This thesis attempts to describe, and to some extent account for, the syntactic, semantic, and pr... more This thesis attempts to describe, and to some extent account for, the syntactic, semantic, and pragmatic properties of a diverse set of constructions in Japanese in which the verbal suffix TE appears as a linking device. TE is the most frequent and the most versatile connective in Japanese, for it can link constituents at three different juncture levels. Because the semantic relations which can obtain between the conjuncts are heterogene ous, the prevailing view is that TE-linkage is a purely syntactic device without any semantic content, and that the interpreter must infer the intended semantic relation on the basis of extraiinguistic knowledge. However, close examination of TE-linkage reveals clear correlations between its syntax and semantics, with the semantics in large part determined by the syntactic type of the conjuncts and the nexus type of the linkage. Such correlations have been obscured in previous studies because they did not investigate TEconstructions as pairings of form and meaning. Although TE-constructions can theoreti cally be ambiguous as to which units are conjoined, actual ambiguities rarely emerge because native listeners, when encountering a TE-construction, can detect subtle cues for identifying which type of juncture is intended by the speaker. In this thesis such cues are discussed, among them phonological phrasing, argument structure, scope of potential operators, and agentivity of the predicates. It is demonstrated that although the possible semantic relations between the conjuncts are inferrable from the meaning of the con juncts themselves, the grammar must state all such TE-compatible relations as part of the meaning of TE-linkage, for two reasons. First, conjuncts typically hold multiple possi bilities of semantic relation. These possibilities are narrowed down, and possibly reduced to a single option, when they appear in a TE-construction. Even analysts who R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission. F. Potential semantic relations expressed by the linkage: e.g. cause-effect, means-end, conditional, concessive. G. Relationships, if any, between syntactic and semantic structures associated with a particular linkage. The theoretical framework employed in this thesis is a hybrid of Role and Reference
This paper analyzes the ko-so-a demonstratives in Japanese as they occur in experimentally-obtain... more This paper analyzes the ko-so-a demonstratives in Japanese as they occur in experimentally-obtained soliloquy data. In soliloquy, the Japanese deixis system consists of a two-way opposition, viz. ko- vs. a-. So- appears only as an anaphor; its antecedent can be either familiar or unfamiliar to the speaker, contrary to Kuno’s (1973) analysis. It is also argued that ko- and a- are always deictic in soliloquy. Adopting Chafe’s (1994) theory of consciousness, it is hypothesized that (i) a- is used when the referent is in the speaker’s peripheral consciousness, and (ii) ko- is used to refer to an entity if it is already focused at the moment of speech. Finally, it is demonstrated that so- and a- exhibit the attributive-referential distinction, which is proposed by Donnellan (1966).
Preface 1. Introduction 2. Basic concepts in role and reference grammar of Japanese 3. TE-Linkage... more Preface 1. Introduction 2. Basic concepts in role and reference grammar of Japanese 3. TE-Linkage with nuclear juncture: Part I: 4. TE-linkage with nuclear juncture: Part II: 5. TE-Linkage with core juncture 6. TE-Linkage with clausal juncture 7. Concluding remarks References Index.
This paper investigates the cross-linguistic applicability of the concept of frame as developed i... more This paper investigates the cross-linguistic applicability of the concept of frame as developed in the Berkeley FrameNet project. We examine whether the frames created for the annotation of English texts can also function as a tool for the assessment of the accuracy of English-to-Japanese translations. If the semantic structure of a source text is analyzed in terms of the frames evoked by its constituent words and the ways in which the elements of those frames are realized, then those frames, their constituent elements, and their interconnections must somehow be present in the translation. The paper concentrates on passages involving causation, as causal relationships are considered by many to exhibit the most salient differences in rhetorical preference between the two languages.
Since the work of Grice (1975), it has been widely accepted that there are two types of meaning f... more Since the work of Grice (1975), it has been widely accepted that there are two types of meaning for any given utterance: what is asserted and what is implicated. One salient example of the assertion/implication distinction involves the English conjunction and. In and-linked sentences, the and-then reading emerges naturally if the first conjunct can be interpreted as perfective. The same temporal sequentiality also can arise, however, even when the two clauses are co-present paratactically without and. The semantic relation of TEMPORAL SEQUENCE, accordingly, should not be attributed to the conjunction and per se (i.e. asserted), but to such pragmatic principles as the iconicity between clause order and intended temporal order (i.e. implicated). It is commonly accepted that the first type of meaning (assertion) is strictly a property of the sentence and hence is a subject for semantics proper, while the second (implication) should be accounted for by pragmatics. The present study challenges this prevailing view of linguistic meaning by examining TE-linkage in Japanese, a translational equivalent of English and-linkage. 1 Japanese TE, like and, is used to express a diverse range of semantic relationse.g. TEMPORAL SEQUENCE, CAUSE-EFFECT, MEANS-END, CONTRAST. When such a relation is understood to be intended by the speaker, it is always inferable solely from the conjuncts themselves. Furthermore, these relations are cancellable and thus can be regarded as implicatures. Most researchers, therefore, have considered that TE
This paper is a discussion of strategies to express intimacy and exaltation of the addressee-the ... more This paper is a discussion of strategies to express intimacy and exaltation of the addressee-the notions morphologically incompatible in the Japanese honorific system, wherein addressees are marked as either psychologically distant and exalted or intimate and not exalted. Selection of such speech style reflects not only social relationships between the interlocutors (e.g. superior vs. inferior); it also constructs such a relationship. Consequently, the selection is dynamic, and speech styles can shift back and forth during a single span of discourse. These shifts are by no means arbitrarily made; different strategies are required when a superior interlocutor initiates a polite-to-plain shift as opposed to when the inferior does so. In order to account for the latter, recognition of the two kinds of expression-public and private-is needed. Public expression presupposes the existence of an addressee and fulfills the communicative function of language; private expression does not presuppose an addressee and satisfies the non-communicative, thought-expressing function of language. This study demonstrates that when the inferior initiates a polite-to-plain shift, the speech will consist exclusively of private expressions, and that by means of restricting plain expressions to revealing one's private thoughts, the speaker communicates psychological closeness to the addressee without switching to the canonical plain speech style.
Introduction Language is recognized as an instrument of communication and thought. The research e... more Introduction Language is recognized as an instrument of communication and thought. The research exploring language for communication is enormous, and our knowledge of various linguistic devices for communicating effectively has advanced significantly in recent decades. By contrast, exploration of language for thought and thinking is scarce, with the study of private speech in psycholinguistics the notable exception. The Japanese language is equipped with numerous markers of interactional functions that many languages do not have grammaticized, e.g. evidentials, honorifics, interpersonal particles, various pronoun-like elements. For this reason, Japanese has been a focus of attention, especially in the field of pragmatics for many decades. Several years ago, I began to wonder how the Japanese language would look like if such interactional layers were removed, i.e. when it is used to express thought without communicative intent. I then embarked on a study of soliloquy (cf. Hasegawa 2004, 2006a,b). Soliloquy is the utterance of one's thoughts without addressing another individual. It is sometimes thought of as talking to oneself, inasmuch as many researchers consider that speech and thought are always dialogic, e.g. Bakhtin (1984). That is, even in soliloquy the speaking-self and the talked-to-self always exist, and they invariably mirror normal conversational exchanges. However, there may well be profound differences between these two modes of discourse even if soliloquy is essentially dialogic. If the speaker and addressee are identical, there is no gap or discrepancy in the interlocutors' background knowledge; therefore, the speaker is not obligated to consider the addressee's knowledge and perspective, and, as a result, the information structure of soliloquy inevitably differs from normal dialogic conversations. For another example, we could easily speculate on the absence of linguistic politeness in soliloquy because the speaker does not need to consider the possibility of threatening the addressee's face. Analyzing language used in non-canonical settings can serve to inspire the researcher. I am therefore advocating in-depth investigation of soliloquy as a new approach in pragmatics research. As an instance of such research, the present study reports an analysis of the ko-so-a demonstratives in Japanese as they occur in experimentally obtained soliloquial data. The organization of this paper is as follows: Section 2 provides a brief outline of the characteristics of ko-so-a. Section 3 explains my experiment design and data selection. Section 4 examines and analyzes the ko-so-a tokens in the data and discusses their implications. Section 5 reconsiders the notions of deixis and anaphora, and Section 6 concludes this paper.
There is a perceptual difference between male and female speech in fbndamental frequency and, les... more There is a perceptual difference between male and female speech in fbndamental frequency and, less significantly, in formant frequencies. These characteristics are physiologically determined to a great extent, and thus speakers exert little control over these characteristics in their normal utterances. The present study investigates whether some characteristics of hndamental frequency are also manipulated by speakers to make speech quality more feminine. The results of our experiment show that the delay of the FO fall that signals a lexical accent is associated with femininity in Japanese.
Annual Meeting of the Berkeley Linguistics Society, 2014
Proceedings of the Seventeenth Annual Meeting of the Berkeley Linguistics Society: General Sessio... more Proceedings of the Seventeenth Annual Meeting of the Berkeley Linguistics Society: General Session and Parasession on The Grammar of Event Structure (1991), pp. 121-129
Annual Meeting of the Berkeley Linguistics Society, 2014
Proceedings of the Twenty-First Annual Meeting of the Berkeley Linguistics Society: General Sessi... more Proceedings of the Twenty-First Annual Meeting of the Berkeley Linguistics Society: General Session and Parasession on Historical Issues in Sociolinguistics/Social Issues in Historical Linguistics (1995)
This thesis attempts to describe, and to some extent account for, the syntactic, semantic, and pr... more This thesis attempts to describe, and to some extent account for, the syntactic, semantic, and pragmatic properties of a diverse set of constructions in Japanese in which the verbal suffix TE appears as a linking device. TE is the most frequent and the most versatile connective in Japanese, for it can link constituents at three different juncture levels. Because the semantic relations which can obtain between the conjuncts are heterogene ous, the prevailing view is that TE-linkage is a purely syntactic device without any semantic content, and that the interpreter must infer the intended semantic relation on the basis of extraiinguistic knowledge. However, close examination of TE-linkage reveals clear correlations between its syntax and semantics, with the semantics in large part determined by the syntactic type of the conjuncts and the nexus type of the linkage. Such correlations have been obscured in previous studies because they did not investigate TEconstructions as pairings of form and meaning. Although TE-constructions can theoreti cally be ambiguous as to which units are conjoined, actual ambiguities rarely emerge because native listeners, when encountering a TE-construction, can detect subtle cues for identifying which type of juncture is intended by the speaker. In this thesis such cues are discussed, among them phonological phrasing, argument structure, scope of potential operators, and agentivity of the predicates. It is demonstrated that although the possible semantic relations between the conjuncts are inferrable from the meaning of the con juncts themselves, the grammar must state all such TE-compatible relations as part of the meaning of TE-linkage, for two reasons. First, conjuncts typically hold multiple possi bilities of semantic relation. These possibilities are narrowed down, and possibly reduced to a single option, when they appear in a TE-construction. Even analysts who R ep ro d u ced with p erm ission o f the copyright ow ner. Further reproduction prohibited w ithout p erm ission. F. Potential semantic relations expressed by the linkage: e.g. cause-effect, means-end, conditional, concessive. G. Relationships, if any, between syntactic and semantic structures associated with a particular linkage. The theoretical framework employed in this thesis is a hybrid of Role and Reference
This paper analyzes the ko-so-a demonstratives in Japanese as they occur in experimentally-obtain... more This paper analyzes the ko-so-a demonstratives in Japanese as they occur in experimentally-obtained soliloquy data. In soliloquy, the Japanese deixis system consists of a two-way opposition, viz. ko- vs. a-. So- appears only as an anaphor; its antecedent can be either familiar or unfamiliar to the speaker, contrary to Kuno’s (1973) analysis. It is also argued that ko- and a- are always deictic in soliloquy. Adopting Chafe’s (1994) theory of consciousness, it is hypothesized that (i) a- is used when the referent is in the speaker’s peripheral consciousness, and (ii) ko- is used to refer to an entity if it is already focused at the moment of speech. Finally, it is demonstrated that so- and a- exhibit the attributive-referential distinction, which is proposed by Donnellan (1966).
Preface 1. Introduction 2. Basic concepts in role and reference grammar of Japanese 3. TE-Linkage... more Preface 1. Introduction 2. Basic concepts in role and reference grammar of Japanese 3. TE-Linkage with nuclear juncture: Part I: 4. TE-linkage with nuclear juncture: Part II: 5. TE-Linkage with core juncture 6. TE-Linkage with clausal juncture 7. Concluding remarks References Index.
This paper investigates the cross-linguistic applicability of the concept of frame as developed i... more This paper investigates the cross-linguistic applicability of the concept of frame as developed in the Berkeley FrameNet project. We examine whether the frames created for the annotation of English texts can also function as a tool for the assessment of the accuracy of English-to-Japanese translations. If the semantic structure of a source text is analyzed in terms of the frames evoked by its constituent words and the ways in which the elements of those frames are realized, then those frames, their constituent elements, and their interconnections must somehow be present in the translation. The paper concentrates on passages involving causation, as causal relationships are considered by many to exhibit the most salient differences in rhetorical preference between the two languages.
Since the work of Grice (1975), it has been widely accepted that there are two types of meaning f... more Since the work of Grice (1975), it has been widely accepted that there are two types of meaning for any given utterance: what is asserted and what is implicated. One salient example of the assertion/implication distinction involves the English conjunction and. In and-linked sentences, the and-then reading emerges naturally if the first conjunct can be interpreted as perfective. The same temporal sequentiality also can arise, however, even when the two clauses are co-present paratactically without and. The semantic relation of TEMPORAL SEQUENCE, accordingly, should not be attributed to the conjunction and per se (i.e. asserted), but to such pragmatic principles as the iconicity between clause order and intended temporal order (i.e. implicated). It is commonly accepted that the first type of meaning (assertion) is strictly a property of the sentence and hence is a subject for semantics proper, while the second (implication) should be accounted for by pragmatics. The present study challenges this prevailing view of linguistic meaning by examining TE-linkage in Japanese, a translational equivalent of English and-linkage. 1 Japanese TE, like and, is used to express a diverse range of semantic relationse.g. TEMPORAL SEQUENCE, CAUSE-EFFECT, MEANS-END, CONTRAST. When such a relation is understood to be intended by the speaker, it is always inferable solely from the conjuncts themselves. Furthermore, these relations are cancellable and thus can be regarded as implicatures. Most researchers, therefore, have considered that TE
This paper is a discussion of strategies to express intimacy and exaltation of the addressee-the ... more This paper is a discussion of strategies to express intimacy and exaltation of the addressee-the notions morphologically incompatible in the Japanese honorific system, wherein addressees are marked as either psychologically distant and exalted or intimate and not exalted. Selection of such speech style reflects not only social relationships between the interlocutors (e.g. superior vs. inferior); it also constructs such a relationship. Consequently, the selection is dynamic, and speech styles can shift back and forth during a single span of discourse. These shifts are by no means arbitrarily made; different strategies are required when a superior interlocutor initiates a polite-to-plain shift as opposed to when the inferior does so. In order to account for the latter, recognition of the two kinds of expression-public and private-is needed. Public expression presupposes the existence of an addressee and fulfills the communicative function of language; private expression does not presuppose an addressee and satisfies the non-communicative, thought-expressing function of language. This study demonstrates that when the inferior initiates a polite-to-plain shift, the speech will consist exclusively of private expressions, and that by means of restricting plain expressions to revealing one's private thoughts, the speaker communicates psychological closeness to the addressee without switching to the canonical plain speech style.
Introduction Language is recognized as an instrument of communication and thought. The research e... more Introduction Language is recognized as an instrument of communication and thought. The research exploring language for communication is enormous, and our knowledge of various linguistic devices for communicating effectively has advanced significantly in recent decades. By contrast, exploration of language for thought and thinking is scarce, with the study of private speech in psycholinguistics the notable exception. The Japanese language is equipped with numerous markers of interactional functions that many languages do not have grammaticized, e.g. evidentials, honorifics, interpersonal particles, various pronoun-like elements. For this reason, Japanese has been a focus of attention, especially in the field of pragmatics for many decades. Several years ago, I began to wonder how the Japanese language would look like if such interactional layers were removed, i.e. when it is used to express thought without communicative intent. I then embarked on a study of soliloquy (cf. Hasegawa 2004, 2006a,b). Soliloquy is the utterance of one's thoughts without addressing another individual. It is sometimes thought of as talking to oneself, inasmuch as many researchers consider that speech and thought are always dialogic, e.g. Bakhtin (1984). That is, even in soliloquy the speaking-self and the talked-to-self always exist, and they invariably mirror normal conversational exchanges. However, there may well be profound differences between these two modes of discourse even if soliloquy is essentially dialogic. If the speaker and addressee are identical, there is no gap or discrepancy in the interlocutors' background knowledge; therefore, the speaker is not obligated to consider the addressee's knowledge and perspective, and, as a result, the information structure of soliloquy inevitably differs from normal dialogic conversations. For another example, we could easily speculate on the absence of linguistic politeness in soliloquy because the speaker does not need to consider the possibility of threatening the addressee's face. Analyzing language used in non-canonical settings can serve to inspire the researcher. I am therefore advocating in-depth investigation of soliloquy as a new approach in pragmatics research. As an instance of such research, the present study reports an analysis of the ko-so-a demonstratives in Japanese as they occur in experimentally obtained soliloquial data. The organization of this paper is as follows: Section 2 provides a brief outline of the characteristics of ko-so-a. Section 3 explains my experiment design and data selection. Section 4 examines and analyzes the ko-so-a tokens in the data and discusses their implications. Section 5 reconsiders the notions of deixis and anaphora, and Section 6 concludes this paper.
There is a perceptual difference between male and female speech in fbndamental frequency and, les... more There is a perceptual difference between male and female speech in fbndamental frequency and, less significantly, in formant frequencies. These characteristics are physiologically determined to a great extent, and thus speakers exert little control over these characteristics in their normal utterances. The present study investigates whether some characteristics of hndamental frequency are also manipulated by speakers to make speech quality more feminine. The results of our experiment show that the delay of the FO fall that signals a lexical accent is associated with femininity in Japanese.
Annual Meeting of the Berkeley Linguistics Society, 2014
Proceedings of the Seventeenth Annual Meeting of the Berkeley Linguistics Society: General Sessio... more Proceedings of the Seventeenth Annual Meeting of the Berkeley Linguistics Society: General Session and Parasession on The Grammar of Event Structure (1991), pp. 121-129
Annual Meeting of the Berkeley Linguistics Society, 2014
Proceedings of the Twenty-First Annual Meeting of the Berkeley Linguistics Society: General Sessi... more Proceedings of the Twenty-First Annual Meeting of the Berkeley Linguistics Society: General Session and Parasession on Historical Issues in Sociolinguistics/Social Issues in Historical Linguistics (1995)
Uploads
Papers by Yoko Hasegawa