Towards the use of phonological markedness and extraprosodicity in accounting for morphological errors in Specific Language Impairment

Abstract

Certain grammatical morphemes are variably produced in the speech of children with Specific Language Impairment (SLI). Previous research tends to view this as a consequence of either a deficit in linguistic knowledge or a limitation in processing capacity; however, both approaches raise problems. For example, linguistic accounts are unable to explain why these children’s problems are mostly with production rather than comprehension. Processing accounts, on the other hand, have difficulty explaining why affected children have differing levels of problems with grammatical morphemes that are similar on the surface (e.g. English plural -s vs. third person singular -s). In this paper, a new, phonological account is proposed which avoids these problems, and better captures the wide array of data presented in the literature. It is proposed that children with SLI have problems with organizing segmental data into prosodic structures that are linguistically highly marked, in particular those that involve various forms of extraprosodicity.

L2 Acquisition of a Complex Stress Pattern: UG-Constrained Learning Paths in Khalkha Mongolian

“…In fact, I will demonstrate that this unique acquisition task presents strong evidence that L2 prosodic grammars are constrained by the options made available by UG ( Broselow and Finer, 1991 ; Archibald, 1993 ; Goad and White, 2008 , 2019 ; Özçelik, 2016 , 2018 ), thereby offering support for UG-based approaches to L2 acquisition ( White, 1989 , White, 2003b ; Schwartz and Sprouse, 1996 ). As will be made clear later, in restructuring their grammar, learners of Mongolian consider only UG-constrained options and do not entertain options that are not permitted by UG.…”

“…I hypothesize that whether Khalkha initial prominence, the main focus of this paper, is formally footless or not, English-speaking learners of the language will always have footed representations, as expunging the foot, I assume, is impossible, once it is projected in the L1 ( Özçelik, 2011 , 2018 ). Further, given the Prosodic Acquisition Path Hypothesis (PAPH) ( Özçelik, 2016 ) [see also the Full Transfer/Full Access Hypothesis (FTFA), Schwartz and Sprouse, 1996 ], in initial stages, they will construct feet as in English, that is, right-to-left iterative weight-sensitive trochees, with Extrametricality and Weight-Sensitivity both set to Yes and with moraic codas [see (6) and (12)], even though this will not, on the surface, consistently stress the correct syllables in the target language.…”

“…First of all, if, as we have illustrated above, the default initial prominence in Mongolian is footless intonational prominence rather than (footed) stress, English-speaking learners will eventually need to have footless representations, at least for words that contain no heavy syllables (i.e., initial default stress). However, since representing footless words in an L2 is an extremely challenging task for learners with footed L1s ( Özçelik, 2011 , 2016 , 2018 ) and assuming thus that English-speaking learners of Khalkha will always produce footed outputs, in order to accommodate the L2 input with a footed grammar, one could claim that Khalkha L2ers will possibly be forced to make several UG-unconstrained assumptions about the language, where End-Rule is sometimes set to Right , sometimes to Left and, even more surprisingly, sometimes to Middle , as has already been mentioned in Section “Introduction” and illustrated in (13) below. In fact, for words that that are long enough and only contain a light syllable, Leftmost-Wins, one could say (though see below), will replace End-Rule [see (14)]:…”

“…Even if we assumed initial default prominence in Khalkha to be trochaic (i.e., foot-based), English-speaking learners of the language will face similar challenges, except for having to expunge the Foot, of course. In other words, assuming that they may not ever be able to expunge the Foot ( Özçelik, 2016 , 2018 ), whether one assumes initial prominence to be trochaic or (footless) intonational prominence, they will, in the end, have to make similar rearrangements (which all involve the Foot) in parameter settings in consistently stressing the initial syllables of words composed only of light syllables. More specifically, learners will need to create either a binary trochee that is constructed from the left edge (i.e., with left-to-right foot construction) or an unbounded weight-insensitive trochee that encompasses the whole word, as illustrated respectively in (15a) and (15b) below.…”

“…In addition, this paper also sheds light on the issue of variability in interlanguage grammars, a topic that has recently generated much fruitful discussion in syntax and morphology, particularly with respect to variable omissions of functional morphology (see e.g., Lardiere, 1998a , b ; Ionin and Wexler, 2003 ; White, 2003a ; Ionin et al, 2004 for different accounts of variability), but has received almost no attention in phonology, even though successful phonological explanations have been offered to explain variability in morphology and syntax (see e.g., Goad et al, 2003 ; Goad and White, 2004 , 2006 , 2019 ). Explaining variability in phonology itself is crucial, because along with variability in suppliance of functional morphology, phonological variability in interlanguage grammars is perhaps the leading indication of non-native-like performance, as it is persistent even in end-state grammars (see e.g., Özçelik, 2016 , 2018 ).…”