Re: person specification (polite/familiar)

From: Neeraj Mathur (neemathur_at_hotmail.com)
Date: 02/17/05 

Date: Thu, 17 Feb 2005 12:34:43 -0000

"John Atkinson" <johnacko@bigpond.com> wrote in message
news:OPAQd.162921$K7.99150@news-server.bigpond.net.au...
> But in Bantu languages, for example, gender (noun class) of subject and
> object *does* have a relationship with forms of the finite verb. The
> prefixes involved go into the same slots and behave in just the same way
> as
> those which are used when the subject or object is 1st or 2nd person.
>
> Even if you insist on sticking to languages "of the Indo-European type",
> there are cases where gender does "have a relationship with forms of the
> finite verb". Admitedly, it's mostly restricted to particular tenses
> (past
> in Russian and, IIRC, Hindi), and the gender suffixes don't work in
> exactly
> the same way as the person suffixes.
>
> In any case, if you're trying to establish as a linguistic universal that
> 'person' categories *can't* be based on gender, surely it's cheating to
> restrict yourself to just one family.

With regards to the IE languages you've mentioned, the relationship is not
between gender and forms of the finite verb, it's with participles which are
morphologically adjectives. So in Hindi 'me~ kar raha: hu~' (a present btw),
the participle 'raha:' agrees for gender and number, while the finite verb
is 'hu~' and agrees for person and number.

The Bantu example is quite interesting. I do need to broaden my linguistic
horizons beyond IE.

But I was not trying to 'establish as a linguistic universal that 'person'
categories can't be based on gender'. I was trying to demonstrate that the
person categories are not ad hoc creations of grammarians but are actually
pschologically real to native speakers of those languages that have them in
their grammars. This is because I was responding to a claim that 'The whole
concept of "1, 2, 3" is largely a construct of grammarians,' to which it was
responded that this was created to 'describe a factual phenomenon of the
language'. I am trying to show more support for the idea that the
'construct' of gender is in fact motivated by a description of a feature
that is present subconsciously in the grammars of certain languages. This
feature is not arbitrary in these languages but based on what is present in
the grammar - thus gender categories are separate from person categories in
the languages that were being discussed. I am not interested (here) in
linguistic universals.

If in the Bantu languages (or other families) the categories are different
and include what IE treats as gender, then that's fine - the job of the
linguist is to determine what category is psychologically real for Bantu.
The end product should be given a label that identifies this feature. I am
perfectly happy to accept that to call this 'person' and interpret that
label in the same way as with IE when describing the language would then be
arbitrary - or just plain wrong.

Thanks for the opportunity to differentiate between descriptive grammar and
linguistic universals.

Neeraj Mathur

Relevant Pages

  • How many proto-languages?
    ... I've never been convinced by the arguments for a single proto-language. ... What I have noticed from the small number of modern languages I have ... Ones without gender, e.g. Finnish ... Holes of ignorance (I know nothing of class and/or gender in Khoi/San, ...
    (sci.lang)
  • Re: Gender in language
    ... You need languages with lots and lots of genders ... English does. ... It may just be that gender reflects something that most humans have in common about the way they construct their mental categories. ... Gender in many European languages is expressed in the morphology of nouns, pronouns, adjectives, and articles. ...
    (sci.lang)
  • Re: Indo-European Languages and Gramatical Gender Loss
    ... plural pronouns as well as masc. ... The more languages we consider, ... about non-Englishcentric grammatical gender, not sex? ... humans since the antecedent could be of mixed gender / sex. ...
    (sci.lang)
  • Re: How many proto-languages?
    ... Nostratic, if it existed, was a *long* time after proto-world. ... If, for example, Bantu languages had 24 classes of which two ... dominant gender. ... Noun classes are widespread in the North-East and North-Central Caucasian ...
    (sci.lang)
  • swahili numbers [was Re: How does Armenian fit into the Indo-European family?]
    ... most other Bantu languages. ... languages it takes prefix li- or di-, which no doubt is their form of PBantu *di-ne-, the class 10 prefix. ... "Kenda", OTOH, seems to have originated as "-enda" in PCentral Bantu. ... Since "kenda" no longer exists in Swahili and neighbouring languages, what's the basis of your statement that it's not inflected there? ...
    (sci.lang)