En réponse à David Peterson <[log in to unmask]>:

>     Let me reproduce a list my pidgins and creoles teacher gave us about
> what
> a language needs in order to be a language:
> What Does a Language Need?
> 1.) Definite/indefinite opposition (possibly via zero marking of one)

Latin did very well without marking that distinction. Japanese too. And I don't
think you can call these languages pidgins.

> 2.) Nouns


> 3.) Adjectives (although in many languages the class is very small, with
> most
> property items as verbs)

In languages where they are identical to nouns or verbs, I think you can say
that there are no adjectives (Occam's razor: don't make unneeded distinctions).
If what you often translate as adjectives in Western languages behave in
language X exactly like nouns, or exactly like verbs, I don't think you need to
call them adjectives. They are just nouns or verbs.

> 4.) Verbs (claims that certain Native American or Southeast Asian
> languages
> have no disctinction between nouns and verbs is at present
> controversial.)

Nootka can put any kind of affix on any kind of word, even those who are
translated by prepositions in Western languages. Unlike Mandarin, it doesn't
even restrict the use of some marks to some kinds of words. In other words, in
Nootka there is no grammatical way to separate words into categories. The only
existing categories are semantic.

> 5.) A dative/benefactive marking strategy


> 6.) An oblique case marking strategy

Do you mean, case/preposition/postposition or like? I guess I agree.

> 7.) A plural marking strategy (although only used emphatically in many
> grammars)

If you mean by that that no language lacks a word for 'several', I agree. But I
disagree that a language MUST have a plural marking strategy other than the
possible use of quantity words (or numerals).

> 8.) Pronouns for three persons (there are languages which do not
> distinguish
> number in pronouns)

And languages who don't have distinct personal pronouns for the 3rd person.
Latin had to use deitics. Of course, if you say that the deictics are part of
the personal pronouns, then the question is trivial. Only the opposition I-you
seems necessary. So I'd say that languages must have pronouns for the first two
persons (of course, those pronouns can be former nouns that were specialised,
like in Japanese).

> 9.) A proximal and a distal demonstrative

I disagree. In French for instance the distinction is optional and has to be
done with adverbs. The demonstrative itself doesn't make the distinction.

> 10.) Spatial deictics (or nominals used in this function)

If you mean, equivalents of adverbs "here", "there", I agree.

> 11.) One general locative preposition

Or postposition, or case mark, or use of a nominal for it. Agreed.

> 12.) One modality marker of obligation and one of probability

Then again, it can be done semantically. It doesn't need to be grammatical.
French "devoir" and "pouvoir" don't behave differently from any other verb.

> 13.) Causative marking

Also can be done semantically.

> 14.) A subordination strategy (not all languages have a distinct
> relativization strategy)

Which can be as simple as full nominalization of the subclauses. I don't know
if you can call that subordination anymore. But I guess you're right.

> 15.) Adverbs

Disagree. Adverb is the most fuzzy and unprecise category. Then term 'adverb'
in Western languages refers to things very different from each other:
circonstancial complements (e.g. where, yesterday, well), determinators of
adjectives or other adverbs (e.g. *very* beautiful, early *enough*), terms
completing nouns (e.g. '*even* John is sleeping'), terms completed by
prepositions (e.g. from *here*), enonciation words (e.g. maybe), connectors
(e.g. however). No common criterion can put those different things in one
category. And some languages go very well without them (complements can be
replaced by nominal phrases - the origin of most 'adverbs' in Western
languages -, affixes can work too - -er in happier does the job as good
as 'more' in 'more beautiful', and german and Dutch go well with only it. When
completing nouns, Latin used adjectives and other nouns for things like 'a lot
of', 'more', etc... -). So I don't think such an undistinct category can be
considered essential.

> 16.) A focus marking strategy
> 17.) A topic marking strategy

Agreed for both.

> 18.) Question words (e.g., WH-words)

Which may be identical to other kinds of words (nominals, demontratives, verbs -
 Comox has a verb meaning 'to be what?' -, etc...). So they are not by
themselves essential, or at least they need not be a separate category.

> 19.) A conjunction "and" (or a word with a broader usage subsuming the
> domain
> of "and")

I would add the necessity of 'or' too (disjunction is as essential as

> 20.) Interjections

It goes without saying :) .

>     That's what my professor, John McWhorter, says every language must
> have
> in order to be a language and not a Pidgin, or, in conlang terms, a
> sketch.

I think this person has quite a narrow vision of what a language can be. A
little Western-centrist vision I'd say.

> I'm interested to see what comments there'll be (if any).

Well, you saw mine :) . They were meant to be constructive critics, so I hope I
didn't offend anyone.


Take your life as a movie: don\'t let anybody else play the leading role.