[T-R]

I'm admittedly having a bit of trouble parsing the first sentence of your last paragraph.

When you say "if head-first and complement-first were robust, universal features of languages", do you mean "head-first and complement-first cover the whole set of possibilities for the property 'word order'", or "all languages (within a family?) are one or the other", or something more like "a given language cannot contain both head-first and complement-first structures", or something different?

Also, with "appearances of word orders demonstrating these features are dependent on each other across language families", in what sense do you mean that the appearances would be dependent on each other?

[williamdix]

I mean something along the lines of if head-first and complement-first are the only possibilites for phrase rules and that they must apply to all phrase rules for a given language, i.e.

  for all X, s.t. X is a grammatic category which can be the head of a phrase ; XP := X _

or

  for all X ; s.t. X is a grammatic category which can be the head of a phrase; XP := _ X 

Then, modulo processes which change a sentence from its underlying form to its surface form, we would only see forms characteristic of head-first or complement-first phrase structure.

In the second case, I intend dependent as the article means dependent. That is co-appearing because of the same underlying feature. To use the genetic analogy, co-appearing because one gene (the head firstness gene) determines them.

[T-R]

Thanks kindly, much appreciated.

[T-R]

For anyone having trouble following, the article author's webpage provides a nice introduction:

http://language.psy.auckland.ac.nz/wordorder/