Flege’s Speech Learning Model predicts that if an L2 learner perceives an L2 speech sound as similar to an L1 speech sound, the two sounds will be combined as a diaphone category, the properties of which will eventually be intermediate between the properties of the L1 and L2 sound. In contrast if the L2 sound is perceived as new, then a new category will be established with properties which may eventually match the properties of the L2 sound. Canadian English has two high front vowels: tense /i/ and lax /I/ differing in spectral and duration properties. Japanese has two high front vowels: long /i:/ and short /i/ differing in duration only. English /i/ and /I/ are expected to be perceived as similar to Japanese /i:/ and /i/, and Japanese learners of English are predicted to
establish diaphone categories. Their identification of English /i/ and /I/ is predicted to initially match their perception of Japanese /i:/ and /i/, but eventually be intermediate between the native norms for the L1 and L2 categories. Spanish has one high front
vowel. Spanish learners of English are predicted to perceive English /I/ as less similar
to Spanish /i/ than English /i/, and are predicted to eventually establish a new /i:/
category. Their identification of English /i/ and /I/ is predicted to initially be poor but
eventually match that of English listeners. These predictions were tested using a
multidimensional edited-speech continuum covering the English words /bIt bit bId bid/.
Properties which varied in the continuum included vowel spectral properties and vowel
duration. A longitudinal study was conducted testing Japanese and Spanish speaking learners of English one month and six months after their arrival in Canada. Japanese listeners were found to have a primarily duration-based categorical boundary between English /i/ and /I/ which did not change between the initial and final tests. Spanish listeners did not have a categorical identification pattern in the initial test, but they did establish duration-based or spectrally-based categorical boundaries by the time of the
final test. Results were therefore consistent with the theoretical predictions.
This paper proposes a phonological analysis for vowel devoicing in Tokyo Japanese using the framework of Optimality Theory (Prince and Smolensky, 1993). Generally speaking, in Japanese the high vowels /i, u/ are devoiced when they
occur between two voiceless consonants. However, there are some contexts where such a simple generalization does not hold, e.g., so called “word-final devoicing” and when devoiceable vowels are accented. This paper attempts to provide a
unified analysis for such issues as well as for the canonical context.
The Japanese motion verb kuru ‘come’ can be used as either a lexical verb or an auxiliary verb. The traditional analysis of kuru involves mere itemization of its various senses (Koga, 2001). An image schema has yet to be presented to characterize its diverse instantiations. This paper undertakes this goal within the framework of Cognitive Grammar, which assumes that linguistic expressions of a single item are routinely polysemous (e.g., Langacker, 1987 & 1991). A network model and a source-path-goal (deictic center) schema are proposed in this paper to capture the relatedness of the various senses of this item. The basic (prototypical) meaning of kuru involves the theme moving toward the speaker's position (vantage position) along a spatial path. Usually the path and the final stage are profiled; however, variations (non-central meanings) of this verb do exist due to profiling different parts of the path or different stages of the motion. Three variants of this verb have been identified in this study. The other senses of kuru as a lexical verb are mainly metaphorical extensions. As an auxiliary verb, kuru mainly serves directional and aspectual functions. Cognitive domain selection or shifting (e. g. extending from the spatial domain to the temporal domain on some perceived commonalities) plays an important role in motivating semantic extensions of kuru, either as a lexical verb or as an auxiliary verb. This paper supports a semantic network model by demonstrating that all the senses of kuru are far from random (Hamada, 1989) and that all the extensions are related to the basic meaning of kuru directly or indirectly through family resemblance. It is human conceptualization of phenomena (mainly metaphor) that directly motives the extensions including the grammaticalization of a particular item.
This study discusses the use of a population test as an empirical method in exploring
semantic content of near-synonyms for use in electronic dictionaries. Chapter 1
reviews some of the problems of a conventional dictionary, and suggests how an
electronic dictionary could meet these challenges. Current lack of semantic
information in linguistic literature hampers the development of electronic
dictionaries, which has raised an urgent need to study the implicit knowledge of
native speakers. Chapter 2 describes the present study, which aims at exploring what
types of semantic information can be obtained with population tests. In this study,
the test field comprised of twenty-one Finnish verbs all used to describe a
complaining speech act. Many of these words are defined as synonyms in mono- and
bilingual dictionaries, and many of them are also classified as expressive
(onomatopoetic-descriptive) words, which are especially numerous in the Finnish
language. The test population (informants) consisted of 154 (16-18 yrs.; 95 women)
native speakers of Finnish. Five semantic features (gender and age of the agent, level
of anger, volume of voice, and furiousness of the patient) were tested with multiple
choice and open-ended tasks. Chapter 3 discusses the results of this study in the
context of their potential use in electronic dictionaries. Population test methodology
per se will also be discussed. It seems that population tests are able to give remarkable amount of new information to objectively distinguish near-synonymic words from each other. This test type could offer effective tools for exploring the dimensions of semantic contents of words, which would directly serve in construction of electronic, multidimensional dictionaries.
English noun-noun compounds are often translated into Russian as relational adjective-noun
constructions with the adjective parallel in function to the non-head noun of a compound. However, a large subclass of English compounds which are sometimes referred to as ‘deverbal’ do not have a relational adjective-noun equivalent in Russian. In deverbal compounds (e.g. van driver), as opposed to so-called ‘root’ compounds (e.g. bookstore), the head noun is derived from a verb and the non-head noun is interpreted as an internal argument of the head noun. In Russian, the same meaning is expressed by means of a genitive construction. It is proposed that this restriction is due to the morphological difference between English compounds and Russian relational
adjective-noun constructions. Following Chomsky (1970) and others, nouns derived from transitive verbs may retain the internal arguments of their base verbs. Kayne (1981) shows that internal arguments may only be expressed as DPs and never as APs. Assuming that internal arguments are realized within the lowest maximal projection of
their heads (Williams, 1981), adjectives may not express internal arguments because they are adjuncts and as such are realized outside of the lowest maximal projection. Thus, in English deverbal compounds, the non-head member can express an internal argument, whereas it is impossible in Russian since the non-head member of the construction is an adjective.
This study reports the experimental data dealing with differences of the
acquisition sequence of Japanese secondary depictive predicates and
affectedness of Japanese verbs between Japanese speaking young children (L1
acquisition) and Mandarin Chinese speaking adults learning Japanese (L2
acquisition). The results support Schwartz and Sprouse’s (1996) and Eubank’s
(1996) hypothesis that L2 grammar is derived from learners’ L1 grammar at an
early stage of L2 acquisition.
This paper reports on a study carried out on referential communication strategies. It used
the theoretical framework of Levelt’s (1989) model of L1 speech production in its application to L2 (DeBot, 1992). The study investigated the underlying processes of utterances of adult speakers who had to solve a referential communication task in L1 and L2. Two groups of participants were formed; the first group acquired the L2 in a second language environment (SLA) and the second group learned the L2 in a foreign language classroom (FLA). Although the distinction between SLA and FLA is significant in any type of L2 acquisition (Ellis, 1994; Rösler, 1995; Edmondson, 1999; Lightbown, 2000),
it has not yet been considered in the heoretical framework mentioned. Results show that the process of generating a comprehensible message in referential communication by L2 adult speakers is influenced by the L2 environment.