No subject

Sat Sep 1 07:55:31 UTC 2007

Dear all,

Investigating into some possible connection between negation and nasality is an exciting prospect but I wish to be sure I fully understand the question and its limits. In a few words:

1)     supposing that at some semantic, cognitive level or whatever there is any such thing like “negation”, does it universally call for nasal marking? 
NO. Counter-examples are no doubt infinitely more abundant than matches.

2)     supposing that in a given language N can be analyzed as a “unified morpheme” with a core value, is it necessarily a negative one? 
NO. In Basque –n is used for forming the inessive and the genitive of NPs and the past and the 2nd person singular feminine in the agglutinative conjugation; and the negative marker is “ez”. So one might wonder if the uses of –n are interconnected but certainly not how they relate with negation, unless I’m missing something.

3)     supposing that in a given language we find nasal markers connected with negation such as not in English, nu in Georgian, na(i) in Japanese (+ arimasen) and nngit in inuktitut, it seems necessary to determine whether the nasal element is a candidate in the first place. It is so in IE languages both because in synchrony the marker is sufficiently widespread to be undeniably attested (on distributional grounds) and because diachrony the same group has been stubbornly consistent in preserving the “nasal connection” of negativity in spite of phonological change – so that diachrony justifies the origin (this is perfectly tautological) but does not explain why the connection is still alive and kicking (the linguist). In other cases whether nasality is a candidate at all as a negative marker is to be explored. In Georgian is nu an isolated case: other occurrences of negative N? If not I am reluctant to admit it as a match. The case of nngit is to be investigated. That of na(i) and arimasen are more promising but not conclusive at this early stage.

4)     Admitting that nasality may be connected with negation this raises other questions: what about other cases like possible connections between dental consonants and the past or the accomplished (apart from IE languages, Japanese ta, to, possibly te), vowel alternations (Georgian: u- / i- / a- as  verbal prefixes; wolof: noun classifier + I (proximal), A (distal), U (undetermined + lexical post-determiner). In other words: if N in Georgian should eventually be accepted as negative, would this raise the question of –u in nu, its connections with other occurrences of u in the same language, therefore a compositional analysis of morphemes…

5)     If the nasal connection is validated in some cases at least; what would it mean?
How is the motivation, if any, oriented: from intended meaning to the selection of sound, or from the construal of meaning out of the sounds’ proprioceptive properties ? And which ones? (multimodality: essential visual, auditory and tactile properties) And how are they distributed in the environment and between the interlocutors? for ex: I / a = proximal / distal if one relies on tactile iconicity (closed / opened) and absolute auditory iconicity (Doppler effect: treble / bass) but in terms of relative auditory iconicity I can be dissociative and A associative (A = “echoing effect” = will be responded by perceivable harmonics in confined environments while /i/ will not. I tested this with my daughter in a cave: all tourists present tested the echo with loud /a/s; just for fun I tested a shrill /i:/, with no success at all, “why does it not work with me? Is there a power cut or something?” and everybody burst out laughing, including those who did not speak French, because they had this intuitive knowledge (acquired through experience) of the interactional properties of sounds on the environment – sounds may not have actual meaning but they are not “cognitively neutral” and I suspect that the root of motivation (not determination), when motivation does happen, is to be found in this kind of phenomenon.

Many linguists are reluctant to raise questions like that of the “nasal connection” because we have a feeling that if we open Pandora’s box all hell is going to break out loose, iconicity, phonosymbolism, cratylian motivation and all that. I wish the current discussions were the opportunity to reconsider those questions methodically – actually make the problematics explicit and elaborate scientific protocols to establish the data in the first place since even “what is data” in this domain cannot be taken for granted; at least we’d know wheter ther is a question at all since obviously there is reason to doubt it.
This might also be an opportunity to bridge a gap between typology and theories which raise serious questions (cognition, embodiment, distribution, intersubjectivity, language formation and acquisition) but from other disciplines’s standpoints and without the linguistic data required to be accurate and avoid postulating universals that ignore diversity.

Best regards
Didier Bottineau

CNRS, UMR 7114 MoDyCo 
Université Paris 10, Bât. L, R12D
200 avenue de la République
92001 NANTERRE
01 40 97 40 72

De : Discussion List for ALT [mailto:LINGTYP at LISTSERV.LINGUISTLIST.ORG] De la part de claude-hagege
Envoyé : vendredi 31 août 2007 20:19
À : LINGTYP at LISTSERV.LINGUISTLIST.ORG
Objet : nasals and negation

Dear James and Matthew,

    James Gair writes that he does not recall that Dravidian has been mentioned. Dravidian is explicitly mentioned in my 30 August message, reproduced below. Besides, I agree with Matthew Dryer that "we need to be careful in a number of ways" and that  "what we need to do is to avoid examining too many languages from the same family". This is precisely why the message below cites languages from widely different phyla and recalls that one among many necessary methodological precautions consists of examining the different types of negations languages may exhibit, rather than looking for negation in general. Finally, I'm not sure it is true that nasal morphemes crosslinguistically tend to be shorter than others.This is not always true, for example, in languages that use verbs to express negation, nor is it true on the level of whole language families (with the proviso above, of course!): Austronesian languages like Indonesian, Tagalog, Malagasy, various Oceanian languages, etc., have a strong tendency towards dissyllabic words, whether nasal or other. Various African languages express main clause negation by reduplication, and even, in the case of Banda-Linda ((Adamawa-Eastern), reduplication + tonal phenomena.

All best, Claude

Dear all,

    It is somewhat surprising to see that a logical, or cognitively motivated, link is proposed between nasality and negation, just because many languages express one by means of the other. It is true that the number of such languages that have been mentioned so far in this discussion is impressing. It is also true that since many languages have more than one way of expressing negation (depending on mood, aspect, subordination, nominal or verbal sentence, etc.), one must be cautious before affirming hastily that a language does not use a nasal in negative forms: Corean has ops-ta "not to exist", but also ani-ta "not to be"; Standard Indonesian has tidak "no" but bukan "is not", and Jakarta spoken Indonesian has nggak "no"; Georgian has ar "no", but also prohibitive nu, and as for the negative potential ver "cannot", it begins with a voiced labiodental which has some features in common with a nasal. 
    But although it is true that theoretical attempts to interpret the link between nasals and negations  may have a local application scope, and that, to that extent, one cannot deny that they may shed some light on the nature of negation, such attempts remain shaky. The main reason for this is not far to seek: the number of languages that give the lie to these attempts is no less impressive that the number of those that support them. To mention only a few: Tagalog has huwag, Malagasy has tsi, Palauan has diak (and for that matter, negations containing nasals are rare overall in Austronesian languages); Israeli Hebrew has lo; neither Hausa nor Fulani have negations containing nasals (but is true that Wolof, though West-Atlantic like Fulani, has nasals in some of the negative pronouns of its verbal paradigms); in many of the languages where the main clause negation is in fact a verb "to not", there is no support for the hypothesis of a nasality~negation relationship: such languages are, for example, Finnish (and Estonian), Comox and other Salishan languages. 
    I am just mentioning here languages that occur to my mind now, and I have no time to pursue the inquiry further, although I suspect that Eskimo-Aleut, Chukotko-Kamchatkan, Na-Dene, Penutian, Dravidian, many Caucasian languages deserve to be examined in that respect. A large-scale inquiry would no doubt bring a lot of other counter-examples to the (interesting) hypothesis of a nasality~negation link, essentialy based on Indo-European languages and others from some other groups.

All best,
Claude Hagège, Collège de France, Paris.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/lingtyp/attachments/20070901/2895708e/attachment.htm>