Talk:Language/Archive 1

This is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.

Archive 1

Archive 2

Archive 3

Archive 4

For transcription used by some authors, check out http://www.phon.ucl.ac.uk/home/sampa/home.htm---- I call that opinion. Accurate according to whom. Yes the IPA is accurate, and perhaps there should be a link to it here. But, Samba is surely counter-intuitive and hardly the only choice for transcription or even computer transcription.

The advantage of SAMPA is that it's very close to IPA (you'll most certainly find a link to IPA on the SAMPA page). And BEING COUNTER-INTUITIVE is not a disadvantage at all, since IPA is also counter-intuitive - especially to speakers of English. Linguistics is very often counter-intuitive, and that probably applies to science in general. I don't think there is a more scientific and more computer-friendly system out there. Of course, /u/ instead of <oo> is counter-intuitive to speakers of English, of course /S/ instead of <sh> seems strange, but only because <sh> is the grapheme you learn in school. In linguistics (especially phonetics and phonology), it doesn't really make sense to write <sh> instead of /S/. Of course those are all opinion, since there is no TRUTH in general. Of course I think it would be agood idea to use SAMPA throughout, but it's maybe better to include a transcription closer to English PLUS a more scientific transcription. Although SAMPA is not THAT hard to learn, I don't wanna force anyone to do so.---- Yes- that is 80% opinion at least. Further, this is a site where anyone can use whatever transcription system they like, and even, though it would be rude, anyone change someone else's transcription system. In my opinion, any transcription system that includes numbers rather than letters is more counter-intuitive than one that tries to confine itself to relatively familiar alphabets. Furthermore, British English speakers are brought up on the IPA, due to the influence of the OED, unlike Americans, and find it quite natural.

--- yeah, sure, I'd prefer to write in IPA, though that's not possible for technical reasons now is it? Sure, people with linguistic training from the UK feel that IPA is quite natural, but even IPA transcriptions may vary. Anyway... I don't think it would be a good idea to exclude SAMPA or IPA transcriptions. SAMPA only has SOME numbers, it's mainly or as far as possible it's identical with IPA.

--- I'll include ORTHOGRAPHY in the LINGUISTICS SECTION, because orthography is not language per se, but it is part of linguistics.

The subject of how to indicate pronunciations on Wiki pages is a complicated one, and simple solutions to complicated problems are frequently a bad idea. For many uses, something very simple and English-centric like the system used in the Jargon File would be adequate. But it's clearly not adequate for many other uses here, particularly when describing foreign languages. Using more than one system has its own drawbacks, and is more complicated overall than using one. Yes, SAMPA (or a competing IPA-based system such as the one used on sci.lang) is probably overkill and difficult to learn for many; but the technology here can be arranged to make it simpler. For example, the server already recognizes "ISBN XXX..." in Wiki text and automatically makes links to more detailed information about a book. It could do a similar thing with "SAMPA XXX...", automatically creating a link to a page explaining SAMPA, perhaps even a sound clip. It is possible that in the future, it could display real IPA symbols when Unicode fonts become more available. Most important, though, is that IPA-based information is information-preserving while simpler systems might not be, so we can add those features without needing more information. For that reason alone, I think it is important that we adopt some IPA-based standard. --LDC

yeah... dunno about the jargon file... the question is: is it used in linguistics?

No, it would be wholly inadequate for that; it's just a simple ad-hoc. I only mention it to say that it would br appropriate to use in some contexts (notably the jargon file itself, which describes jargon to English-speakers who aren't linguists). Surely we will need a more complete IPA-based system for linguistics articles in general, but that doesn't rule out the use of simpler methods for other uses. --LDC

This article says programming languages lack "discreteness". What does that mean? -LC

No answer for the last 4 months. I've removed the word. If anyone wants to put it back, please define it. -LC

Why does the introduction the concept 'language' START with machine language rather than with the human phenonmenon? It makes for a clever paragraph, but really now! --MichaelTinkler

I have removed the sentence that says that language is a technology. Just read the definiton in the technology entry. Calypso

I'm removing this, which has no place here until someone can provide a more in-depth description of both the nature of animal languages and which definitions of "language" they don't live up to:

Most often, the term refers to vocal languages. Animal "languages" can be quite complex; but they do not satisfy the criteria linguists use to define a true language. A language must have a full grammar and be able to be used in any context to express any idea. Among the animals, humans alone have been proven to have a true language instead of a simple means of communicating a limited set of things such as danger, readiness to mate, or location/proximity.

--Ryguasu 07:14 Jan 9, 2003 (UTC)

i wanna to know about the theoretically base for verbal memorising (in China, we use the word "recite")in EFL/ESL for middle school students. 1. Does "verbal memorising" equally mean " rote memorising", if not, why? 2. If "verbal memorising" is effective, then , how can it benefits the future real communication? 3. If "verbal memorising" isn't effective, and the immersion method (eg,in Canada)is still far to be proven as effective, in countries such as China, Japan where English input is badly not enough, what can we do?

Thanks for any suggestion, more detailed information is esp. welcome.
Email: pan_nelson@21cn.com

Mathematics and computer science use artificial entities called formal languages (including programming languages)

Is it actually correct to say that a programming language is a kind of formal language? I admit that there is a set S of syntactically valid programs for a given language L. Now, following Formal language, we could say that L is nothing more than S. However, this seems misleading, because L has not only a syntax by a semantics. Thus I'm confused if someone is seriously claiming that programming languages are formal languages, no more, no less. --Ryguasu 04:18 Apr 3, 2003 (UTC)

Hey VeryVerily, have you ever learned Esperanto, that you can judge, if and how good and easy Esperanto works? How many other languages do you speak? Tell me why you found the remark about Esperantto had to be removed? Give me reasons!!! --Halsbandsittich from Wikipedia deutsch

The text says: Chomsky (1986) points out that "some dialects of German are very close to dialects that we call 'Dutch' and are not mutually intelligible with others that we call 'German'". ' Hahaha this looks very funny. It is like citing George Washington who says that 'it is rainy today' That fact does not make it more noteworthy because George Washington himself said it, does it? How many people may have "pointed it out" before? :-))) Flyingbird 23:41, 18 Dec 2003 (UTC)

How about adding context-free and context-sensitive languages and introduce the concept of context? That way we could ellaborate on context in linguistics and include the word meaning in the categories, which I have tried in the sandbox in the magyar wikipedia meeting with the disgust of the moderator. Apogr 10:46, 9 Jan 2004 (UTC)

Introduction

I find the term "self-appointed linguists" to be somewhat pejorative. Moreover, I find it highly controversial to claim that Saussure created linguistics as an academic discipline. You might just as well claim that the study of linguistics dates back to antiquity. It is a bit like claiming that Peano created mathematics as an academic discipline.

This comment is unsigned and undated. I assume it refers to the opening paragraph of an earlier version of the article. The opening paragraph now reads:

Language is a system of finite arbitrary symbols combined according to rules of grammar for the purpose of communication. Individual languages use sounds, gestures and other symbols to represent objects, concepts, emotions, ideas, and thoughts.

I [ David Pierce ] am going to change this paragraph on a number of grounds:

Saying "language is a system" suggests that there are other systems; is this the intended meaning? I suppose not.
I don't understand this; the sentence was perfectly clear (of course there are other systems; mathematics and logic, for example). --Mel Etitis (Μελ Ετητης) 14:14, 17 August 2005 (UTC)
Do mean to suggest that mathematics and logic are also systems of finite arbitrary symbols combined according to rules of grammar for the purpose of communication? If it is so, then the present article should say so: "Language is one of the systems...Other such systems are mathematics and logic". But I think this formulation misleadingly suggests that language, mathematics, and logic are on the same level. It seems to me rather that logic is a formalization of certain aspects of language, and that mathematics makes use of this formalization. David Pierce 08:21, 18 August 2005 (UTC)

You're reading the scope rather oddly, I think: languages, maths, logic, etc., are systems of arbitrary symbols; the first of these are combined... --Mel Etitis (Μελ Ετητης) 09:06, 18 August 2005 (UTC)
I simply do not understand your criticism here. Are you saying that you would define maths as "a system of arbitrary symbols..."? That's not how Wikipedia defines maths, and it's not how I would define maths. (In fact I am a professional mathematician and logician, although this does not mean that I can define what I work on. Still, Euclid and other ancients, at least, did maths in ordinary language, without special symbols.) By the current Wikipedia definition, language is a system of finite[ly many] arbitrary symbols..." You seem to agree that this means that there are other such systems, and that this is correct. If so, then should there be a Wiki category of such systems? Do I understand correctly that you would include maths and logic in this category? Please clarify how you see the relation between language, maths, and logic. David Pierce 12:56, 18 August 2005 (UTC)

I don't see that it matters which other systems there are; it's obvioulsy a system (not two or more), and it's obviously logically possible that there be others. If you disagree with regard to maths and logic, fine (we've got enough to deal with alredy without straying into those areas). --Mel Etitis (Μελ Ετητης) 16:04, 18 August 2005 (UTC)
You have not made clear your view of maths and logic, so I don't know whether I disagree with it. I think the main point here is that, in the broadest meaning of the word, language is, not a method, but the method of communication (or human communication, if you prefer, although the present article allows the possibility of animal language). Below I quote some dictionary definitions in justification of this view. David Pierce 07:27, 19 August 2005 (UTC)
The word finite serves no obvious purpose, and it may even be wrong, especially when one allows artificial languages.
No, they're always finite, though I suspect that what was meant was that the system is finite, which is also true. --Mel Etitis (Μελ Ετητης) 14:14, 17 August 2005 (UTC)
A particular logical language is generally supplied with infinitely many variables. A first-order logical language may be supplied with infinitely many non-logical symbols. As for human languages, I believe it is claimed that infinitely many grammatical sentences can be formed; each of these sentences can be considered as a symbol in itself. You may say that those sentences are built up from finitely many parts; but is this observation essential to language, so that it should be used categorically in the first sentence of the present article? I don't think so; but please read on. David Pierce 08:21, 18 August 2005 (UTC)

Perhaps you mean to refer to a potential infinity of variables. no language or other system has an infinite number (one of the key facts about language is precidely that it can be used to generate a potential infinity of well-formed strings using a finite number of elements). --Mel Etitis (Μελ Ετητης) 09:06, 18 August 2005 (UTC)
No, I meant an infinite set of variables or other symbols. I and my colleagues work with such sets. You may say that this is an illusion, because no set is actually infinite. That would be a philosophical position, and probably a minority position, at least among mathematicians. Some mathematicians work with infinite sets that they call "languages", sets whose elements are called "symbols". Is the present article not supposed to cover this use of the word "language"? David Pierce 12:56, 18 August 2005 (UTC)

So far as I'm aware it's the majority view among both mathematicians and philosophers; in fact I'm in the minority, in that I think that the real existence of an actual infinity is possible — but I certainly don't think that any language has an actual infinity of symbols. If mathematicians call sets "languages" then they're using the term metaphorically; the article might mention such uses, but not in the summary. --Mel Etitis (Μελ Ετητης) 16:04, 18 August 2005 (UTC)
I am quite content to consider the mathematical use of "language" as metaphorical or analogical or something. In fact, I think my core objection to the current Wiki definition is that it appears to be based on things like computer languages that are called "languages" only by analogy with ordinary language. I just don't see finiteness as part of the essence of language in the ordinary sense. If one believes that all sets are finite, then in particular sets of symbols will be finite. What purpose is served by introducing the word "finite" in the initial definition of language? To me it suggests a philosophical POV. David Pierce 07:27, 19 August 2005 (UTC)
In some views, the word arbitrary is wrong: If a deity uses a particular language, should that language be called arbitrary?
???? First, what on Earth have deities to do with it, and secondly, why wouldn't they be arbitrary in any case? --Mel Etitis (Μελ Ετητης) 14:14, 17 August 2005 (UTC)
Perhaps it was not a good idea to mention deities. However, I think we must be agnostic on the question of their existence. In the Jewish tradition as I understand it, for example, God speaks the world into existence using Hebrew. To suggest that Hebrew or any other language is arbitrary (in the sense of "it could have been otherwise") is to raise theological questions that I don't think belong in the opening paragraph. But to be more down to earth: "Arbitrary" suggests "established by convention". Was English established by convention? Would not such an establishment have required a prior language? What language was that? Because such questions do not have obvious answers, I do not think the word "arbitrary" belongs in our definition. David Pierce 08:21, 18 August 2005 (UTC)

I don't think that the opening paragraph should be constrained by theological concerns at all. Moreover "arbitrary" doesn't suggest "established by convention"; if I complain that a judge's decision is arbitrary, my complaint is precisely the reverse, in fact. --Mel Etitis (Μελ Ετητης) 09:06, 18 August 2005 (UTC)
The judge's decision is established by convention amongst the members of a set of size one. I understood the present definition of language to mean that a community chooses the symbols of its language arbitrarily. The community are choosing collectively; hence I suggest that the choice is conventional. Have I misunderstood the meaning of "arbitrary" in the definition? David Pierce 12:56, 18 August 2005 (UTC)

I take "arbitrary" to mean: "having only relative application or relevance, not absolute" (The Collins English Dictionary). I can't find a dictionary (on my shelves or on-line) that defines "arbitrary" as "established by convention". --Mel Etitis (Μελ Ετητης) 16:04, 18 August 2005 (UTC)
OK, but as with "finite", I ask: What purpose is served by using the word "arbitrary" in the definition of language? Again the word suggests a POV to me. Take the definition of "arbitrary" at m-w.com: "depending on individual discretion,...based on or determined by individual preference or convenience rather than by necessity or the intrinsic nature of something...". The notion of individual choice does not seem compatible with language as a shared activity. The denial of necessity and intrinsic nature is perhaps not compatible with (some) theories of Universal grammar. David Pierce 07:27, 19 August 2005 (UTC)
Human language is generally spoken before it is written; what are the symbols of spoken language? We may analyse speech into phonemes, but the possibility of this analysis does not explain what language is.
Why do you think that symbols have to be visual? --Mel Etitis (Μελ Ετητης) 14:14, 17 August 2005 (UTC)
I don't, and that's why granted the possibility of aural symbols. But is not a symbol established by convention? Then we are back to my objections to the word "arbitrary". David Pierce 08:21, 18 August 2005 (UTC)

See my previous response. --Mel Etitis (Μελ Ετητης) 09:06, 18 August 2005 (UTC)
We assume that a language has rules of grammar, just as we assume that physical bodies behave according to physical laws. But is the physical world to be defined as that which behaves according to physical laws? Also, I suppose humans could recognize that they used language before they recognized that they were following grammatical rules.
We don't assume that — it's part of the definition of "language". Anything without such rules wouldn't be a language. The rhetorical question about the physical world doesn't convey anything to me I'm afraid, and the final sentence is irrelevant. --Mel Etitis (Μελ Ετητης) 14:14, 17 August 2005 (UTC)
Part of whose definition of language? We have baby-talk and Shakespeare; it's all language. As scientists, we try to understand aspects of this language; one way to understand is to look for rules of grammar. The fact that we can go some distance with this approach does not mean that the existence of rules of grammar should part of the definition of language. Similarly, we can find symbolism in painting; but we would not define painting in terms of symbols. David Pierce 08:21, 18 August 2005 (UTC)

In so far as baby talk and Shakespearian English are languag, they're grammar-governed; if they're not grammar-governed, then they don't count as language (this is one of the problems with those who use the metaphor of language when talking of music or architecture, and then go on to take it too literally). It's not a question of what should be the definition, but of what is; it's not up to us to decide, but to report. --Mel Etitis (Μελ Ετητης) 09:06, 18 August 2005 (UTC)
OK, but whose definition of "language" are we reporting? (I already asked you this.) In a cursory search, I find an article by Chomsky stating two possible definitions of language: "the totality of utterances that can be made in a speech community" [Bloomfield], along with a supposedly Aristotelian definition: the set of ordered pairs (s,m), where s is a sentence or utterance, and m is a meaning. Nothing here about finitely many symbols. Chomsky goes on to speak of a grammar as a [finite] system that generates the language, but mentions the view of Quine and others that choice of grammar is a matter of convenience, not truth. I don't suppose this view is imcompatible with the view that a language must be grammatically analysable to be a language. Still, it seems to me that to define language as a sort of thing that has a grammar is to define language as the object of study of linguistics: "language is that to which linguistic tools like grammatical analysis apply." I think this would be taking a philosophical position, namely, that what the ordinary person considers as language will always turn out to be amenable to linguistic analysis. David Pierce 12:56, 18 August 2005 (UTC)

But (as with Collingwood) you'r appealing to definitions given in specific circumstances for specific purposes (Chomsky might be the exception; do you have a source for that?) by people with specific theories. We need to find a much more general and mainstream account for the summary. --Mel Etitis (Μελ Ετητης) 16:04, 18 August 2005 (UTC)
I gave those definitions to show that some people who have thought about language have come up with definitions quite different from the one on Wiki now. Chomsky does not find the definitions that I quoted to be useful. I don't either, although I think some dictionary definitions are approximations to those. Perhaps our article should begin something like: "Language is commonly defined as...Some people find this definition inadequate for such-and-such reasons, and propose the following definition as an alternative..." Anyway, if you are asking for what I was reading, it was "Language and Problems of Knowledge" (1990) by Noam Chomsky, in The Philosophy of Language, edited by A.P. Martinich, 4th ed., Oxford, 2001. Yet again I must ask you where in print (or on line) one can find something like the current Wiki definition of language. David Pierce 07:27, 19 August 2005 (UTC)
Communication is one use of language; but expression is another. (For example, interjections are expressive, but not communicative.)
It's not entirely clear what you mean by "expression" here, but are you sure that purely expressive interjections are strictly part of language? --Mel Etitis (Μελ Ετητης) 14:14, 17 August 2005 (UTC)
Are you sure that they are not? I think the opening paragraph of the article should make as few assumptions as possible. I suppose baby-talk at first consists exclusively of interjections. If baby-talk is not language, it is intimately connected with the acquisition of language. David Pierce 08:21, 18 August 2005 (UTC)

But what is intimately concerned with the acquisition of a thing isn't that thing. ("Baby talk" is a bit vague, and covers both what clearly is not language and what clearly is.) --Mel Etitis (Μελ Ετητης) 09:06, 18 August 2005 (UTC)
I do not share your clarity concerning what is and is not language, but it could be owing to my lack of experience with babies. Do you have a practical criterion for determining what sounds out of a baby's mouth are language? David Pierce 12:56, 18 August 2005 (UTC)
See above, inter alia. --Mel Etitis (Μελ Ετητης) 16:04, 18 August 2005 (UTC)

In particular then, I shall add "expression" to the definition of language. I don't see the need for the long list "objects, concepts, emotions, ideas, and thoughts", and I shall reduce it to two: emotions (rather, feelings) and ideas. Anybody who wants to expand again, please do so, but explain why. (What did "object" mean on the list, for example?) —The preceding unsigned comment was added by David Pierce (talk • contribs) 12:47, 17 August 2005.

I'm not clear what you find unclear about "objects".

Your new version of the introduction isn't acceptable, I'm afraid. it's not only vague and misleading, it's outright false (language isn't the capacity for anything). --Mel Etitis (Μελ Ετητης) 14:14, 17 August 2005 (UTC)

(Sorry about the original lack of signature; I don't know how that happened.) I didn't know what it meant to communicate an object (as opposed to an idea about an object.) To whom is my definition of language not acceptable? I found the old (and now current) definition to be unacceptable for the reasons I have outlined. Le me quote my reformulation here:

Language is the capacity or means of expression and communication. That we have language means that we can express feelings and communicate ideas. We express and communicate by means of language.

Here I did try to suggest why language could be understood as a capacity: to have a language is to have the capacity to communicate. (If I remember correctly, the original OED gives capacity as one meaning of language, albeit an uncommon meaning.)

But language isn't the capacity — it's the means (to have a hammer is to have the capacity to knock in a nail, but a hammer isn't a capacity). --Mel Etitis (Μελ Ετητης) 09:06, 18 August 2005 (UTC)

See below. David Pierce 07:27, 19 August 2005 (UTC)

I should say that I am influenced by R.G. Collingwood. Here is a passage from his Principles of Art:

Bodily actions expressing certain emotions, in so far as they come under our control and are conceived by us, in our awareness of controlling them, as our way of expressing these emotions, are language...Language here exists in its absolutely original shape. It has a long way to travel. Later, it has to be profoundly modified in order to meet the demands of intellect. But any theory of language must begin here. If we begin by studying the result of these further modifications, the language we use for expressing our thoughts concerning the world around us and the structure of thought itself, and take this highly developed and highly specialized form of language as representating the universal and fundamental character of language, we shall get nowhere. The grammatical and logical articulations of intellectualized language are no more fundamental to language as such than the articulations of bone and limb are fundamental to living tissue. Beneath all the elaboration of specialized organisms lies the primitive life of the cell; beneath all the machinery of word and sentence lies the primitive language of mere utterance, the controlled act in which we express our emotions.

I don't know what Collingwood's standing is today, but his book (Oxford, 1938; paperback, 1958) is still in print, and I think his notion of language as expression of emotion should be accommodated in the opening definition in an article on language; ideas about symbols and grammars should come later. David Pierce 08:21, 18 August 2005 (UTC)

Collingwood's standing depends on whom you're asking, I suspect. Philosophically (he was also an ancient historian) he's known primarily as a philosopher of history, secondarily of aesthetics, in both cases as an Hegelian. I don't think that he's read much any more by mainstream Anglo-American philosophers (I don't know about the Continentals). I generally find him unrigorous, allusive, and unhelpful, I must admit, but he was no philosopher of language, and I think that that shows in the passage that you quote (it's indicative that he's not mentioned at all in, for example Hale & Wright [edd] A Companion to the Philosophy of Language (1997), and that the reference books that mention him (and many don't) say nothing about his views of language. In any case, his view is certainly not central or standard, and shouldn't be in (or influence) the summary. If you wanted to include it at all, it should go in the body of the article (I'd say it would be better in the article on Collingwood, in fact). --Mel Etitis (Μελ Ετητης) 09:06, 18 August 2005 (UTC)

Let me now quote from two dictionary definitions of "language":

The original OED:

"1. The whole body of words and of methods of combinations of words used by a nation, people, or race; a `tongue'. Dead language: a language no longer in vernacular use.

"b. Applied to methods of expressing the [sic] thoughts, feelings, events, etc., otherwise than by words. Finger language = dactylology. Language of flowers: a method of expressing sentiments by means of flowers.

"c. Applied to the inarticulate sounds used by the lower animals, birds, etc.

"2. In generalized sense: Words and the methods of combining them for the expression of thought.

"b. Power or faculty of speech; ability to speak a foreign tongue. Now rare.

"3. The form of words in which a person expresses himself [sic]; manner or style of expression. Bad language: coarse or vulgar expressions. Strong language: expressions indicative of violent or excited feeling.

"b. The phraseology or terms of a science, art, profession, etc., or of a class of persons..."

Points to note:

The word "expression" and its grammatical variants appear several times; "communication" never appears (not in the parts I left out either).
The notion of a symbol or grammar is not appealed to.
Language can be understood as the methods [taken as a whole] of using words to express thought.
Language can be a faculty.

I haven't got the newer OED, but in the Concise Oxford Dictionary, 9th ed. 1995, "language" is defined thus (examples omitted):

"1 the method of human communication, either spoken or written, consisting of the use of words in an agreed way.

"2 the language of a particular community or country etc.

"3a the faculty of speech.

"b a style or the faculty of expression; the use of words, etc.

"c (also bad language) coarse, crude, or abusive speech.

"4a a system of symbols and rules for writing computer programs or algorithms.

"5 any method of expression or communication.

"6 a professional or specialized vocabulary.

"7 literary style."

Here I note:

"Communication" is introduced, but "expression" is retained.
"Symbol" is used only in the context of computing.
"Grammar" is never used; "rule" is used, but only in the context of computing.
Language is still a faculty; it is even the faculty of expression.

I called language a means and capacity; the dictionaries call it a method and a faculty. I don't mind using the dictionaries' words. It still seems to me that the Wiki article should start with something like the COD definition, and I think my proposed definition was an approximation to this. Words like method and faculty, expression and communication should be used. Words like symbol and grammar should come later, in the context of theories of language.

I would say that same thing about the word "word"—it should come only later—, since it is not clear to me that the division of speech into words is natural; but the dictionaries do use the word.

Finally, I do think that the COD may be misleading to allow language to be the faculty of expression simply. Collingwood makes a distinction: language is controlled expression. We (or some people) have a faculty for blushing; blushing is expressive; but I don't think anybody would refer to blushing as a part of language. My guess is that the lexicographers took control to go without saying as a criterion for language. David Pierce 07:27, 19 August 2005 (UTC)

Here was my next proposed introduction, again removed:

In the most general sense, language is the faculty or the totality of methods of expression and communication. Etymologically, language involves the voice, since "language" derives ultimately from the Latin lingua "tongue". However, the method of communication used by a particular community can be called a language, regardless of whether the method is vocal. A language might involve gesture, for example. Thus both English and American Sign Language are languages.
In a more precise sense, language is the faculty or method of communicating with words. Linguistics is the study of language in this sense. Communication in general is the subject of semiotics and communication studies.
There are computer languages, and languages used in mathematics and logic. These are formal languages, so called because of their analogies with natural language.

From the again-current definition, my version preserves the observation that a language can use gesture. I think I have already discussed why I find the current definition inappropriate or wrong. Briefly, the sentence

Language is a system of finite arbitrary symbols combined according to rules of grammar for the purpose of communication.

is ambiguous in its use of the indefinite article a: is Language a system among many, or is it the system; or is a language a system?
expresses a philosophical position and hence is not written from a NPOV.

My proposed definition begins with the most general sense of the word "language" that is justified by English dictionaries. (See the quoted definitions above.) My definition moves on to a more precise definition, still justified by English dictionaries, but more in line with what linguists actually study. By contrast, the current definition seems to be inspired by formal languages. To suggest that natural language is language only insofar as it resembles formal language is to take a philosophical POV.

For such reasons, I find my proposed definition to be superior to the existing one. This is why I made the replacement, now reverted. My reasons for replacement were already implicit in what I had written here; now I have made them more explicit.

However, the current definition has not been justified, though I have repeatedly asked for some reference that justifies it. Again I request that some justification be offered, or else that my own suggested definition be accepted or improved. David Pierce 08:17, 23 August 2005 (UTC)

My comments continue below in the section called "Definitions" David Pierce 07:05, 15 September 2005 (UTC)

Human languages

I am writting linguistic pages in Serbian Wikipedia. English versions are base for that. During the translation of the part "Human languages" of this page, I think I have description for that part. As my English is not good enough (as well as I don't work on this page), I will not paste it into the article. I ask someone who is working on this page to take my suggestions.

It is necessary to make distinction between "real", used languages and standard languages. Standard language exists util supporting political structure exists; "real" language doesn't have so strong relation with political structures. Standard language has strong political and ethnical influences; "real" language doesn't have. Even standard and "real" language has relations.
The best example for this is South Slavic language area: from Black Sea at the East to the Alpes at the West; from South Hungary at the North to the North Greece at the South. There are seven standard languages (Bulgarian, Macedonian, Serbian, Montenegrian, Bosnian or Bosniak, Croatian and Slovenian) with one "dead" standard language (Serbo-Croatian). At the other hand, there is only one dialect continuum: going from the West of Balkan to the East and from the North of Balkan to South -- there is no so distinctive differences between two places so we can say that two of them belongs to two different languages.
Standard Bulgarian language is based on Eastern dialects of Bulgarian language area, and standard Serbian language is based on Western dialect of Serbian language area. Speakers of Western Bulgarian dialects and speakesr of Eastern Serbian dialects speaks with the same language and there is no difficulties in communication between them. But, speakers of standard Serbian language and speakers of standard Bulgarian language have difficulties in communication even they can communicate (if they have good will). The same situation is between Serbian and Bulgarian at one side and Macedonian at another; as well as between Croatian and Slovenian.
Things become more bizare when we try to make distinction between four standard languages born after the death of standard Serbo-Croatian language. Even differences between standard Serbian, Montenegrian, Bosnian or Bosniak and Croatian language are minimal and we can't say that there is more then one "linguistic" language. But, political and ethnical feeling of speakers of that "languages" say that they are using only "lanugage" wich has the same name with therir ethnicity and country.
Because of that it is important to know that standard lanugage is political, not linguistic category. And linguists try to describe both: standard and "real" language.
--Millosh 08:16, 4 Jul 2004 (UTC)

Excellent points. This discussion is a must, especially in light of the constant problems we have with the Serbo-Croatian language article. See below in From Basic Principles to Specifics for more. --Tox 13:14, 12 November 2005 (UTC)

Human Language

I included some context for just what language 'is' and moved around the linguistics section. If anyone has any comments on the change I'd be glad to hear them.

--Kharhaz 06:21, 29 Sep 2004 (UTC)

using s and oes

Animal Language

I have added some information to the Animal Language section. It may be a little contoversial, however, I think the previous entry did not acuratly reflect the current scientiffic consensus.

--Selket 21:34, 29 Jan 2005 (UTC)

Junk

This article is full of junk. Two examples:

"The origins of the word language can be found in the Middle Ages when Latin was considered the language par excellence, so the very word "Latin"�l�den in Aglo-Saxon or leden in Middle English came to mean language."

What on earth does that mean? 1) the word "language" does not come from the word "Latin", and 2) people had the concept of language before the Middle Ages.

"Some say that our reality is in fact created as it is described by our language. Albert Einstein spoke of a pocket watch sitting on a table. If you had never seen a clock or a pocket watch and had no clue what it was, how might one describe the sound "ticking". Our language and linguistics can alter our reality."

Vague, unreferenced, badly expressed mysticism by someone who's never studied linguistics.

Tamil

Am I missing something? Someone claims that Tolkaappiyam broke sounds into consonants and vowel and that is was a huge leap in linguistic studies. But isn't this ridiculous since any alphabet which has seperate glyphs for consonants and vowels already has considered this. And certainly Tamil alphabet is not the first to do so (Brahmi script which the Tamil script is based on, Egyptian hieroglyphs, Phoenician alphabet just to name a few).

Egyptian hieroglyphs did not represent vowel, only consonant. There were a few glyphs that have the appearance of vowels (such as the semiconsonant 'y'), but they were not vowels. Phoenician likewise had no vowel, only consonants. Brahmi did not have vowels, either, but only a way of indicating the vocal 'color' of its consonants. Again in Brahmi, there were a few letters that seem to Westerners to be vowels, but it was only a superficial resemblance. The quality of the letter 'k' could be shown by the addition of one of several diacritics, yielding ka, ki, ku, ke, ko, k?, k?, k?, kai, kau. In addition, there were some letters that look to modern Western eyes like the pure vowels a, i, u, e, and o...but these were not vowels at all, but nul consonants. There was no concept whatsoever of a vowel.

And that is the case with the modern languages of those regions. Arabic, for example, is only consonants, and the consonants may be colored by application of diacritics (fatha, kasra, dhamma) and marks of prolongation. Almost all the scripts used in India today, as well as most of the scripts of Southeast Asia, work like Brahmi, and the letters are all consonants that receive vowel coloring by the use of various diacritics.

The Tamil script is unique among India�s scripts, and is the only one that has come close to a true alphabet. Tamil writing is easy for Westerners to learn, while almost all of the other scripts of South and Southeast Asia are exotic and difficult (for us) syllabaries. Thai is said to have become alphabetic, but vowels in Thai words are often not indicated, and you really have to know a word before you can pronounce it correctly. Tamil vowels are always very clear, and even unfamiliar words are easy to pronounce from the spelling. —Stephen 7 July 2005 07:18 (UTC)

Improvement Drive

The article on Acholi language is currently nominated to be improved on Wikipedia:This week's improvement drive. If you can contribute or want it to be improved, you can vote for this article there.--Fenice 16:42, 17 July 2005 (UTC)

Definitions

This continues my comments under "Introduction" above. As nobody had said anything about those comments after my last extensive addition to them, yesterday I changed the language article accordingly. Then, for a third time, User:Mel Etitis reverted my changes; he has offered no reason here.

It is not clear whether Mr Etitis had noticed that I had added a new section on definitions of language. There I assembled, with documentation, the definitions that I had been able to find. I wish people with broader knowledge would add what they know (also with documentation). I might be able to improve it myself, but I would first want to know that not everything I did would be summarily deleted. If there is an objection to the new section as such, I can't imagine what it is, but I am open to persuasion.

Meanwhile, I am restoring the new section and am placing there the introduction that I had preferred for the article itself. The introduction that Mr Etitis prefers is retained, with a tag of dubiousness, discussed in "Disputed" below. David Pierce 07:05, 15 September 2005 (UTC)

Disputed

I do not think the introductory sentence about language says what language is. I think it expresses a point of view about language, but not the only one. I have expressed some of my concerns in the section "Introduction" above. I have repeatedly requested some authority for that introductory sentence, but no citation has come. I have cited other authorities whose definition of language is quite different.

It seems to me that an article like the present one should first establish that we are all talking about the same thing; then the article can go on to give theoretical accounts of the subject. I think most people know that they use language, and they recognize that other people use language; but they don't know grammar and don't think of language as essentially grammatical or as analysable into symbols. They may be wrong; but I think the language article should begin with a description of language that everybody can make sense of.

In any case, let authorities be cited. David Pierce 07:05, 15 September 2005 (UTC)

The section on "The study of language" pointedly makes no mention of European linguists of antiquity: for example, Donatus the Grammarian who was active aroud AD 354. The jump to a Persian active in AD 760 followed immediately at the beginning of the next paragraph by the words "Later in the West..." implies there was none. Is Wikipedia racist? Or does anti-European bias not constitute racism under Political Correctness?

recent changes

I reverted because the edits were a combination of unencyclopædic style and controversial content. To be frank, how could anyone think that "There is language, and there are particular languages." – not just as a sentence, but as as a one-sentence paragraph – is an appropriate way to start a section, much less an article ([1])? Apart from the style, it merely states baldly what is made clear in the article. The rest of the new material is similarly infelicitous in style, and involves appeals to often (out-dated) reference works, bald and often inaccurate (usually because too sweeping) claims, and original research.

I'm surprised and disappointed that an article of this importance hasn't attracted more atention from other editors. I'll try to rectify that, but until there's been serious and proper discussion, please don't make these large-scale changes to the article. --Mel Etitis (??? ??????) 10:55, 15 September 2005 (UTC)

Mr Etitis, as I understand Wikipedia:Disputed statement, you should not remove my "dubious" tag without citing sources to justify the dubious statement. (Also your reversion is not minor.) Citing sources may not be an absolute rule, but in this case, if you think the claim about language is accurate, you ought to be able to name somebody else who makes it. I do not understand why you continue to ignore my request.

I do not know what is controversial about my other contributions. If I misrepresented my sources, explain how. Which sources are outdated? And so what if they are: is there no historical value to them? Where is the original research? Where else in the article is universal grammar (for example) mentioned? Can you please be specific about your objections?

These are all serious questions. I am quite mystified. I wish somebody else would get involved here, because I am quite at a loss to understand your antipathy for what I write. Have you looked at the new articles I contributed? Do you find them unencyclopedic? David Pierce 12:51, 15 September 2005 (UTC)

I'm not "Mr Etitis"...
Rather than sticking unsightly templates on the article, it would be better to discuss issues here; the template is something of a last resort. In this case, it's not at all clear why you find this statement dubious, in fact; it's so basic, and widely accepted, that your objection is difficult to take seriously. But, from the first book I looked at:
"we can start with relatively obvious features of language. (1) A language is made up of units interrelated in in some sort of systematic way. [...] (2) Units at the word level have meaning by "convention." (3) This system (language) is used for communication." (William P. Alston, "Language", in Edwards [ed.] The Encyclopedia of Philosophy vol. 4, p.384)
The minor is automatically added by the admin's "rollback" function.
Do you not think that the 1911 Britannica is just a touch out-dated (at nearly a century old)?
I have no general antipathy towards you or what you write; I haven't followed you around checking your other articles. --Mel Etitis (??? ??????) 14:34, 15 September 2005 (UTC)

I'll step in and give my opinions.

David Pierce's suggested opening paragraph is really awful.
Some of David's paragraphs that followed were okay and informative, but they should not come at the expense of Mel's concise description of what a language is generally agreed to be. Every article should begin by describing what the topic is.
Marking a revert as "minor" seems like bad form to me. Are rollbacks really always marked as "minor"? Why in the world would they be like that?

RSpeer 18:59, 15 September 2005 (UTC)

I agree that some of David Pierce's additions are rescuable, but they need considerable rewriting; I intend to go through them when I get the chance. I don't know why rollback reverts are automatically minor — I've often thought that it was a bad idea. It's the only reason that I'm occasionally reluctant to use it, but it saves so much time, and I'm so often swamped by vandalism-fighting, etc. that I find myself seduced... --Mel Etitis (??? ??????) 20:23, 15 September 2005 (UTC)

I am glad somebody else has joined this discussion. Rspeer, which of my suggested opening paragraphs is really awful? I think I've made three. The last one said there was language and there were particular languages. Mel Etitis also didn't like this. Can you give me more explanation? The point is that the word "language" is used for this or that language, and also for the concept of language—the concept, or some other abstraction; I'm not particular about the word used. By contrast, the word for the concept of (idea of?) a human being is not "human", but "humanity". An anthropologist studies humans or humanity; a linguist studies languages or language. I don't have a name for this "dual" property that "language" has, but "human" has not. I do observe:

the property seems to exist (many gerunds in English have it: there are happenings, and there is happening);
"language" seems to have the property;
the opening sentence to the article, as it is now, is unclear to somebody (like me) who recognizes the dual property of the word.

Mel Etitis, thank you for giving a reference. I note the following from your quote:

"Units" is used, rather than "symbols". I think the latter word carries too much baggage.
"Grammar" is not used (neither is "finite").

Since you haven't looked at the other things I have worked on, I'll tell you that I created Turkish grammar and related articles. I can tell you more about Turkish grammar than most Turkish people, probably even most educated Turkish people; but I don't really know Turkish. What are people doing when they use language? I think it is misleading to say that they are "following the rules of grammar". If you are driving a car, are you following the rules of the road? Probably, but I don't think this rule-following does a lot to explain what you are doing. Also, the rules of the road are there to keep you from doing things you might otherwise consider doing, whereas, most people, when speaking, are not worried about following rules.

Yes there are rules that are followed willy-nilly. The Law of Gravity is a rule that is not optional. Here then is another kind of "dual" meaning: some rules can be broken, albeit with bad consequences; other rules are in principle unbreakable. Which kind of rule is meant by the reference to the "rules of grammar"?

In any case, it doesn't ultimately matter what we think; we want common knowledge or understanding here; but I still haven't seen any authority for defining language in terms of grammar.

Here is a fundamental question about articles on subjects like language. They should start with a definition, but what kind of definition? In terms of the article Definition, I think an article should start with a lexical definition, giving a theoretical definition later. I think the language article now starts with a theoretical definition, because of the mention of grammar.

I cited the 1911 Britannica because somebody pointed it out to me; but it just said in terser form what the old OED said. I haven't got ready access to the latest OED, but I have looked at recent lesser dictionaries. None mention grammar. One calls language an "aspect of human behavior...". The older references are outdated if one is looking for a theoretical definition of language, but perhaps not if one wants the lexical definition. In fact the older references might be better for not being influenced by later theories. The age of the reference should be indicated, and I did indicate it.

But again, the main question: Is there a distinction between a lexical and a theoretical definition, and if so, should the former be given, and given first? I think Yes and Yes. Hence, for example, I think the Sun article should begin by pointing out that the sun is the bright disk that makes day and night; only then should it be said that the sun is a star, just like the ones that come out at night. The latter understanding took centuries to develop and is a fascinating story in the history of science; the story shouldn't be hidden by the suggestion that the sun is simply a certain star. (I did try to make the change, but ineptly it seems; anybody interested can see the history, and the discussion page there under "intro".)

I still wonder where my "original research" was, but I've already asked the most important question. David Pierce 08:23, 16 September 2005 (UTC)

This Article Should Begin Differently!

I open an article about Language and immediately I'm buffeted by a section on the Tol'kapiyar and Tamil grammar! That's ridiculous and is no way to open such an article. RCSB 23:08, 27 September 2005 (UTC)