Talk:Natural semantic metalanguage

From Wikipedia, the free encyclopedia

This article is within the scope of the WikiProject Constructed languages, an attempt at creating a standardized, informative, and easy-to-use resource about constructed languages, aka conlangs. If you would like to participate, please visit the project page, where you can join the project and see a list of open tasks.

If listing this article for deletion or if there is an active edit war, please post a note here.
Constructed languages
Portal
Start This article has been rated as Start-Class on the class scale.
Mid This article has been rated as Mid-importance on the importance scale.


To-do list for Natural semantic metalanguage: edit · history · watch · refresh
  1. Restructure article.
  2. Expand theoretical background
  3. Allolexy, polysemy
  4. More on history
  5. More practical examples of its use (Goddard 1998, Wierzbicka 1988, G&W 1994, G&W 2002)
  6. References: weed out the less central ones, or better yet, divide them in basic and secondary literature
  7. Theoretical context: relation to other semantic theories and to cognitive linguistics; opposition to 'objectivist semantics' (like Lakoff).
  8. Psychological reality and status of semantic primitives
  9. Criticisms & problems (Ameka 1994, Van Driem 2004, Lakoff 1987:279-280, Goddard 1998)

Last update: — mark ✎ 02:35, 11 Dec 2004 (UTC)

Anna Wierzbicka structured the list of semantic primitives in groups

The 61 Semantic Primitives (addition of LONG is proposed)

ABOVE, AFTER, ALL, BAD, BECAUSE, BEFORE, BELOW, BIG, BODY, CAN, DIE, DO, FAR, FEEL, FOR SOME TIME, GOOD, HAPPEN, HAVE, HEAR, HERE, I, IF, INSIDE, KIND OF, KNOW, LIKE, LIVE, A LONG TIME, MANY/MUCH, MAYBE, MOMENT, MORE, MOVE, (LONG), NEAR, NOT, NOW, ONE, OTHER, PART OF, PEOPLE/PERSON, THE SAME, SAY, SEE, A SHORT TIME, SIDE, SMALL, SOME, SOMEONE, SOMETHING/THING, THERE IS, THINK, THIS, TOUCH, TRUE, TWO, VERY, WANT, WHEN/TIME, WHERE/PLACE, WORD, YOU

Contents

[edit] Chinese radicals

Hi, You guys should include chinese radical characters as part of your study. They are words by themselves, and the combinations of these radicals form mroe complex concepts. An interesting place to start would be the formalized Han dynasty script, and the oracle bone scripts. - KevinSim

Where is the reference to the most recent structuring?

It's easy to find from the NSM homepage. Here's the direct link: http://www.une.edu.au/arts/LCL/nsm/index.htm --Crag 21:54, 2004 Jun 8 (UTC)

[edit] Bias

This article presents the theory as if it were universally accepted. As with any theory in modern linguistics there are reasonable arguments for and against. How about some coverage of the opposing viewpoint (beyond a link to an article titled "bad arguments against...")? --rkharrison

Noted. Started. Though the "bad arguments against ...." paper is a pretty good starting point for a survey of criticisms. Yakushima (talk) 15:45, 6 June 2008 (UTC)

[edit] Notice

Edit in progress. I'm planning to give the article more structure by adding sections & headlines; after that I'll be adding some references, text (history/background information, central assumptions, some criticism), and links.

I'm not a native speaker of English, so don't hesitate to point out problems in my style, vocabulary or grammar.

strangeloop 18:46, 13 Aug 2004 (UTC)

[edit] Pointless ramble about 'Atomic'

At the moment the article says that 'the semantic primes are believed to be atomic, primitive meanings present in all human languages.'

There is some subtle neatness in the analogy of calling them atomic. For a long time, the things we call atoms were thought to be indivisible- hence their name comes from Greek meaning 'non-divisible'. Of course we now believe they are divisible, made up of protons and electrons. In essence, atoms are not 'atomic'. These semantic primes are also not really 'atomic', they are not indivisible. The words in this list really only have a lesser, related property: None of them can be fully defined in terms of the others. Infact, many of their meanings overlap. For example, I, YOU, and SOMEONE all define categories of things which all contain at least what the category SOMEONE contains. The fact that they can even be usefully categorized themselves implies that concepts within a category share some meaning. For anyone reading this that is reasonably fluent in math, what I am saying is that they form a vector basis of some set of meaning but they are still not orthogonal.

Anyways, I'm not saying all this to suggest that this endeavor has failed nor even to suggest that the article's choice of word needs to be revised. I just want to show how really neat this is. Atoms failed to be 'atomic' in almost exactly the same way! Carbon, Lead, and Hydrogen turned out to all contain at least what Hydrogen contains. Of course, none of this implies that Carbon and Lead aren't in some sense elementary or primitives, and while they aren't truely indivisible they are atomic in a different sense. Anyways, the whole neatness is that Atoms were named after an old Greek idea about the world, and this turned out to be a misnomer. The adjective 'atomic' now either refers directly to these new atoms or to the old greek idea- related, but distinct concepts. These semantic primes, on the other hand, are atomic in a new sense, a generalization over the atomicity of the physical atoms. The idea being generalized is that atoms are now thought of as the smallest division of something(element) which retains some property(chemistry). In perfect analogy, the semantic primes are the smallest division of meaning which retains the property of being a word in these natural languages. Neat. Anyways, good work, sorry for babbling, nothing is wrong with the article.

-P.S. The analogy to be drawn between the primes and the nucleons is even better. --Intangir 01:32, 31 December 2005 (UTC)

[edit] This article should include at least one link that explains what is actually meant by each semantic primitive

Since English is a very ambiguous language, and only one meaning intended by each semantic primitive, there should at least be one link to something that specifies "the intended meaning" of each one (especially since this is an encyclopedia). Can anyone find where they specify which meaning is intended, and link to it somewhere (or incorporate this explanation into the article)? I was unable to find this. - Doubleg 20:12, 8 June 2006 (UTC)

The problem inherent to this issue is that these primes are claimed to be atomic, undivisible meanings. Therefore, the primes of this metalanguage, 'carved out of natural language', are by definition undefinable.
I don't think there is an online source detailing the relation between the set of semantic primes and the corresponding words in English. I believe Goddard (1998) discusses this issue, and I'm sure you can find relevant information in the 2002 volume edited by Goddard & Wierzbicka, too. — mark ✎ 20:10, 11 September 2006 (UTC)
Still the original sources may provide enough explanation of the primes to understand which usages of the English words that are taken as exponents of the primes represent the primes, and which are not, that could allow for a description (rather than definition) of the form “X, as used in Sentence Y, but not as in Sentence Z”, or some other clarification of the relation of the English words to the universal primes that is not a definition per se.--Cmdicely 02:33, 6 November 2006 (UTC)
I think original objection is valid -- NSM researchers are using restricted senses of the English words used to code the primitives. For example, YOU is never the plural sense, nor the vague I/we sense seen when people say things like "you try as hard as you can and what does it get you". Maybe the article should at least outline the senses of the primitives, point out where they were introduced, what (if any problems) were encounted with them, and in the case of those deprecated (e.g., "IF ... WOULD ...") explain why they were cut and what explications were found to substitute for them. Yakushima (talk) 15:50, 6 June 2008 (UTC)

[edit] Pirahã language and numbers

"The semantic primes (below) are believed to be atomic, primitive meanings present in all human languages." -- According to Wikipedia itself Pirahã language has no numbers, no words for "one" or "two." But this doesn't mean I doesn't believe these words to be "atomic" in a sense, they just don't exist in all languages. Thus "It is said that after eight months of enthusiastic but fruitless daily study ... not a single Pirahã had learned to count up to ten or to add 1 + 1" according to Pirahã and the Sapir-Whorf hypothesis. Maybe some words about this?

Second, aUI has another set of "primes" (aUI, The Language of Space). But this language use metaphorical compounds and so on. But the point is, I think even the "litteraly" meaning differs among languages, and they combinating different primes to archive a certain meaning.

This is actualy my first edit on Wikipedia ever [except a minor in user area], so it's a test also :P -- Veoler 18:38, 11 September 2006 (UTC)

Welcome to Wikipedia! Wikipedia being a tertiary source, we can only report on Pirahã and its possible implications for NSM if this issue has been discussed before in a reliable source. I have not yet come across such a discussion in the linguistic literature. Judging from what I read, the NSM researchers would probably investigate whether the Pirahã words for 'few' and 'some' would be susceptible to a polysemy/heterosemy analysis, i.e., whether these words in some contexts can be taken to signify the meanings of the allegedly nonexistent primes. If they aren't (and this issue is much less easy to determine than the wording of the Pirahã language article suggests), then NSM's hypothesis about ONE and TWO being conceptual primes probably would have to be adjusted. — mark ✎ 20:01, 11 September 2006 (UTC)
A quick skim of a paper by the main linguist studying the Pirahã language suggests to me that Mark is probably right. With almost wearying regularity, NSM researchers are shown a language seems not to have one primitive or another, only to discover that there is a way to say it after all. My favorite so far: a language with the same word for "see" and "hear". They can express the distinction when they need to, with something like "I could @@@ the bird's voice, but I couldn't @@@ the bird." Yakushima (talk) 16:10, 6 June 2008 (UTC)

[edit] UNE Broken Links

The links to material from the University of New England are broken, it looks like there's been some reorganization on their site. —Cmdicely 01:36, 6 November 2006 (UTC)

[edit] Puzzling punctuation and an apparent inconsistency between lists

This is my first post, so please forgive me if I say anything senseless or annoying.

I was building a table of semantic primitives in order to map them onto other short lists of basic words. In the process I noticed some unexplained use of punctuation, and an apparent inconsistency between the list on wikipedia and the list on the NSM website.

The wikipedia page states there are 61 semantic primitives. The NSM homepage states that same number. However, there originally appeared to be 65 semantic primitives on the wikipedia page.

I finally noticed under the category of "space" that the pair of primitives "far, near" and the next pair "side, inside" are separated from their neighbors by semi-colons. That same punctuation is used on the NSM website. I don't know what that means, but one explanation could be that the English words in each pair represent different aspects of the same semantic primitive. From a neophyte's point of view, the semi-colons are very difficult to pick out, (nigh on invisible,) and now that I've noticed them I'm uncertain as to what they actually indicate. I wonder if it might be helpful to add a note that clarifies the situation, as was done with the proposed primitive (long.)

Assuming that each of those two pairs really do represent the same primitive, that still leaves 63 primitives on the wikipedia list. I finally discovered that the list at NSM under the category of "actions, events and movement" shows the semantic primitive "touch" where the wikipedia page shows two semantic primitives "go" and "put". Substituting "touch" for "go" and "put" brings the count to 61 plus 1 for "long". I've read much of the NSM website, plus a couple of other articles that are cited. So far I have not discovered an explanation.

I would not be surpised to learn that one web site is more up-to-date than the other. I thought it might be helpful to mention this in the interest of clarity. If at some point in the near future I stumble upon an explantion, I will post a follow up suggesting a posible change. Even if I think I've discovered an answer, as a non-expert I would not presume to edit the page.

Peace be with you,

{Gdeclute 21:54, 5 March 2007 (UTC))