We investigate effects of prosodic cues on interpretation of ambiguous sentences containing relative clauses (RCs) in English by Spanish-speaking learners. English and Spanish differ in default preference for RC attachment: English has a weak low attachment (LA) preference (RC modifies NP2); Spanish has a stronger high attachment (HA) preference (RC modifies NP1). We conducted an interpretation task with auditorily presented stimuli to examine whether prosodic cues determine attachment. Target items were manipulated for position of break and length of RC, NP1, and NP2. For both groups, break and length are significant. For the learners, proficiency interacts with break suggesting L1 transfer: lower proficiency learners choose HA more when break points to LA; higher proficiency learners choose HA more when break points to HA. Lower proficiency learners are more likely to choose LA overall, suggesting a recency effect. Our results confirm the importance of using aural stimuli when testing interpretation of ambiguous sentences.

It is proposed that failure to consistently produce inflectional morphology by Mandarin-speaking learners of English is due to properties of the LI prosodic phonology which are transferred into the interlanguage grammar. While English requires inflection to be adjoined to the Prosodic Word, Mandarin does not permit this structure. Inflection in Mandarin is instead incorporated into the PWd of the stem to which it attaches. It is shown that Mandarin speakers fall into two groups in their treatment of English inflectional morphology. One group of learners is sensitive to the need for a unified analysis of inflection. They recognize that English does not permit a stem-internal analysis of this morphology, but as their grammars do not permit adjunction, inflection is deleted across-the-board. For the other group, inflection surfaces variably, for those stimuli where the shape of the stem enables it to be incorporated into the PWd, as in the L1.

The position that languages require both coda and onset options for the syllabification of word-final consonants is adopted. The latter option is further divided into languages where final consonants are onsets of empty-headed syllables and those where final consonants are syllabified through onset-nuclear (ON) sharing. ON sharing is reserved for languages where final consonants display fortition (overt release): the nucleus hosts the release of the consonant. Empirical evidence from across populations demonstrates that ON sharing is unmarked. It is favoured among the outputs of first and second language learners and individuals with Specific Language Impairment. It is further argued that final onsets are optimal for parsing in end-state grammars, as they demarcate the right word-edge more effectively than codas. Among the two types of onsets, ON sharing is preferred: through the nuclear release, it is better able to host the range of contrasts that right-edge onsets display. The parsing argument serves to illustrate how ON sharing provides an advantage to end-state grammars, beyond being an emergent property from acquisition.

Stoel-Gammon & Cooper (1984) have recently presented phonological analyses in this journal of three children acquiring English. Their general conclusion is that the three children are each following their own unique pattern of acquisition. This paper is a response to that conclusion. We first point out that such research needs to distinguish different possible causes of variation. We then go on to criticize the method of analysis they used, arguing that it disposes the results towards individual variation. Next, we present an alternative analysis, using a different methodology, and show that the children's patterns of acquisition are actually relatively similar. Lastly, we conclude that both their and our results provide evidence for early phonological processing in opposition to the biological model presented in Locke (1983).