2. Taking a Free Ride in Morphophonemic Learning McCarthy, 2005
The second review which is also relevant to this research comes from John J. McCarthy from University of Massachusetts. The research was written in an
article of Catalan Journal of Linguistics journal volume 4, 2005, entitled Taking a Free Ride in Morphophonemic Learning. The study focuses on the basic learning
of morphophonemic process. McCarthy suggested that in morphophonemic learning, the underlying representations influence the grammar and the grammar
influences the underlying representations. As learners begin to analyze morphologically complex words, they discover morphophonemic alternations for
which the identity map is insufficient. In the article, McCarthy has pursued the idea that learners simultaneously consider various hypotheses about underlying
representation, rejecting any for which no grammar is possible and preferring the one that allows the most restrictive grammar. Morphophonemic learning
necessarily intersects with phonological opacity which presents its own learning challenges. It remains to be seen whether the proposal developed here can be
modified and extended to address this notably thornier problem. In the last review, the study done by McCarthy is about the basic learning
of morphophonemic. The similarity of this study with the present research is on the phonological concerns from analyzing the morphologically complex words.
However, this study is different from the present study because the present research analyzes the morphophonemic process by applying the theories of
morphophonemic rules or changes that will be specified further.
B. Review of Related Theories 1. Theory on Cliticization
According to Stephen R. Anderson in his article entitled Clitics 2010, he clearly explains that the property of being a clitic in this sense is not necessarily a
characteristic of a lexical item, but rather of a phonological form which can realize that lexical item. The same item may well have both clitic and non-clitic
forms. The classic example of this is the case of the auxiliary verbs in English. Many of these have both full, non-clitic forms is, has, had, would, will, etc. and
clitic forms ’s, ’d, ’ll, etc.. From the point of view of the grammar, these are essentially free variants. If a reduced clitic form is chosen to lexicalize the
auxiliary in a given sentence, however, this may result in prosodic ill-formedness, as a consequence of the impossibility of incorporating the prosodically deficient
item into the overall sound structure of the sentence in a well-formed way. Apart from these differential phonological effects, however, the reduced and unreduced
auxiliaries are instantiations of the same grammatical element.
1.1 Theory of Auxiliary Reduction
The first type of cliticization, auxiliary reduction, has been clearly analyzed by Arnold M. Zwicky in his book Auxiliary Reduction in English 1970.
In the paper he examines the English contraction rule Auxiliary Reduction, by way of investigating the extent to which it and rules related to it are dependent
upon information not available in surface structure, or are restricted by complex conditions referring to syntactic and phonological information available in surface
structure. In as much as only a handful of problematic cases are presented in detail
here or elsewhere, it shall not attempt to extend phonological theory to accommodate these facts. This extension must wait upon the accumulation of
further data of similar type. Further explanation regarding auxiliary reduction can be divided as follows.
a. The rule glide deletion This rule drops morpheme-initial [h] quite generally, [w] only in will,
would, was, and were, and [ð] in they, them, than, this, these, that, those, and there. It does not affect [y]. Zwicky 1970:326 briefly mentioned that in slow,
careful speech the rule does not apply. At moderate rates of speech it applies to certain unstressed pronouns and auxiliaries he, him, his, her, have, having, has,
had, will, would, them, and than, and in faster speech it is extended to the other listed forms with [w] and [ð], and to all occurrences of [h] before syllables with
relatively weak stress. The vowel following the [h] bears stress, but relatively little in contrast to nearby syllables. The restriction of glide deletion is applied to
only a few forms in [ð] and [w]. It does not seem to have an explanation in terms of other facts about English, although considerations of general phonological
theory are undoubtedly relevant here, [h] being, in general, more likely to drop than [ð] or [w], and [w] more likely than [y].
b. Initial observations on auxiliary reduction Zwicky 1970:327 stated that his rule and NOT-Contraction are the most
familiar contraction rules of English, the ones regularly represented in the orthography. This section explains that auxiliary reduction applies to eight forms
only: is and has, which reduce to [z]; would and had, which reduce to [d]; have,
Kategori