Re: Related languages (Re: A China-Sumer connection)

From: Neeraj Mathur (neemathur_at_hotmail.com)
Date: 03/24/05 

Date: Thu, 24 Mar 2005 13:19:32 -0000

<phippsmartin@hotmail.com> wrote in message
news:1111640265.515654.194990@l41g2000cwc.googlegroups.com...
> Comm wrote:
>> "Neeraj Mathur" <neemathur@hotmail.com> wrote in message
>> news:d1qb9t$9l7$1@news.ox.ac.uk...
> Yes, BUT when I was studying Mandarin Chinese the teacher made the
> comment that if you trace back the pronounciation of words in Chinese
> (presumably if you found ancient rhyming dictionaries?) then you will
> find that words used in Vietnamese to mean the same thing are, in fact,
> the same word. He went so far as to speculate that Vietnamese sounds
> more like ancient Chinese than modern Chinese does (presumably due to
> the influence of Manchu pronunciation on Mandarin). Yet people on this
> group have insisted that Chinese (Mandarin, Cantonese, etc.) and
> Vietnamese belong to unrelated language groups. Comm suggests that
> this is a paradigm, that people are classifying Chinese languages as a
> group for political reasons whereas, in reality, Cantonese differs from
> Mandarin to the same extent that Vietnamese differs from Cantonese.
> Who is right?

You're talking about one set of words only. What about the other ones? There
are two types of words (or morphemes) in a language: we can generalize them
to 'content' words and 'function' words. In English, content words include
most nouns and many verbs: 'sing', 'song', 'computer', 'desklamp' etc -
these are the things the language talks about. On the other hand, the
prepositions, pronouns, auxiliary verbs, little bits that you add to content
words for grammar (morphemes) like -s, -ing, -ed are all 'function words':
it is these which make the language what it is, how it makes meaning out of
random content words. Ask your teacher if the words in Vietnamese that are
like Chinese are content words or function words. I don't know, but I would
expect that the vast majority are content words only.

Now the thing is, anybody in the world can talk about a computer, and they
will probably use the word 'computer'. They might even use the word 'surf'
when talking about the internet. Or any number of content words could be
borrowed, at any time and from any language to any language. But the way in
which the language talks about these things is what defines their 'essence'
as it were. So in English, the plural of 'computer' is formed by adding
an -s, which -s has a history in English that goes back to Proto-Germanic.
Hindi would (probably) make a plural by adding -eM (I'm using M to mean a
nasalized vowel); this suffix has a history within Hindi that takes it back
to Sanskrit. This is what makes Hindi different from English, what defines
it.

I was talking about tracing the history of the individual words of a
language. If you trace back all of the words in Vietnamese, and you then
find that an awful lot were borrowed from Ancient Chinese at some point
while almost all of the function words or functional morphemes were not
borrowed and don't seem to ever have had anything to do with Chinese, then
why would you claim that Vietnamese has any genetic relation to Chinese? You
wouldn't. There is nothing 'political' about it - it's based on observation
and the comparative method (which is a lot more rigourous than you seem to
think).

> Fair enough, but note your use here of the word "inherited". I've been
> criticized for saying that there should be a correlation between
> genetics and linguistics: if you trace back different groups speaking
> different languages and assume that they were once one group speaking
> the same language then you would have to argue that they must also have
> a genetic relationship. The only other possibility is that there was a
> language shift, that people, for whatever reason started speaking
> another language. It has been pointed out by many people that this
> does happen when one group is conquered by another (even though I was
> criticized, yet again, for making this very suggestion).

Linguistics doesn't deal with human genetics. You noted my use of the word
'inherited' but I think you misunderstood my usage - I used this term in a
linguistic sense only, as a piece of linguistic jargon. There are lots of
people who confuse this with biology - that's not a linguist's problem. I
haven't studied archaeology or historical / prehistoric anthropology in
anywhere near as much detail as I have linguistics; I cannot presume to say
anything about that. Hitler liked Wagner's music; that's not Wagner's
problem, and it doesn't mean Wagner was a Nazi (chronological accident
perhaps, but the point stands). In the same way, I can set up genetic
relations between languages using methods that I can test and justify; if
you want to turn that into some argument about genetics, I wash my hands of
it. I'm not interested, and I won't answer for any others on the group (I
remember the arguments too).

> Now, in the case of English, we are descended from Angles and Saxons,
> Germanic tribes, so it is natural to say that the oldest English words
> are from German, even if most of the modern words were borrowed from
> other languages, mostly French and Latin. Fair enough.

When you say 'most of the modern words' you mean content words only, not
function words/morphemes.

> But in the
> case of people in South America, for example, who speak Spanish but are
> presumably descendents of local people, couldn't we argue that they are
> still speaking their traditional language but that they have borrowed
> vocabulary, grammar and pronunciation from Spanish?

You could argue it; your argument would probably be unsuccessful at
convincing others (including myself). If there is nothing of their local
language in their current speech other than a few content words, they are
speaking a different language.

> This is an extreme
> example. I'm just asking if vocabulary can be borrowed then why not
> everything else?

With prolonged contact, the pronunciations of languages can change to
resemble each other more (Armenian has a phonology like its neighbouring
Kartvelian tongues with things like glottal ejectives; tones developed in
southeast Asia through convergent development with Chinese); grammatical
structures can tend to converge and mirror each other (can't think of many
examples off the top of my head - need sleep, been up all night watching
Almodovar and cleaning - but there is the use by Vergil of passive
participles with an accusative direct object, an obvious mirroring of the
Greek middle voice) and, to be honest, even a few random function words can
travel (Hindi often uses 'lekin' which ultimately has an Arabic origin to
mean 'but').

You are trying to get me to define a line between borrowing aspects of a
language and borrowing the entire language itself; at least, I suppose you
want me to fail in that attempt, so that I will 'see the light' and accept
that borrowing is a more important descriptive tool than inheritance. You
want me to conclude that English is a Romance language or something
ridiculous like that. I'm not about to do it, but I'm too tired to provide a
watertight argument right now. I'm drained from a busy term; it's Holi
tomorrow and Saturday; I'm going to go recharge at my grandmother's. Then
I'll take you on, k? :D

Sorry, I hope you don't take me as being rude; I do enjoy these friendly
discussions that challenge my foundations. Making me restate arguments that
I've internalized for so long is only a good thing for my intellecutal
advancement, so thanks to you and Comm for playing devil's advocate.

> In which case we are all speaking whatever language
> we inherited from our ancient ancesters plus any elements that were
> borrowed along the way. We would then expect a 100% correlation
> between genetics and "inherited" as opposed to "borrowed" language.

I won't talk about human genetics - it's not my field.

> On the other hand, related languages should sound similar AND have
> words in common and should therefore it should be easier for people
> speaking realted languages to understand each other.

I'm still completely mystified as to why you insist that they should sound
similar, as if sound changes don't exist! Does a native English speaker from
rural Scotland sound anything like one from New Delhi to you? Or consider:
Panjabi is a tonal language, with phonemic tones (it has three tones). Hindi
has no tones, neither does Gujurati. That's a pretty big difference in how
the languages sound, and yet it in no way makes them any less related than
they were before Panjabi developed those tones.

Here's another one for you. English 'wheel' and Sanskrit 'cakra' (the <c>
stands for an English <ch> sound) are pretty dissimilar in terms of their
sounds. And yet, each and ever sound in both words is in a regular
correspondence. English 'wh' regularly corresponds to Sanskrit 'c' or 'k'
depending on the vowel following; English 'e' regularly corresponds to
Sanskrit 'a', a Sanskrit internal 'k' regularly corresponds to a blank in
English, and English 'l' regularly corresponds to Sanskrit 'r' or 'l'
depending on the variety of Sanskrit. There are hundred of other examples
for each of these correspondences. Moreover, the correspondences occur in
the words and morphemes that we are certain each of the language inherited,
but not in the ones that they borrowed (depending on when they were
borrowed). They don't sound similar, but they are thus most certainly
related.

> But pronounciation, the use of tone and rhythm, these are what
> determine how a language sound. Surely these things are inherited (or
> borrowed as the case may be) along with vocabulary and grammar, so that
> you would EXPECT realted languages to sound similar. I suggested that
> could be a "prejudice".

Pronunciations change; use of tones changes; rhythms change. I just
mentioned Panjabi has developed tones. Latin went through a period of having
a heavy stress on the first syllable of every word; by the time it developed
into Romance it had a completely different stress pattern. As for
pronunciation, I've mentioned the divergence between the UK and the USA,
where one has changed its pronunciation of 'r' in a post-nucleic position in
a syllable. All of these types of changes happen; over time, they add up.
But Glaswegians, Bostonians, and people from Sydney all speak related
languages nevertheless.

Neeraj Mathur

Relevant Pages

  • Re: Related languages (Re: A China-Sumer connection)
    ... > comment that if you trace back the pronounciation of words in Chinese ... > Vietnamese belong to unrelated language groups. ... So in English, the plural of 'computer' is formed by adding ... language and borrowing the entire language itself; ...
    (sci.lang)
  • Re: A beautiful Gaelic song by Wiliam Ross
    ... The advantage of a phonetic language is of course beyond rubies, ... When speaking English the only rule is that to be understood, ... The Chinese languages, of which there are many, with variations spoken ... Get the tone wrong and you ...
    (soc.culture.scottish)
  • Re: singapore identity
    ... It is an English language paper which is top heavy with Indians and half-Indians. ... the Indian bias is shining. ... Chinese language, perfectly, at work, at play or at home. ...
    (soc.culture.singapore)
  • Re: English as a creole.
    ... simply means there was no such thing as tomahawk. ... quadrilingual (English, Norse, Norman French, Latin). ... A century of post-Conquest language change ... which (along with the borrowing already ...
    (sci.lang)
  • Re: The "spinoza" programming language
    ... anything, it shows in natural language a distinct entropy as seen, ... I have heard that when books, particularly science books, are translated ... using long strings of Chinese characters. ... English or another IE language before beginning their studies of CS and ...
    (comp.programming)