Review on the Development of Evidentiality and Analysis of Non-Grammatical Evidential Systems

CHEN Zheng,ZENG Xiang-hong

Evidentiality is the linguistic representation encoding the source of the said proposition and indicating its credibility， and also embodies the speaker’s assessment of information sources and the involvement in the said proposition. In the field of typology as rationale， the present study introduces the development of evidentiality in five stages， explores the definitions of evidentiality in the broad and narrow senses as well as grammatical and lexical evidential markers， and then illustrates non-grammatical evidential systems with English and Chinese examples， which provides a perspective for future empirical discourse studies in languages with only lexical evidential markers.

Keywords： evidentiality， evidential markers， English and Chinese evidentials

I. Brief Introduction to Evidentiality

Evidentiality has been extensively studied as a linguistic category coding the grounds behind assertions in typologically different languages. As a primary medium to spread knowledge and provide information， language has various ways of expressing the sources of knowledge and the reliability of information provided within a linguistic unit. The linguistic category that primarily encodes source of information and meanwhile expresses speakers’ evaluation and attitude is called “evidentiality”. Evidentiality is the linguistic coding in which a speaker verifies a proposition by referring to the source of the information and expresses his assessment of the reliability of that information.

II. Development of Evidential Studies

2.1 The Starting Point in 1950s

It is acknowledged by many scholars that evidentiality and evidential markers in languages are in principle an independent category， a semantic-functional domain in its own light， but not a subdomain of epistemic modality. In linguistics， the concept of evidentiality was introduced in the first half of the 20th century， but as a linguistic category， evidentiality was established and recognized attributing to a series of work on Kwakiutl by Franz Boas （1938， 1947） and the first occurrence of the term evidential in his posthumously published grammar of Kwakiutl （1947）.

2.2 The Development from Middle 1950s to Early 1980s

The year 1957 has witnessed the turning point in the history of the study of evidentiality. The publication of Roman Jacobson’s works has started the typological studies of evidentiality. In the book， he describes evidentials as a “tentative lable” for the generic verbal catergory （Jacobson， 1957， p. 392）， and introduces the first basic classification of verbal categories. It was generally acknowledged that in Jakobson’s works evidentiality was clearly distinguished from modality for the first time and became one grammatical category to indicate the source of information in a given proposition.

This period also featured some new findings and insightful works （Aronson， 1967； Haarmann， 1970； Friedman， 1979； Hardman & Barnes， 1981， 1984； Slobin & Aksu， 1982； Giv？n， 1982）， which has laid the foundation for the opening of the first international conference on evidentiality in Berkeley in the spring of 1981.

2.3 A Milestone in the Evidential Study in Middle 1980s

Apparently， it is not until the middle of 1980s that evidentiality has come into the core attention of a larger number of linguists， mainly attributing to the publication of the collected volume of papers from the Berkeley conference （Chafe & Nichols， 1986）. This volume examines evidentiality cross-linguistically from heterogeneous perspectives， and proposes a simplified but reasonable classification of evidential types in diverse languages， which practically becomes the prototype of most， if not all， subsequent studies. The classification and some ideas in both this volume and Willett’s article （1988） gave a new impulse， in one way or another， to the studies of scholars interested in this category. Since then， evidentiality has been firmly established its own status in linguistics and more scholars have studied it from various perspectives ranging from typology to cognitive linguistics， syntax， pragmatics and sociolinguistics.

2.4 The Focus from Old World to New World

The interest in the category of evidentiality has been rapidly growing in the following years， when researchers have not just described language-individual evidentials， but also attempted to do cross-linguistic and typological generalizations， and even explored the relationship between evidentiality and other categories in semantics. Some scholars further have taken up some of the ideas proposed in Chafe and Nichols’ volume and made descriptive analysis in a broader scope and with richer and more authentic data from languages of the “Old World”. These works have sustainably contributed to the development of evidentiality and classification of evidential markers on the whole， which ends up polemic statements with reference to Willett’s classification largely based on materials from the languages of the New World （Plungian， 2010， p. 28）.

2.5 The Monographic Issue of Pragmatics and Related Works for Evidentiality

In 2001， the prestigious Journal of Pragmatics had a monographic issue， which included seven papers on the topic of evidentiality， which were selected from the two panels of the 6th International Pragmatics Conference in 1998. It should be noted that at that time one tendency was to go beyond individual language and its typology to define evidentiality inter-linguistically.

During the same period， Ilana Mushin （2001） explores the discourse pragmatics of reportive evidentiality in Macedonian， Japanese and English through an empirical study of evidential strategies in narrative retelling； Elly Ifantidou （2001） uses Sperber and Wilson’s Relevance Theory to show how evidential expressions can be analysed in a unified semantic/pragmatic framework； the edited volume of Gabriele Diewald and Elena Smirnova（2010） provides further cross-linguistic empirical evidence about languages that have various lexical and grammatical evidential expressions.

Likewise， the edited volume by Aikhenvald and Dixon （2003） and the monograph by Aikhenvald （2004） have made a comprehensive analysis of evidentiality with the exclusive focus on inflectional languages. Due to the authentic materials and crystallized classification， subsequent researches on evidentiality， particularly on that in inflectional languages， will largely draw on Aikhenvald’s monograph.

III. Evidentiality and Evidential Markers

For evidentiality， scholars and researchers are always inconsistent in definition and scope， which practically means that the application of this category to particular linguistic phenomenon has often been lacking in consistency and definitiveness.

Some linguists consider that evidentiality is not as equally important as other grammatical categories， for only “about a quarter of the world’s languages every statement must specify the type of source on which it is based” （Aikhenvald， 2004， p. 1）. And some linguists hold the opinion that as a semantic category， evidentiality may be expressed grammatically， lexically or paraphrastically. The controversy on the above questions inevitably leads to the inconsistency in definition and the semantic characterization of the forms of evidential markers.

For the reason that typologically， some languages， such as English and Chinese， using other devices—that is， lexical evidential markers such as adverbs， propositional or modal verbs， etc.—to mark evidentiality for lack of obligatory morphosyntactic system of evidentials， are out of the core of studies of evidentiality， and studies on evidentiality in these languages have been somewhat sporadic. A more comprehensive literature review on the previous work of these languages has been presented here， and the analysis of the necessity and feasibility of the study focusing on these languages is also attempted. Also a description of evidential system in English and Chinese will be presented as an illustration of lexical and paraphrastical forms of non-grammatical evidential system.

3.1 Definitions of Evidentiality—Narrow and Broad

Different semantic definitions have been formed in individual languages. As it is typically assumed， in order to qualify his commitment to the truth of what is said， the speaker will indicate his judgement of the reliability of truth as well as mark the source of information in the form of evidentials. Therefore， the linguistic property of evidentiality has been characterized on the basis of the relationship among evidentiality， source of information and speaker’s attitude. However， the specification of source of information and that of speaker’s attitude towards the information are not so clear， leading to the inconsistency in defining and scoping evidentiality. In some monographs and thematic volumes， the definition of evidentiality has been discussed in diverse ways and so far， it is obvious that there is not a single and consistent definition of evidentiality.

According to the literature， evidentiality can be mainly defined in two senses： broad and narrow. The main differences between these two approaches lie in whether the core semantics of forms of evidentiality is about to identify the source type of the information or to specify the speaker’s attitude towards the information they convey in a proposition.

The narrow definition of evidentiality restricts this category to the specification of source types of the information. Jakobson （1957， p. 135） describes evidentials as coding “the alleged source of information about the narrated event”. Similarly， Hadumod Bussmann （1996， p. 157） defines that evidentiality is the “structural dimension of grammar that codifies the source of information transmitted by a speaker with the aid of various types of constructions”. Aikhenvald （2004） classifies lexical means for specification of the source of information as forms of extensions of grammatical categories to evidential-like meanings， which refers to“evidentiality strategies”.

The broad definition of evidentiality has been developed by Chafe in a description of English evidential-coding system. In Chafe’s terminology， evidentiality broadly includes marking the attitude of the speaker towards knowledge or reality. He defines evidentiality in a broad sense to include epistemological assessment， that is， any form with evidential-like meaning is inclusively considered as a potential evidential marker， totally beyond grammatical realization. According to Chafe’s model， the specification of the information source is to identify different types of knowledge a speaker has about the world. The ways of the acquisition of different knowledge invariably leads to the differentiation in speaker’s attitude， which also influences the coding of evidential meaning. Therefore， the semantics of evidentiality， in Chafe’s opinion， is to match different types of knowledge with diverse modes of knowing， against the degree of reliability. In other words， the meaning of evidentiality dose not simply include sources of information， but also the speaker’s epistemology of knowledge derived from these sources. Under this interpretation， the specification of sources of information and speaker’s attitude， inextricably interwoven with each other， should be both included in evidentiality.

The entangling relationship between information source and speaker’s attitude is a kind of common awareness among many linguists. Infantidous （2001， p. 2） accordingly defines evidentiality broadly. She includes evidentiality both the marking of the source of knowledge and the speaker’s commitment to the truth of what is being said. And even Aikhenvald， the proponent of narrow definition of evidentiality， has noticed this semantic entanglement between reference to sources of information and indication of speaker’s evaluation.

In the broad sense， evidentiality has taken both grammaticalized evidentials and lexical evidentials into consideration. Since every language has one way or another to specify the source and reliability of information， evidential studies are no longer confined to the languages with grammaticalized evidential systems， in which morphological variations indicate sources of information.

3.2 Evidential Markers—Grammatical or Lexical

All languages have some means of making reference to the source of information or indicating speaker’s degree of commitment to the information. The way or the form of evidentiality structured in languages is called evidentials， or evidential markers （the latter is adopted in this paper）.

Evidential markers are linguistic expressions encoding the information of the speaker’s degree of commitment. David Crystal （1991） defines evidential markers as “constructions express a speaker’s strength of a commitment to a proposition in terms of the available evidence （rather than in terms of possibility or necessity）”. According to the above definitions， evidential markers are exclusively used for illustrating the type of justification for a claim that is available to the person making that claim， whether in grammatical or lexical forms.

As for linguistic realization of evidentiality， scholars holding the broad view cast a wider net to encompass lexical and paraphrastic evidential markers together with grammaticalized ones. Though focusing on studies of grammaticalized evidential markers in different inflectional languageas， Aikhenvald （2003） also recognizes that the source of information may be expressed through non-grammatical elements. Since almost all languages in the world have certain methods to linguistically encode evidential meaning but not all of them have grammaticalized evidential markers， it is probably universal and more practical to have lexical and paraphrastic evidential markers to specify the information source. These lexical means can be of different forms.

IV. Evidentiality in English and Chinese

Evidentiality， like other linguistic categories， enjoys the high universality among many languages and has received a lot of attention in recent literature. Evidential studies focus on either a cross-linguistic study of the topic， or on individual languages. However， being a large part of typological map， many European and Asian languages only have lexical means to specify information sources， but they have diverse lexical means to code evidential meaning. Here we will use English and Chinese evidential markers to illustrate how evidentiality in languages with only lexical means indicates the source of information.

4.1 English Evidential Markers

As a language without grammaticalized evidential markers， English is mapped onto a heterogeneous set of lexical forms， such as adverbs， modal verbs， adverbs or parentheticals etc.， to encode evidential meanings.

In an analysis of spoken and written languages based on evidentiality， Chafe （1986） initially studies English evidential markers. He notices that English doesn’t have grammaticalized evidential markers， so he extends evidentiality to all possible evidentials， independent of any grammatical realization. Accordingly， Mushin’s（2001， p. 56） argues “English lacks in clear grammatical markers of evidentiality， but English compensates for such lack by other identifiable means”. It is important to point out that in her opinion there are many adverbials of“propositional attitudes” in English， such as certainly， probably， obviously possibly， undoubtedly， etc.

According to the above studies， although lack of pure grammaticalized evidential markers， English can still mark sources of information and specify degree of speaker’s commitment in a given proposition in explicitly linguistic forms. For example， modals such as will， should， must， sentential adverbs like undoubtedly， presumably， famously， sentence-initial conjunctions such as but， because， prepositional phrases like as a matter of fact， of course， in reality， and parentheticals such as in my opinion， according to， as we all know， can all function as evidential markers to encode speaker’s evidence for an asserted proposition in English.

4.2 Chinese Evidential Markers

Like English， Chinese is typically a language without morphological evidential markers. In China， this topic has been studied by researehers such as Hu （1994a， 1994b， 1995）， Zhang （1997）， Yan （2000）， Xu （2004）， Niu （2005）， Fang （2005， 2006， 2008）， Zhu （2006）， Wang （2006， 2010）， Le （2011） and etc. Domestic researches on this topic can be classified into four types： introduction of foreign theories and current research orientations of evidentiality， evidential analysis of Chinese evidentiality and evidential markers， application and reanalysis of the nature of evidentiality. Among all these sporadic domestic studies， Jiang Di’s （2005） study is one of the few analysis on languages with grammaticalized evidential markers. Otherwise， Chen Ying’s （2009） monograph is probably the first research book exclusively on evidentiality of Chinese， which provides a critical analysis of Chinese evidential markers and their grammaticalization on the basis of the theory of subjectivity.

In Chinese， different lexical forms of evidential markers will be employed to indicate sources of information and specify degree of speaker’s commitment. The following examples are explicitly linguistic forms of Chinese evidential markers：

In the above example， Chinese people use different lexical means to convey the proposition “She moved to Shanghai” according to the reliability of the information they have. In （a） and （b）， 猜 and 應該是 are two verbs， epistemic verb and auxiliary verb， indicating that the information is a subjective judgement or inference made by speaker on the basis of certain visual or reportative evidence. In （c） and （d）， 估計是 and 確實 are two adverbs of different degrees of certainty， indicating that the information expressed in given proposition is derived from speaker’s inference. In （e）， the evidential meaning is conveyed by the modal particale 吧 （in Chinese， modal particles， such as 的， 呢，嘛，嘍， are also used as evidential markers to convey speaker’s attitude towards the information provided. Some scholars hold the viewpoint that the evidential markers like these are close to grammaticalized evidential markers， or these markers are in the process of grammaticalization.）. In （f） and （g）， 說是 and 據(jù)我所知 are parenthesis with the indication that the information in the proposition is reportative， heard or seen by the third party.

V. Conclusion

In conclusion， the current study has explored evidentiality from the aspects of the development， definition and case analysis， with focus on non-grammatical evidential systems， especially in English and Chinese.

From the previous work， it can be seen that despite the recent surge of interest in studies of evidentiality， no consensus， at least at present， has been reached concerning the definition and scope of evidentiality and its linguistic representation. As it has been mentioned above， scholars hold different opinions on this category and adopt diverse research approaches while analyzing evidential-related phenomena. But we had the following findings which will help us to redefine and reframe evidentiality for future researches.

First and foremost， the term evidentiality used in English or Chinese evidential-related studies refers to a semantic category defined in a broad sense. In other words， evidentiality is the linguistic representation encoding sources of information and indicating its credibility in the said proposition， and also embodies the author’s assessment of information sources and involvement in the proposition.

Secondly， the evidential markers in English or Chinese evidential-related researches are independent of purely grammaticalized realization. It is admittedly that grammaticalized evidential markers are of special concern in the field of typology for its closely morphological correlation with grammar and grammaticalization. However， English and Chinese are both languages lack of morphological devices for coding evidential meanings but with abundant lexical means to mark evidentiality. Actually， in the field of evidential research， many a scholar hold the view that as long as a language employs certain mechanism from its own linguistic system to express evidential meaning， whether grammatically or lexically， even paraphrastically， this mechanism from the perspective of evidentiality is undoubtedly worth researching. From what has been discussed and quoted above， it is safely to state that the forms of linguistic realization should not be the barrier for a semantic category， in particular evidentiality， to be researched in a more comprehensive way among a wider range of world’s languages.

For the focus of English or Chinese evidential-related studies， a relatively broad view of evidentiality should be adopted as the linguistic encoding of information source and speaker’s commitment of a factual claim.

References

Aikhenvald， A. Y. （2004）. Evidentiality. Oxford： Oxford University Press.

Aikhenvald， A. Y. （2003）. Evidentiality in typological perspective. In A. Y. Aikhenvald & R. M. W. Dixon （Eds.）， Studies in evidentiality. Amsterdam： John Benjamins Publishing Company.

Boas， F. （1947）. Kwakiutl grammar， with a glossary of the suffixes. Transactions of the American Philosophical Society， （37）， 201-377.

Bussmann， H. （1996）. Routledge dictionary of language and linguistics. London/New York： Routledge.

Chafe， W.， & Nichols， J. （Eds.）. （1986）. Evidentiality： The linguistic coding of epistemology. Norwood， NJ： Ablex.

Crystal， D. （Ed.）. （1991）. A dictionary of linguistics & phonetics （3rd ed.）. Oxford： Blackwell.

Diewald， G.， & Smirnova， E. （Eds.）. （2010）. Evidentiality in German： Linguistic realization and regularities in grammaticalization. Berlin， DEU： Walter de Gruyter.

Ifantidou， E. （2001）. Evidentials and relevance. Philadelphia， PA， USA： John Benjamins Publishing Company.

Jakobson， R. （1957）. Shifters， verbal categories and the Russian verb （Department of Slavic Languages and Literatures， Harvard University）. （Reprinted in Selectied writings 2： Word and language， R. Jakobson （Ed.）， 1971， pp. 130-147. The Hague and Paris： Mouton）.

Mushin， I. （2001）. Evidentiality and epistemological stance： Narrative retelling. Philadelphia， PA， USA： John Benjamins Publishing Company.

Plungian， V. A. （2010）. Types of verbal evidentiality marking： an overview. In G. Diewald & E. Smirnova （Eds.）， Empirical approaches to language typology： Linguistic realization of evidentiality in European languages （pp. 15-58）. Berlin， DEU： Walter de Gruyter.

Willett， T. A. （1988）. Cross-linguistic survey of the grammaticalization of evidentiality. Studies in Language， （12）： 51-97.

陳穎. （2009）. 現(xiàn)代漢語傳信范疇研究. 北京： 中國社會科學出版社.

江荻. （2005）. 藏語拉薩話的體貌、示證及自我中心范疇. 語言科學， （1）， 70-88.