics. London: Thomson. 5th edition.

Right... I didn't mean to suggest that phonetics should be promoted in a way that conflates it with phonology. They're separate for sure. What I'm saying is that it is a separate subfield, but it is a subfield of linguistics on par with the other levels of linguistic description (phono, morph, syn, sem, prag). Or, another way to put it is: just because it doesn't involve abstractions like modern syntactic theory shouldn't exclude it from being linguistics proper. --Joshua Tauberer 11:35, 7 February 2007 (CST)

References and subject-verb agreement

Comments from an interested amateur. On subject-verb agreement - could you add examples of dialect speech where this rule is broken, to make the example clearer? I guess "I sez-he sez" is one. On references - could I suggest please that you try if possible to choose references available on or via the web, or give IBSN numbers or database numbers. I think we need to see references as there to be used by the reader, not as remote but perhaps inaccessible sources of authority.Gareth Leng 04:52, 4 February 2007 (CST)

I'm stripping out some of the more obscure WP references; most of the ones I've used in the main article are fairly easy-to-find works. ISBN numbers, yes; working on it. :-) But why strip out the commas for the references? John Stephenson 23:08, 4 February 2007 (CST)
See Help:Citation_style; we've been trying to adopt a consistent reference style to make copy editing easier. This is provisional and for discussion, so please feel free to join in Gareth Leng 03:42, 5 February 2007 (CST)
OK; I think it looks a bit silly but no problem with going along with it. John Stephenson 21:58, 5 February 2007 (CST)


I changed the paragraph order. I think it is clearer to someone not acquainted with linguistics to proceed from some historical and philosophical considerations to more specific and technical linguistic themes.--Martin Kalck 05:37, 5 February 2007 (CST)

The problem with this is that the 'history of linguistics' is pretty peripheral to most modern students; I've done a BA and an MA and we never had courses on Ancient Greek philosophers for the simple reason that most study arises out of 20th-century approaches; yes, of course it draws on a lot of previous ideas, but what we study now is quite different from the comparative work of a century ago, and the writing of grammars on specific languages even further back. I think starting with the history rather than saying what linguistics is fundamentally about will confuse people. There's an article on the history stuff anyway; no need to further promote it. John Stephenson 21:58, 5 February 2007 (CST)
So, maybe we should skip it completely: I don't think that putting it at the end of the text makes it more accurate. But I think that is more relevant at the beginning of the text if we keep it. It may show how linguistics is different from older forms of language studies, while if we put it at the end, it has no connection with the rest of the article. The historical part is very short, and I think we can keep it as a brief introduction to what linguistics is by telling where it comes from. (In France, everybody is supposed to study a bit of philosophy of language during his last year in highschool but nobody studies linguistics before universtiy and many don't even know what it is about-well, I know that French education system is sometimes a bit peculiar)-Martin Kalck 02:32, 6 February 2007 (CST)
I think this plan is pretty clear: intro: what is modern linguistics; then how it differs from older approaches; then how language can be studied, then how it is studied. I don't think a list of linguistic "core areas" is very telling at the beginning of the article. Martin Kalck
Skipping most of it might not be a bad idea, as might more on how it differs from other language study. As I said above, I have little experience of 'history of linguistics' so am dubious about including all this Ancient Greeks stuff. However, in putting it towards the end we'd only be following the approach in many books: e.g. Aitchison's (2003) introductory work goes on about properties of language and the 'inner circles' (syntax etc.) before exploring the history - and even that largely ignores work before 1786. In other books I've seen there's not much of this stuff either. I just think that if I had no idea what linguistics was, I'd want a succinct definition and overview of the main fields, rather than being informed what people believed in the past, even though it is relevant for those wishing to know more. John Stephenson 03:58, 6 February 2007 (CST)
I do agree that references to the Greeks are unnecessary. I just feel somewhat uneasy letting people think that linguistics emerged out from nothing, while it was a central problem to some philosophers. I also removed the claim that modern linguistics was specific by trying to establish unviersal laws: this makes a difference with 19th century linguists but not necessarily with earlier philosophers (e.g Leibniz, or Anrault and Nicole) Martin Kalck 05:53, 6 February 2007 (CST)----
IMO, the introduction should immediately be followed by the core areas and then the other areas (which I think should be significantly expanded on). These are crucial for understanding what linguistics is about because the non-linguists will have absolutely no idea how complex language is and why it's worth studying. Everything else that comes before it (prescriptivism, speech vs writing) is not interesting unless you know that there is something *in* the field to be studied. --Joshua Tauberer 12:23, 7 February 2007 (CST)
On another hand, the first thing to know about linguistics is that it is descriptive. It needs to be clearly stated in the intro, and I believe it is. Then, you have to decide whether it is worth more development in the main article or not. If it is, it would be logical to put it toward the beginning: it is a very general topic that determines the nature of linguistics; and I think we should proceed from the general to the particular, i.e from something general about linguistics to a description of its various areas. YEt another solution may be to put the part on prescription in the "applied linguistics" section since it is explained that some prescriptions are "linguistically founded" and some are not. -As for me, I tend to think that we could do without an entire part on prescription in the main article, but I admit this is a purely personal view.
About the historical part, I suppose that Aitchinson (and other authors) didn't want to put off their readers who were not interested in history, but wanted nonetheless to write a short historical part for those who were, so they put it at the end, in an annex-like position. But cititzendium does not need such annexes: it can contain as many articles as we can write. So if we think something does not need to be put in the core of the article, we can move it to another article. (though again I think a very short historical paragraph at the beginning would not be a bad idea).
Anyway, I know I'm not the most competent on the subject here, and I will conform to your choices. --Martin Kalck 16:58, 7 February 2007 (CST)

"Instinct to communicate"

I don't think this is quite right: "Theoretical linguists are concerned with questions about the apparent human 'instinct' to communicate,[1]". I haven't read Pinker, so the reference may be correct, but my interpretation of the phrase "the language instinct" has always been not to take it in the sense that child birds will fly instinctually -- i.e. without being prompted to do so -- because children raised in isolation obviously don't just start speaking English, but rather in the sense that there is something deep, unconscious, and possibly native about language. I don't think that's expressed by "instinct to communicate." As a theoretical linguist in-training, I'm not concerned with something apparent, psychological/biological, or native (for that matter). --Joshua Tauberer 06:22, 10 February 2007 (CST)

Pinker doesn't say that language is biologically determined, either; input is needed as a trigger and subsequently. Maybe change the analogy to bird flight? John Stephenson 03:43, 12 February 2007 (CST)
Just changed it to 'bird song'. :-) John Stephenson 03:44, 12 February 2007 (CST)
Further to this, I've just flicked through The Language Instinct, and in the opening section there's an interesting analogy between language and spiders' webs. Perhaps it would be worthwhile to mention that if a spider isn't in the right environment (trees etc.) it'll never spin a web. But this doesn't mean that webspinning isn't innate. Also, Pinker makes the point that all prelingual infants babble, which suggests the system at work in thr absence of any experience. John Stephenson 22:24, 17 February 2007 (CST)

"Fields of linguistics"

I find it quite odd to call a part "fields of linguistics" and another "core areas". Aren't core areas fields of linguistics too? --Martin Kalck 03:37, 12 February 2007 (CST)

I thought about something like 'disciplines' for the latter; maybe a bit too high-brow? John Stephenson 03:43, 12 February 2007 (CST)
Mark Liberman at Penn has used "levels of linguistic description" versus "goals" of linguistics. --Joshua Tauberer 05:55, 12 February 2007 (CST)

See also

[Deleted from the article and archived here in case some links need to be restored; I've done this to tidy up the page. John Stephenson 22:16, 21 February 2007 (CST)]


Do you think it's worth illustrating the linguistics page with something like this? (This is just a test version for illustrative purposes, so it's rubbish.) John Stephenson 00:46, 22 February 2007 (CST)

The image has now been replaced. John Stephenson 07:43, 19 August 2008 (CDT)
Levels of linguistic knowledge involved in producing the utterance 'the cats'.

Properties of language (archiving material here

Deleted this:

It has been understood since the time of the ancient Greeks that languages tend to be organized around grammatical categories such as noun and verb, nominative and accusative, or present and past. The vocabulary and grammar of a language are organized around these categories.

In addition, language organizes elements into recursive structures; this allows, for example, a noun phrase to contain another noun phrase (as in the chimpanzee's lips) or a clause to contain a clause (as in I think that it's raining). Though recursion in grammar was implicitly recognized much earlier (for example by Otto Jespersen), the importance of this aspect of language was only fully realized after the 1957 publication of Noam Chomsky's book Syntactic Structures,[1] which presented a formal grammar of a fragment of English. Prior to this, the most detailed descriptions of linguistic systems were of phonological or morphological systems, which, especially in English, tend to be closed and admit little creativity.

Chomsky used a context-free grammar augmented with transformations. Since then, context-free grammars have been written for substantial fragments of various languages (for example GPSG, for English), but it has been demonstrated that human languages include cross-serial dependencies, which cannot be handled adequately by Context-free grammars. This requires increased power, for example transformations.

An example of a natural-language clause involving a cross-serial dependency is the Dutch[2][3]

Ik denk dat Jan Piet de kinderen zag helpen zwemmen
I think that Jan Piet the children saw help swim
'I think that Jan saw Piet help the children swim'

The important point is that the noun phrases before the verb cluster (Jan, Piet, de kinderen) are identified with the verbs in the verb cluster (zag, helpen, zwemmen) in left-right order.

This means that natural language formalisms must be relatively powerful in terms of generative capacity, which is to say that they must be able to account for very complex word orders and relations between the words. On the other hand, formalisms must not be too powerful so as to predict word orders that do not occur in any language. Some of the commonly studied formalisms include LFG, HPSG, Minimalism, Tree Adjoining Grammar, and Categorial Grammar. The formalisms all tend to be too weak or too powerful in various ways, and an important concern for theoretical syntax is to find how to constrain the formalisms to match the languages that we see.
  1. Chomsky, Noam. 1957. "Syntactic Structures". Mouton, the Hague.
  2. Bresnan, Joan, Ronald Kaplan, Stanley Peters, and Annie Zaenen. 1982. Cross-serial dependencies in Dutch. Linguistic Inquiry 13:613-636.
  3. Shieber, Stuart. 1985. Evidence against the context-freeness of natural language. Linguistics and Philosophy 8:333-344.