The Anaphoric Use

Demonstratives that make reference to a participant or to a noun phrase are suggested as the tracking use (Himmelmann 1996) or the anaphoric use (Diessel 1999). Different from the situational use or the discourse deictic use, the referent of the anaphoric use has existed somewhere in the preceding discourse and the anaphoric demonstrative helps the hearer to track what is happening to the referent. A typical case of the anaphoric use is to introduce a noun which has been mentioned once in the previous discourse; this is a common strategy to introduce a new participant in

languages which lack articles (Himmelmann 1996: 229). This typical use can be found in Tagalog.

(2.32) Tagalog

TAG 003 May kasaysayan sa isang manlalakbay; {0.7 seconds}

May ka-saysay-an sa isa-ng man-RED-lakbay;

EXIST ??-statement LOC one-LK IRR.ACT-RED-travling;

(One incident) is told about a traveler;

TAG 004 ang manlalakbay na ito ay si Pepito.

ang maN-RED-lakabay na ito ay si Pepito.

31

SPEC IRR.ACT-RED-traveling LK PROX PM PN Pepito.

This traveler (his name) was Peopito.

(Himmelmann 1996:229)

In example (2.32), the traveler is a first mention in the initial sentence; as its second occurrence, it is marked by the anaphoric demonstrative ito. The anaphoric

demonstrative and the noun are said to be coreferential. The use of the demonstrative specifies that the main focus is the new participant which has already been mentioned in the preceding discourse. This strategy is widely seen in languages that do not have definite articles.

Similarly, anaphoric demonstratives in languages with articles are preferable in some cases that the participant is mentioned for the second time. This might be somehow averse to the general idea that a noun is marked by the definite article the after its first mention in languages where articles exist. However, a demonstrative is preferred when the second-mentioned referent is a major participant (Christopher 1939; Lichtenbark 1996; Himmelmann 1996). Take English for example. English is known to be the language with articles; in general a noun is marked by the definite article the after it is introduced in the preceding discourse. Nevertheless, a

demonstrative is more usual to indicate a noun at its second mention if the noun is thematically prominent (Christopher 1939; Lichtenbark 1996; Himmelmann 1996).

Observe the examples in (2.33); the sequence in (2.33 a) is more natural than the one in (2.33 b).

(2.33)

a. Once upon a time, there was a king. This king had…..

b. ? Once upon a time, there was a king. The king had…..

(Himmelmann 1996:229)

32

Such sequence is usually seen in narrated storied. In the first sentence, there is a new participant, a king, which is the main character of the story. At its second mention, it is common to be indicated by the anaphoric demonstrative, compared to the one introduced by the definite article, since it is the focus of the story. Likewise, there is always a major participant in communicative events; the major participant is

introduced by the anaphoric demonstrative, specifying it holds a focus position. Thus, the anaphoric demonstratives in languages with articles serve to establish a major discourse participant. Two further examples are provided in (2.34) and (2.35).

(2.34) Toʔabaʔita

Si uʔunu ʔeri ʔe lae suli-a teʔe wane bia kwai-na bia Class story that it:FACT go about-them one man and spouse-his and ʔa-daroʔa teʔe wela, wela wane.

BEN-their.DL one child, child man.

Wela ʔeri kali wela faʔekwa ni bana. ʔe aʔi si tala Child that little child small PART only it NEG NEG be.possible ʔa-na kai lae ʔa-si kula n-e nii daa.

BEN-this ne:NONFACT go to-Class place REL-it:FACT be.located far.

This story is about a man, his wife, and their child, a boy. The child was very little so he wasn’t able to go faraway places.

(Lichtenberk1996 :387-8) (2.35) Toʔabaʔita

… keka soetoʔo ta ai ura wela ʔeri ma impole ʔe-ki

…they ask some woman for children that and person that_PL keka soreʔe: “Kamili’a ʔe aʔi si. ”

they say: we.EXCL it:FACT NEG NEG.

…They asked some woman about the child, and those people said: “We don’t know.”

33

(Lichtenberk1996 :385-7)

The paragraph in (2.34) narrates an event about a man, his wife and their child; all of the three nouns are first mentions in the initial sentence. At this moment, it is yet not sure which participant among the three occupies the focus of the attention. Later in the subsequent clause, wela “the child” is mentioned for the second time and it occupies the subject position; this leads us to infer that the child is the new discourse topic. Therefore, it is marked by the anaphoric demonstrative ʔeri, specifying that the referent is the central focus in the following discourse. In this regard, the anaphoric demonstrative serves as a signal that the new discourse topic is established. This holds true in example (2.35) as well. The noun phrase imoleʔe ʔe-ki “those people” is

coreferential with the noun ta ai “some women”. In the initial sentence, the noun phrase those women does not occupy the topical position either; thus it is marked by the anaphoric demonstrative for its second mention when it turns to be the focus in the second clause. The slight difference between the two is that the anaphoric

demonstrative in example (2.34) is selective whereas the one in example (2.35) is not selective. The anaphoric demonstrative in example (2.35) refers to a referent in the preceding sentence which includes two other participants. As for example (2.35), the referent refers to a group of women and there is no other possible referent.

So far, it is known that anaphoric demonstratives serve to introduce a new participant and establish its focus position no matter the language has articles or not.

In this way, whether a language has articles or not seems to have little impact on the use of anaphoric demonstratives. However, anaphoric demonstratives in languages where articles exist do not track continuing topics (Diessel 1999: 138). In other words, anaphoric demonstratives help a second-mentioned referent to establish its prominent status. Once the referent is in the focus of attention, it is indicated by other tracking

34

devices instead. Table 2.3 summarizes the steps of a referent’s occurrences interacting with tracking devices.

Table 2.3 The use of anaphoric demonstratives after first mention

(Diessel 1999: 98)

In this regard, anaphoric demonstratives in languages with articles more often function as a signal of topic shift (Diessel 1999: 138). This idea is supported by a number of studies investigating anaphoric demonstratives in comparison with other tracking devices such as personal pronouns, definite articles, zero anaphors, and so on (Linde 1979; Ehlich 1982; Givón 1983; Sidner 1983; Ariel 1990; Gundel et al. 1993;

Lichtenberk 1988, 1996; Himmelmann 1996; Comrie 1997). These studies suggest that the continuing topics in general are tracked by third person pronouns, or noun phrases that contain definite articles in languages that has articles. In contrast, non-continuing topics, which is not the current focus and is less unexpected, are tracked by anaphoric demonstratives.

In some languages that has articles such as Dutch, German and Russia, the use of anaphoric demonstratives even eliminates the possibility that the topic in the

preceding discourse is a possible referent (Comrie 1997). Example (2.36) demonstrates the difference between the personal pronoun and the anaphoric demonstrative in German, in which there exist articles.

times of occurrence track devices discourse status the first mention (indefinite) NP new referent

the second mention anaphoric demonstratives established as topic Subsequent mentions 3^rd person pronouns

definite articles, etc.

Continued as topical referent

35

(2.36) German

a. Tracked by the personal pronoun

Der Anwalt sprach mit einem Klienten. Da er nicht viel zeit hatte, The lawyer talked with a client. Since he not much time had, verreinbarten sie ein weiteres Gespräch nächste Woche.

agree on they a further conversation next week.

The lawyer talked to a client. Since he [the lawyer] didn’t have much time, they agreed to have another meeting next week.

b. Tracked by the anaphoric demonstrative

Der Anwalt sprach mit einem Klienten. Da der nicht viel zeit The lawyer talked with a client. Since this not much time hatte, verreinbarten sie ein weiteres Gespräch nächste Woche.

had, agree on they a further conversation next week.

The lawyer talked to a client. Since he [the client] didn’t have much time, they agreed to have another meeting next week.

(Diessel 1999: 96)

In example (2.36a), the third person der is coreferential with the der anwalt ‘the lawyer’. When the lawyer is mentioned for the second time, it is tracked by the personal pronoun since it is considered as a continuing topic. The lawyer in the initial clause stands at the subject position, which usually is assumed to be a focus position and attracts hearers’ attention. In the following sentence, the lawyer continuing to occupy the subject position shows that it is the main character in this event. Therefore, it is the only major participant in this communicative event and tracked by the

personal pronoun. In example (2.36b), the anaphoric demonstrative in contrast refers to the noun phrase Klienten “the client”. The client does not occupy the topic position in the initial clause. Hearing the first clause only, the listener would naturally assume that this communicative event is mainly about the lawyer, which however is not the

36

case in (2.36b). In the second clause, it is the client that stands at the subject position, which means that the main focus is no longer the lawyer but the client. Accordingly, anaphoric demonstratives serve to specify the occurrence of topic shift rather than to track the continuing topic. If the client continues to hold the focus of attention, it is indicated by other tracking devices instead. The phenomenon for anaphoric

demonstratives to indicate topic shit is quite common in languages which have articles.

The sequences of this strategy may be explained by Levinson’s approach to anaphora in terms of conversational implicature(Levinson 1983, 1991); he states that personal pronouns imply unproblematic coreferential relationship while definite full NPs do not imply coreference relation. In this regard, demonstrative expressions are considered to be a compromise since repeating the full noun does not display the coreferential relation on one hand. On the other hand, employing a personal pronoun immediately yields too strong coreferential relation on the other hand, which may not make a sufficient cue for the addressee to relate to the given referent correctly. Take a look at the example (2.37).

(2.37)

VII.10 Something that I noticed about the /movie / particularly unique was that the colors... were {.35}

VII.11. Just {.5}

VII.12. very strange {.6{.2}}

VII.13. Like {.3}}

VII.14. the green was a {2.2}

VII.15 in ordinary bright green, {.55}

VII.16. for the pears, {.4.. and {.25}}

VII.17 These colors just seem a little {.5}

VII.18 Kind of bold, almost to the point of {1.15}

VII.19 Being artificial. {2.25 {.6} tsk {.1} A—nd {.75}}

(Himmelmann 1996 : 227)

37

The use of the proximal demonstrative encodes the coreferential relation to the full noun phrase the color. The occurrence of the demonstrative assists the given referent to establish the topical status of the referent. However, the coreferential relation will be unclear if this position were replaced by personal pronoun; the antecedent could either be the green or the color. Hence, a second full NP mention is necessary for a given referent to become a topic and demonstrative expressions are employed in such cases. In addition, demonstratives are also widely employed to reactive referents which are mentioned in a considerably long distance in some languages like

Toʔabaʔita. Although the use of anaphoric demonstratives has slight variations, they do share a characteristic. That is, they indicate that the antecedent is not the referent which the addressee would expect in the current discourse, i.e. not the most topical noun phrase.

In addition to tracking a referent, demonstratives are employed to track inanimate participants, which holds true cross-linguistically. Observe example (2.38) for

reference to an inanimate referent in Tagalog (Himmelmann 1996: 228).

(2.38)

Ang mgn dinaratnan naman ng bala ay Spec PL RED-come_upon_LOC really GEN bullet PM sinasangga ito ng kanyang kamay, Real-RED-ward_off PROX GEN 3.SG.DAT-LK hand parang sumasangga sa isang pukol lamang.

Like-LK um-RED-ward_off LOC one-LK hit only.

Those who are reached by a bullet try to ward off with their hand, like one who is warding off a mere here.

(Himmelmann 1996:229)

The use of third person pronouns in Tagalog is limited to animate entities, in which

38

case a third person pronoun therefore cannot be the tracking device here. Accordingly, the proximal demonstrative ito is selected to track the inanimate entity, i.e. bala “the bullet”.

In sum, the function of anaphoric demonstratives is to track participants. In languages that possess definite articles or third personal pronouns, demonstratives are used to track nouns that are not in the current focus and function as a signal of topic shift. In contrast, if a language lacks personal pronouns or definite articles,

demonstratives are employed to track major participants instead.