The Mora and Syllable Structure in Japanese: Evidence from Speech Errors

Abstract

The goal of this paper is to demonstrate the role and reality of the linguistic unit called ‘mora’ in Japanese through analysis of several types of speech error and other linguistic phenomena. It is shown that the mora is an indispensable notion for the description and generalization of the patterns underlying speech errors in Japanese. Specifically, analysis of blend errors suggests that mora boundaries are the most common switch point in Japanese, which supports the conception of the linguistic construct as being a psychologically real unit of speech production. in addition to the question of the mora, some related questions are also addressed, notably the psychological reality of the syllable and the syllable constituents of Japanese.

Frequency of occurrence for units of phonemes, morae, and syllables appearing in a lexical corpus of a Japanese newspaper

“…Japanese phonemes, morae, and syllables can be portrayed in a single hierarchical structure (Haraguchi, 1996;Kubozono, 1989Kubozono, , 1995Kubozono, , 1999Kubozono, , 2002Kubozono & Ota, 1998;Terao, 2002). As is shown in Figure 1, the lowest phonological level is the phonemic level, representing consonants (C), semivowels (S), and vowels (V).…”

Investigation of phonological encoding through speech error analyses: Achievements, limitations, and alternatives

“…1973;Nooteboom 1969;Shattuck-Hufnagel, 1983Stemberger. 1982; but see Abd-El-Jawad & Abu- Salim, 1987, on Arabic andKubozono, 1989, on Japanese errors). The description of certain errors as sound exchanges implies that the segments' positions are specified independently of the segmenu themselves (see also Fry, 1%9;Garrett, 1975).…”

“…1983b;Stembcrger & Trciman, 1986). but such cases are less frequent than replacements of complete complex constituents (ShattuckHufnagel, 1983(ShattuckHufnagel, , 1986; but see Kubozono, 1989, for counter-evidence from Japanese errors). Occasionally, errors are observed in which the error unit includes nucleus and coda (i.e., the rhyme), or onset and nucleus of a syllable (Nooteboom, 1%9;Shattuck-Hufnagel.…”

Rhythmic Categories in Spoken-Word Recognition

“…This body of work has shown that each of these units plays a role in the way listeners of the languages in question segment spoken input in order to find the words in a continuous speech stream as rapidly and as efficiently as possible. The units are also relevant for many other aspects of language processing; thus work on English, French, and Japanese has shown the importance of stress units, syllables, and morae in language acquisition (Echols, 1996;Inagaki, Hatano, & Otake, 2000;Mehler, Dupoux, & Segui, 1990) and in language production (Cutler & Young, 1994;Ferrand, Segui, & Grainger, 1996;Kubozono, 1989Kubozono, , 1990. Each of the units listed also corresponds to the unit most relevant for describing speech rhythms (and the rhythms of verse forms) in the language in question; from this it is concluded that language-specific rhythmic categories play an important role in the processing of spoken language.…”