1.Introduction

Definiteness is a grammatical category, often marking the pragmatic category of identifiability, which is important in all languages; however, it is not grammaticalized as a definite marker in all languages. Many linguists have investigated the semantic categories of definiteness in order to account for the uses of definite articles, especially the referential function of the definite article the in English. Uses of definite articles with referents are due to their ‘familiarity’ (Christophersen 1939Christophersen, Paul 1939 The Articles: A Study of Their Theory and Use in English. Copenhagen: Einar Munksgaard.; Heim 1982Heim, Irene R. 1982 “The Semantics of Definite and Indefinite Noun Phrases.” Ph.D. diss. University of Massachusetts.), ‘identifiability’ (Chafe 1976Chafe, Wallace L. 1976 “Givenness, Contrastiveness, Definiteness, Subjects, Topics, and Point of View.” In Subject and Topic, ed. by Charles N. Li, 25–55. New York: Academic Press; DuBois 1980DuBois, John W. 1980 “Beyond Definiteness: The Trace of Identity in Discourse.” In The Pear Stories: Cognitive, Cultural, and Linguistic Aspects of Narrative Production, ed. by Wallace L. Chafe, 203–274. Norwood, NJ: Ablex.; Lambrecht 1994Lambrecht, Kund 1994 Information Structure and Sentence Form: Topic, Focus and the Mental Representations of Discourse Referents. Cambridge: Cambridge University Press. ), ‘unique identifiability’ (Gundel et al. 1993Gundel, Jeanette K., Nancy Hedberg, and Ron Zacharski 1993 “Cognitive Status and the Form of Referring Expressions in Discourse.” Language 69: 274–307. ), ‘uniqueness’ (Kadmon 1990Kadmon, Nirit 1990 “Uniqueness.” Linguistics and Philosophy 13: 273–324. ; Roberts 2003Roberts, Craige 2003 “Uniqueness in Definite Noun Phrases.” Linguistics and Philosophy 26: 287–350. ; Ortmann 2014Ortmann, Albert 2014 “Definite Article Asymmetries and Concept Types: Semantic and Pragmatic Uniqueness.” In Frames and Concept Types: Applications in Language and Philosophy, ed. by Thomas Gamerschlag, Doris Gerland, Rainer Oswald, and Wiebke Petersen, 293–321. New York: Springer. ) and ‘inclusiveness’ (Hawkins 1978Hawkins, John A. 1978 Definiteness and Indefiniteness. Atlantic Highlands, NJ: Humanities Press.). The form of the referring expression depends on the cognitive status of the referent. Identifiability involves a representation that is shared in the hearer’s and speaker’s mind, whereas specificity of an indefinite noun phrase involves a representation that exists only in the speaker’s mind (Lambrecht 1994Lambrecht, Kund 1994 Information Structure and Sentence Form: Topic, Focus and the Mental Representations of Discourse Referents. Cambridge: Cambridge University Press. , 77–80). The cognitive category of identifiability does not always correspond to the grammatical category of definiteness. The correlation between the (un)identifiability of a referent and the (in)definiteness of the noun phrase is different in different languages. Chafe (1974Chafe, Wallace L. 1974 “Language and Consciousness.” Language 50: 111–133. , 1976Chafe, Wallace L. 1976 “Givenness, Contrastiveness, Definiteness, Subjects, Topics, and Point of View.” In Subject and Topic, ed. by Charles N. Li, 25–55. New York: Academic Press) attempts to connect ‘givenness’ and ‘consciousness’. Givenness is defined as the knowledge that the speaker assumes to be in the consciousness of the addressee. So-called new information is what the speaker assumes they are introducing into the addressee’s consciousness by what they say (Chafe 1976Chafe, Wallace L. 1976 “Givenness, Contrastiveness, Definiteness, Subjects, Topics, and Point of View.” In Subject and Topic, ed. by Charles N. Li, 25–55. New York: Academic Press, 30). Chafe discusses how long givenness lasts and considers that givenness is fundamentally a matter of the speaker’s belief that the item exists in the hearer’s consciousness. Something previously treated as new may be later treated as old. Nominal referring expressions are ranked on a continuum from highest accessibility to relatively lower accessibility. Definite descriptions involve low accessibility because definite referents are not highly active in one’s consciousness and usually refer back to relatively distant antecedents. Although previous studies on referential forms have not been challenged, they do not explain all uses of definite articles.

This study on definite markers in Longxi Qiang demonstrates that identifiability is neither necessary nor sufficient for the appropriate use of definite markers. In addition to the definiteness encoded by definite articles, some scholars have noticed the discourse functions of definite referring expressions. Epstein (2002)Epstein, Richard 2002 “The Definite Article, Accessibility, and the Construction of Discourse Referents.” Cognitive Linguistics 12: 333–378. argues that neither of the two main approaches to definiteness, familiarity or unique identifiability, provides the necessary or sufficient conditions for the use of the definite article the in English. He further determined that speakers often choose the even when they know that the hearer is not yet able to identify the referent. In addition to unique identifiability and familiarity, the article the is compatible with other functions, including discourse prominence, role/value status and point-of-view shifts. Article selection is an aspect of the active, dynamic process in which speakers induce hearers to accept the referents into the discourse under distinct guises. Apothéloz and Reichler-Béguelin (1999)Apothéloz, Denis, and Marje-José Reichler-Béguelin 1999 “Interpretations and Functions of Demonstrative NPs in Indirect Anaphora.” Journal of Pragmatics 31: 363–397. describe non-coreferential demonstrative NPs by applying a dynamic model for the production and reception of discourse reference. Furthermore, it has been shown that demonstrative NPs have various discursive and communicative functions or effects, such as avoiding a generic interpretation, signaling a problematic presupposition, and signaling that identifying the intended referent will be somewhat difficult.

This study focuses on the conditions for the uses of the definite marker -tì in Longxi Qiang, a Tibeto-Burman language spoken in China. In Longxi Qiang, there are two definite markers -tì and -ʨì, and one indefinite marker -kà. The definite marker -tì appears after the head noun, and sometimes after a verb for discourse functions, while the indefinite marker either precedes or follows the head noun. -Tì can be used with common nouns and proper names. This article primarily focuses on the uses of the definite marker -tì, which developed from the distal demonstrative -thì ‘that’. In Section 6.3, the referential and discourse functions of -ʨì, a definite marker developed from the proximal demonstrative ʨì ‘this’, is discussed. Recorded data show that -tì occurs far more frequently than -ʨì. However, lower frequencies do not equate with less importance.

This paper is organized as follows. Section 2 briefly introduces the data collection and methodology adopted in the study, while Section 3 reviews the previous studies on approaches to definiteness. Section 4 analyzes the types of definiteness with the definite marker -tì, and Section 5 discusses the discourse functions realized by -tì. Section 6 provides evidence to support the active and dynamic aspects of -tì in communication. Finally, Section 7 presents the conclusion of the paper.

2.Data and methodology

The Longxi Qiang language belongs to the Southern Qiang section of the Qiangic branch of the Tibeto-Burman family of Sino-Tibetan stock (Sun 1981Sun, Hongkai 1981 Qiangyu Jianzhi [A Brief Description of the Qiang language]. Beijing: Nationalities Press.). According to the data collected in the 1990s, 3,300 native Longxi Qiang speakers were scattered throughout Longxi Township, Wenchuan County, Aba Tibetan and Qiang Autonomous Prefecture in Sichuan Province, China (Liu 1998Liu, Guangkun 1998 Mawo Qiangyu Yanjiu [Studies on the Mawo dialect of the Qiang language]. Chengdu: Sichuan Nationalities Publishing House.). Longxi Qiang is endangered and underdescribed. No Longxi Qiang monolingual speaker was found in Wenchuan County during my fieldwork. Even those elders who are older than ninety are bilingual. Because the teaching languages in local primary and secondary schools are Standard Mandarin and Wenchuan Mandarin, an increasing number of young people cannot speak Longxi Qiang (fluently). Typologically, Longxi Qiang is a verb-final, agglutinative language, in which its verbal morphology includes person, aspect, mood, direction, causation, evidentiality and negation morphemes.

The data of Longxi Qiang cited in this article are from a corpus composed of ten hours of transcribed natural data, including two hours of monologues and eight hours of dialogues. From 2013 to 2018, data were collected from Madeng Village, Longxi Township,11.On 18th December 2019, the Department of Civil Affairs of Sichuan Province decided to combine Keku Township and Longxi Township and name it Bazhou Township. Wenchuan County, Aba Tibetan and Qiang Autonomous Prefecture, Sichuan Province, China. During the data recording, I worked with nine participants, five females and four males, whose ages ranged from forty to eighty-two years. More information regarding the participants is provided in Appendix B. Although native Longxi Qiang speakers can communicate with Wenchuan Mandarin monolinguals from other villages in Wenchuan County, they are unable to communicate with speakers of other varieties of Qiang in Wenchuan County, i.e., the Miansi and Yanmen varieties of Qiang.

In this paper, an interactive understanding of communication is adopted in which the speaker uses certain forms based on their inference of the hearer’s knowledge and the context of the utterance. I analyze the definiteness of -tì from the perspectives of identifiability and the uses of -tì in various situations, for example, even if both parties cannot identify the referent; referents with the same identifiability present either presence or absence of -tì.

3.Previous approaches on felicitous uses of definite articles

The concept of familiarity accounting for the use of the was first presented in Christophersen (1939)Christophersen, Paul 1939 The Articles: A Study of Their Theory and Use in English. Copenhagen: Einar Munksgaard.. The English definite article the is used when the referent denoted by a noun phrase is familiar to both the speaker and the hearer. In contrast, the English indefinite article a is used in the case that the referent may be familiar to only the speaker. Christophersen (1939Christophersen, Paul 1939 The Articles: A Study of Their Theory and Use in English. Copenhagen: Einar Munksgaard., 79) admits that familiarity theory is not always correct. For example, when talking of a certain book, it is correct to say “The author is unknown”. Hawkins (1978Hawkins, John A. 1978 Definiteness and Indefiniteness. Atlantic Highlands, NJ: Humanities Press., 106–129) presented different usage types of definite articles based on Christophersen’s work. The first use of the definite article is an anaphoric use, which is a situation in which an entity is a member of the previous discourse set, namely, it has been talked about before.

The concept of familiarity fails to explain the uses of definite articles in some cases. Because of such considerations, many scholars prefer the approach of identifiability to familiarity to discuss definiteness. As Chafe (1976Chafe, Wallace L. 1976 “Givenness, Contrastiveness, Definiteness, Subjects, Topics, and Point of View.” In Subject and Topic, ed. by Charles N. Li, 25–55. New York: Academic Press, 39) notes, ‘identifiable’ would be a better term than ‘definite’. The case is not just “I assume you already know this referent”, but also “I assume you can pick out, from all the referents that might be categorized in this way, the one I have in mind”. Chafe (1996Chafe, Wallace L. 1996 “Inferring Identifiability and Accessibility.” In Reference and Referent Accessibility, ed. by Thorstein Fretheim, and Jeanette K. Gundel, 37–46. Amsterdam: John Benjamins. , 38) defines ‘identifiability’ as follows: “Speakers judge that the words they use to express it will enable the listener to identify it”. Gundel et al. (1993)Gundel, Jeanette K., Nancy Hedberg, and Ron Zacharski 1993 “Cognitive Status and the Form of Referring Expressions in Discourse.” Language 69: 274–307. distinguish different referring expressions on the basis of the assumed cognitive status of the referents. Hence, the status of unique identifiability is a necessary condition for all cases of definite reference. Unique identifiability means the addressee can identify the speaker’s intended referent on the basis of the nominal alone.

Identifiability, familiarity and uniqueness are not mutually exclusive. In actual discourse, exclusive unique or familiar readings are rare. These three characteristics are the same thing as those included in accessibility and cognitive status. Extensive studies have been performed on identifiability (for example, Hawkins 1978Hawkins, John A. 1978 Definiteness and Indefiniteness. Atlantic Highlands, NJ: Humanities Press.; DuBois 1980DuBois, John W. 1980 “Beyond Definiteness: The Trace of Identity in Discourse.” In The Pear Stories: Cognitive, Cultural, and Linguistic Aspects of Narrative Production, ed. by Wallace L. Chafe, 203–274. Norwood, NJ: Ablex.; Chafe 1994Chafe, Wallace L. 1994 Discourse, Consciousness and Time: The Flow and Displacement of Conscious Experience in Speaking and Writing. Chicago: University of Chicago Press.; Lambrecht 1994Lambrecht, Kund 1994 Information Structure and Sentence Form: Topic, Focus and the Mental Representations of Discourse Referents. Cambridge: Cambridge University Press. ; Lyons 1999Lyons, Christopher 1999 Definiteness. Cambridge: Cambridge University Press. ). In the present study, the definitions of the cognitive categories ‘identifiability’ and ‘activation’ given by Lambrecht (1994Lambrecht, Kund 1994 Information Structure and Sentence Form: Topic, Focus and the Mental Representations of Discourse Referents. Cambridge: Cambridge University Press. , Chapter 3) are employed. Identifiability refers to a speaker’s assessment of whether a discourse representation of a particular referent already exists in the hearer’s mind or not at the time of utterance, while an unidentifiable referent is when a representation exists only in the speaker’s mind. Chafe (1987Chafe, Wallace L. 1987 “Cognitive Constraints on Information Flow.” In Coherence and Grounding in Discourse, ed. by Russell S. Tomlin, 21–52. Amsterdam: John Benjamins. , 22) distinguishes three types of activation states: active, semi-active (accessible) and inactive. A semi-active (accessible) concept is one “that is in a person’s peripheral consciousness, a concept of which a person has a background awareness, but one that is not being directly focused on”. An inactive concept is one “that is currently in a person’s long-term memory, neither focally nor peripherally active”. The active concept is one “that is currently lit up, a concept in a person’s focus of consciousness at a particular moment”.

Lambrecht (1994Lambrecht, Kund 1994 Information Structure and Sentence Form: Topic, Focus and the Mental Representations of Discourse Referents. Cambridge: Cambridge University Press. , 105–106) argues that there can be no absolute one-to-one correspondence between cognitive states ‘(un)identifiability’ and formal categories ‘(in)definiteness’. Certain unidentifiable referents may be coded with definite NPs, and certain identifiable ones may be expressed with indefinite NPs. For example, even if one assumes that the addressee can identify neither the king’s daughter nor the king himself, the speaker may say ‘I met the daughter of a king’, where the possessee in the possessive noun phrase is marked as definite (the daughter) even if both the possessor and the possessee are unidentifiable in the discourse. Furthermore, the definite form this may have unidentifiable referents. For example, in ‘I met this guy from Heidelberg on the train’, the definite noun phrase this guy expresses the speaker’s intention to add further information about the person in question. The morphologically definite noun phrase this guy is in fact ‘semantically indefinite’ in the sense that it designates a not-yet-identifiable discourse referent. There exist significant exceptions to the correlation between identifiability and grammatical definiteness.

4.Definite marker -tì and indefinite marker -kà

In Longxi Qiang, -kà and -tì are grammatical markers for (in)definiteness, but they aren’t obligatory. Indefinite NPs can denote either a specific or a non-specific referent. A specific indefinite NP is one whose referent is identifiable to the speaker but not to the addressee, while a non-specific indefinite NP is one whose referent neither the speaker nor the addressee can identify at the time of utterance (Lambrecht 1994Lambrecht, Kund 1994 Information Structure and Sentence Form: Topic, Focus and the Mental Representations of Discourse Referents. Cambridge: Cambridge University Press. , 81). The cognitive status of identifiability is marked by the definite marker -tì. I summarize three criteria of definiteness in Longxi Qiang, as shown in (1).

The examples in (2) appear at the beginning of two stories. These two examples illustrate the differences between definiteness and indefiniteness. Definiteness is signaled by the definite marker -tì; indefiniteness is marked by the indefinite marker -kà or numeral-classifier structure. In (2), the newly introduced referent ‘a family’ is marked by indefinite marker -kà and numeral-classifier structure à-kò (line 1). The first-mentioned referent ‘two sons (a pair of sons)’ is expressed by numeral-classifier structure à-tueì (line 3). In the second occurrence of this reference, the definite marker -tì is added to the referential noun of each son (tûbà ‘elder brother’ and tûtsù ‘younger brother’). This story is about the brothers, so both tûbà ‘elder brother’ and tûtsù ‘younger brother’ are marked by topic marker -zù (lines 4 and 6).

The data show that the bare nouns can be interpreted with definite or indefinite reading, which depends on the context. In Excerpt (3), an example cited from ‘The Pear Story’, the bare noun tsə̌ɕìntshé is interpreted with a definite reading, even though it is not marked by the definite marker -tì. In the preceding context, the representation of the referent tsə̌ɕìntshé ‘bike’ exists in the speaker’s and the hearer’s minds, so tsə̌ɕìntshé ‘bike’ is an identifiable referent.

The indefinite marker -kà is also not obligatory. As in (4), the bare nouns are interpreted as indefinite. The unmarked noun sə́ká ‘drug’ is first mentioned in the story, whose representation is only in the narrator’s mind.

Then I tried to clarify the correlation between definiteness and identifiability in Longxi Qiang by looking for the uses of -tì in various contexts. In (5), the speaker H was telling the hearer K how he cooked lunch for his sick father. His father needed to drink cabbage soup at noon. The cabbages were not in front of him. In line 1, the first-mentioned phàtshè ‘cabbage’ would be understood as a generic reference. The second clause (line 3) is a repetition of the first clause (lines 1–2). Both clauses indicate the meaning ‘cabbages are needed’, so these three occurrences of phàtshè are all in unmarked form. In lines 4–5, when the speaker H was describing the process of cooking cabbage soup, he supposed that the hearer already had the mental representation of cabbage in mind. The cabbages were bought in the market.

Example (6) illustrates associative anaphoric uses of the definite marker -tì. In (6), what the speaker is referring to with the pronoun qá ‘I’ is assumed to be identifiable, and then the representation of its associated referent ɕìmì ‘(my) heart’ is also identifiable.

The interpretation of the noun phrase structure ‘noun + numeral + classifier’ depends on how it is used in the context, as given in (7–8). Excerpt (7) appears at the beginning of a story, where the referents of the noun phrases naînaì à-ʨè ‘a grandma’, zə́tsù nə̀-uò ‘two grandsons’ and ʨì à-uò ‘a son’ are newly introduced into the discourse. The addressee is not expected to identify these referents. In this case, an indefinite noun phrase indicates unidentifiability.

In contrast, the numeral-classifier structure nə̀-ʨè given in (8) is not marked as a definite referent, but it is still identifiable to the hearer and the speaker. The reference to the two persons had been introduced in the preceding context.

A headless relative (i.e., a nominalized clause) can provide more information to help the addressee identify the referent. A nominalized clause alone could function as an argument of the predicate. As in (9), the nominalization ìtə̀ ɕyà-mù marked by definite marker -tì and topic marker -zù serves as the single argument of the intransitive predicate lò ‘come’.

5.Prominence marker -tì

6.Discussion: Further support for the discourse function of -tì

In this section, five types of evidence are provided to support the analysis of the discourse functions of -tì: (a) -tì can cooccur with làmò or connectors, both of which have nothing to do with definiteness; (b) omissions of case markers (e.g., genitive marker, nominalizer and locative marker) are induced by -tì for the sake of prominence; (c) the definite marker -ʨì derived from a proximal demonstrative pronoun has similar discourse functions; and (d) the use of -tì as a prominence marker has been borrowed into Wenchuan Mandarin.

7.Conclusion

In this article, I have discussed the functions of the definite marker -tì in Longxi Qiang. The attestations of the definite marker -tì reflect the identifiability of the referent. -Tì functions as a prominence marker. The discourse function of the definite marker -tì in Longxi Qiang provides support for definiteness marking as a source of discourse prominence marking.

Discourse marking is an extended function of definiteness marking. The definite marker -tì is derived from the distal demonstrative thì ‘that’. Similar semantic change is also found in the development of the proximal demonstrative tɕì ‘this’ into a definite marker. One participant pronounced làmò-tì-ʨì (fill-def-this) as làmò-thì-ʨì. I infer from this that -tì is derived from the demonstrative pronoun thì ‘that’.

The scope of -tì is over discourse, not restricted within a proposition. The use of definite marking is based on the context. Once a referent is introduced into the discourse, its identifiability status is generally preserved throughout the discourse.

As a discourse marker, -tì indicates the speaker’s communicative intention to make the topic prominent. For example, when -tì is marked on a newly introduced referent at the beginning of a story, it does not encode identifiability. The discourse uses of -tì manifest interactional functions between speaker and hearer. The speaker makes an entity more prominent than other entities by using -tì, and then the hearer would infer the speaker’s communicative meaning from it.