En réponse à "H. S. Teoh" <[log in to unmask]>:

> At least the ancient Greeks had double negatives... interestingly,
> double
> negatives could *either* strengthen or negate the previous negative.
> E.g.:
> 1)      ou'k oi^da  oude'n.
>         not  know-I nothing
>         "I don't know anything" -- the ou'k strengthens the negative.
>         (Lit. "I don't know not-even-one-thing")
> 2)      oude`n  ouk oi^da.
>         nothing not know-I
>         "I know something" -- the oude`n *negates* the negative.
>         (Lit. "I do not know nothing")

Latin did the same: 'non' could be used with other negative words. And again,
depending on the order between non and the other negative word, the negation
could be strengthened or negated. I wonder if it was a borrowing of Classical
Latin from Greek or an independent evolution...

Interestingly, French (at least my idiolect) also has something with negation
and word order. In spoken French, when the second part of the negation follows
the 'ne', this 'ne' can be (and is often) omitted: Il (ne) vient jamais: He
never comes. But when the second part preceeds the 'ne' (which is possible when
this second part is an adverb or a pronoun used as subject of the sentence),
this 'ne' cannot be omitted anymore: Jamais il ne vient: He never comes
(*"Jamais il vient" doesn't sound correct to me, it sounds affirmative to me).