Wanna Contraction and Prosodic Boundary

The purpose of this paper is to explore that the wanna-contraction is not a purely syntactic phenomenon but a phonological one. Wanna-contraction in terms of a strict syntactic analysis should be reanalyzed in terms of its phonological process. However, it is true that syntactic movements and deletions might affect the location of the prosodic boundary. When want is followed by a moved element, the prosodic boundary comes in place of the moved element. Since want is lengthened with the prosodic boundary immediately following after, contraction is blocked. Wanna contraction is also blocked where there is some missing element immediately following after to. With the prosodic boundary in place of the deleted element, the duration of to is lengthened at a phrase-final or a sentence-final position, and no contraction will take place. Wanna contraction is by nature a phonological weakening or shortening of the function word to. It is sensitive to the lengthening of the syllables at the prosodic boundary and thus no more reducing arises when the function word is stressed or lengthened. I propose that the additional lengthening of want or to at the prosodic boundary may block the wanna contraction. This can be extended to explain all varieties of contractions. It should be also noted that wanna contraction is often found in rapid and causal speech. This also implies that there is no prosodic boundary between want and to when wanna contraction takes place.