Janet C. E. Watson, The phonology and morphology of Arabic (The Phonology of the World's Languages). Oxford: Oxford University Press, 2002. Pp. xiii+307

[...]while Baker succeeds in raising some of the Big Questions that lie beneath structural similarities and variations in language, he falls short of providing an apologetic strong enough to sway skeptics to his linguistic worldview. [...]the thematic dimensions that power Leezenbergs account also require a certain degree of conceptual stability. [...]Leezenberg doesnt seem to appreciate the extent to which knowledge of conceptual metaphors might help people determine the thematic dimension needed for metaphor interpretation. 705 JOURNAL OF LINGUISTICS For example, Leezenbergs If art is the tip of the iceberg, Im the part sinking below (Lou Reed) (234) recruits for its interpretation a model of the iceberg in which the top part is visible, but the bulk of the mass is below the surface. According to this analysis, in the computational component it is possible to create symmetrical structures (i.e. structures in which neither of the two elements asymmetrically c-commands the other) but these structures have to be destroyed before the derivation enters the PF component, where a process of linearization applies for physiological reasons, as our vocal apparatus cannot produce more than one word at a time. [...]it is not possible to examine all known cases of movement in a single book; however, the reader is left with the impression that, if the basic device for breaking symmetry is movement plus merge of a new head (when needed), it is always possible to translate movement instances into the dynamic antisymmetric approach. [...]Moro notes that a comparative approach is indeed compatible with his proposal, and that it might exploit the fact that dierent languages can create dierent points of symmetry depending on the structure they use (for instance, 707 JOURNAL OF LINGUISTICS languages not having preverbal subjects will not need to move the wh-item higher than SpecIP and, as a consequence, will not need do-support).