...past S,4 with the expectation, then, that such languages should show OVSX and VOSX orders. I conclude that specifier-head-complement, and not the reverse,...order available to the subcomponents of a phrase. Consider again, in this light, (4). The conclusion just stated that SHC order is the only one available...

...categories for which head- specifier order is cross-linguistically predominant. Therefore, Kayne concludes that "specifier-headcomplement, and not the reverse,...order available to the subcomponents of a phrase" (p. 36), which in turn means that asymmetric c-command maps onto precedence, not subsequence. The choice...