zompist bboard

I know that there are typological correlations with head-directionality: for example, head-final languages often tend to be dependent-marking and mark case on NPs, whereas head-initial languages can often be head-marking and rely on verb agreement.

I remember reading somewhere that number systems can be correlated with head-initial vs. head-final as well. Is it the case that languages that are head-final are more likely to list the smaller number in a compound first? (e.g. "nine and twenty")? I am sure I have read that before somewhere but I don't know where.

Just an anecdote: the archetypical head-final language, Japanese, is little-endian(smallest number last) in both its native and Sinitic numeral systems.

Numeral systems are overwhelmingly big-endian, regardless of head directionality, though some languages are little-endian only with regard to tens and ones. Malagasy is the only pure little-endian language I know of.

M Mira wrote:Just an anecdote: the archetypical head-final language, Japanese, is little-endian(smallest number last) in both its native and Sinitic numeral systems.

That makes them big-endian, like most languages.

You're right, I was thinking in a left-to-right order, not distance-from-head, which indeed makes more sense.

Isn't Sanskrit also purely little-endian?