This talk concerns an problematic paradigm in English and other languages that I will call coordinated wh questions (CWh for short). An example of an acceptable CWh is given in (1) below.
(1) What and when did Jane eat?
These sentences are odd in that the superficially seem to involve the coordination of a argument wh-word and an adjunct wh-word. Or, failing that, it appears there is overt movement of multiple wh-words within a single English sentence: an apparent anomaly. In this talk I argue that not only are neither of the above analyses correct, but also that any other syntactico-centric analysis of the construction fails.
The example in (1) will be shown to contrast with examples like those in (2) where the relevant verb is of fixed transitivity (cf the ambitransitive 'eat' above). Further, the particular order of the argument versus the adjunct will be shown to effect acceptability as well. This is shown in (3a-b) where the order of the wh-words is the opposite and both sentences are acceptable (as will be supported by a judgment study)
(2) *What and when did Jane fix?
(3) a. When and what did Jane eat?
(3) b. When and what did Jane fix?
I then posit that the two wh-words bear two different types of dependency between their overt position and their lower position. The difference between these two dependencies is evinced through an online psycholinguistic measure that shows a lack of so-called filled gap effects for the first wh-word (what).