Having read this SO answer, I am curious if another supposedly non-configurational language like Wubuy (Nunggubuyu) has been re-analyzed as "configurational". Work on it was done by Jeffrey Health in 1986, and still being used as example by David Gulumbia in 2004, which was after Legate's 2002 dissertation. I quote:

Here even Warlpiri seems not to be the true limit case. Heath (1986) describes Nunggubuyu, an Australian language of Arnhem Land that is ‘‘more radically nonconfigurational than Warlpiri, particularly in lacking clear evidence for a subject- VP split’’ (Heath, 1986, p. 376). The critical phenomena in question, in addition to the discontinuous constituents, free word order, and extensive use of null anaphora familiar from Warlpiri, include ‘‘extremely complex’’ morphology and morphophonemics.

I would be interested in knowing any research work on that. Thanks.

1 Answer 1


It seems that Legate's work is not conclusive and definitive, per Fanselow & Féry:

"Recall that Hale (1983) and Jelinek (1984) proposed an analysis for Warlpiri working with multiple independent modifications of a single argument slot by different autonomous nominal expressions rather than with discontinuous noun phrases in a strict sense. On the other hand, Legate (2002) argues for a configurational analysis of Warlpiri, in the spirit of Rizzi (1997), in which the language would allow discontinuous noun phrases just like German does. The limitations of our data do not permit us to settle such issues."

According to Falk, Legate's approach is circular:

Attempts have been made in the structurally oriented literature to discover indirect arguments for VP in non-configurational languages; one recent example is Legate (2001) on Warlpiri. Typically, these studies demonstrate that non-configurational languages have subject/non-subject asymmetries in some area of the grammar, such as anaphora, and then argue that such asymmetries require a VP node. However, this argumentation is circular: nobody denies the existence of subject/non-subject asymmetries – the disagreement is over how best to express them formally. Since the basic intuition behind constituent structure is that sentences are built out of hierarchically arranged “pieces” (constituents), what is missing is direct evidence for the VP as a “piece” of structure. While it is true that absence of evidence is not evidence of absence, the absence of such evidence in the literature – despite the theoretical importance of demonstrating that non-configurational languages have a VP – is, to my mind, significant.

It would then follow that Legate's work can't be applied with Wubuy, yet.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.