Re: [tied] Speculation

... pronouns must be short words, clearly composed of easily pronounced sounds, generally without consonant clusters. The consequence is that pronouns are similar in almost all languages, though this does not imply a common origin ... Therefore, pronouns must be used with caution in establishing relatedness of languages.

 

Unmarked and perceptually salient consonants are frequently employed as useful grammatical markers in CV combinations, and that fact alone increases the possibility of chance agreement between unrelated languages. Additionally, there is a synaesthetic tendency to associate certain sounds with certain deictic ("pointing") functions, which accounts, for example, for the frequent occurrence of nasals in first-person markers (Bantu *mi 'I, me', Hausa ni 'I', mu 'we', Hungarian én 'I', mi 'we') and coronal consonants in demonstratives (Bantu *-da/*-de/*-dia 'that', Indonesian dia 'he, she, it', Finnish se 'that, it', IE *so-/*to-). Distant ("that/there/then") deixis signalled by coronals, or by low vowels, may contrast with proximate ("this/here/now") deixis signalled by dorsals, or by high vowels. Nasal negation is common cross-linguistically, also in non-verbal negative signals ([?m?m:] and the like).

 

Happy new year to you, too!

 

Piotr

 

 

 

----- Original Message -----

From: Håkan Lindgren

To: Cybalist

Sent: Thursday, December 28, 2000 8:48 PM

Subject: [tied] Speculation

Thanks for your answers to my question about backgrounding! (Yeah, I know that was more than a week ago, but, well, y'know...)

The paper that made me ask those questions (Richard A. Rhodes, "On Pronominal Systems", in "Indo-European, Nostratic and Beyond: Festschrift for Vitalij V. Shevoroshkin", Washington 1997) proved to be really interesting. Rhodes confirms something I've been suspecting for a long time - that words are not only shaped by linguistic history (I mean, by sound laws acting on earlier stages of the language) or by pure coincidence, there's also a third factor.

According to Rhodes, frequently used words, like pronouns (or pronominal affixes) tend to use less "marked" sounds. In Ojibwa, one of the languages he studies, the sounds (I think "phonemes" is the proper term) a, i, d, g, m, n, etc. are far more frequent in pronominal affixes than in the language as a whole. We often consider pronouns to be among the stable, unchanging core elements of language, therefore we use them in long-range comparison, but because of this tendency to use less marked sounds pronouns may tend to look similar in languages that are not related, says Rhodes. People doing long-range comparisons would have to look out even more for "false friends", similar words in unrelated languages may not just be borrowing or a coincidence, they may be shaped that way by the demands of backgrounding or other similar factors.

Rhodes' ideas connect to some of my own speculations. I've kept these ideas to myself so far, but now I would like to hear what you think of them. I hope what I'm going to say now will be intelligible despite my broken English.

Let's start with an example. If we made a list of demonstrative pronouns from a large number of languages from all continents, perhaps they would look vaguely similar, not because they are all related to Patrick Ryan's distant Proto-World language, but because they are all "pointing sounds", sounds that are typical of humans who want to point something out. Their demonstrative meaning might restrict what sounds are used for these words - some sounds are perhaps more frequent in demonstrative pronouns, and other less frequent. And if we follow this line of thought, could any idea or meaning be connected to just about any sound? Or are certain ideas and meanings predisposed to be associated with certain sounds because of the way all humans think and feel?

Could the pronouns "I" and "you" be represented by just any sounds - are the sounds that represent these pronouns in all the world's languages merely chosen at random? Was it just a coincidence that the Eskimos "chose" to use "uvanga", the Finnish "minä" and the Chinese "wo" for the first person singular - could they just as well have "chosen" any other combination of sounds?

Would a "soft" idea ever be expressed with a "harsh" sound? I admit that "soft" and "harsh" are not exactly well-defined categories, and that the opinion of which sounds are soft or harsh may differ from one language to another - but beneath these differences, isn't there a level where most people agree which sounds are harsh and which are soft? A lion has a harsh sound, while a sparrow doesn't.

If this idea of mine is true (I don't imagine that it could ever be proved true), it means that words from widely separated languages may sound similar, not because these languages are giving us a hint about a distant proto-language, but because there are similarities in the way people from all linguistic families think and feel, and this influences what sounds they choose to represent a certain idea. If we discovered that such similarities exist, we would not have discovered the primordial proto-language, but we would have learnt something deep about human nature.

Hakan Lindgren
wishing everyone on the list a happy new year!