下载掌阅APP，畅读海量书库
立即打开
It is not surprising to think that at the very beginning of the human beings, there is no language, either in the form of speaking or in the form of writing, since language is not needed under any circumstances, if human beings live separately and only take care of themselves. It was not until later in the human evolution that language arose. There were several hypotheses on when, why and how human beings started to have and use language, but in any case, they were urged to use language to communicate. Taking all circumstances into consideration, we can conclude that they all fall in the following three categories: the hunting theory, the hunted theory, and the grooming theory.
For example, in the hunting theory, human beings needed to hunt for food to survive, but there were animals that were too large for a man to trap or kill by himself, lacking the convenience of science and technology. Even a modern human is not capable of conquering a beast of prey on his own. Therefore, human beings needed to collaborate and work together to tame the prey, carry it back to the cave where they lived, and share it along with other members of the group.Therefore, it is understandable that human beings were in the need of sharing information, figuring out the steps and giving orders: they needed to know where the preys were, how to get them and what to do if there were an emergency or incident. At this time, pure changing of the length of the sounds or using gestures was not enough to provide precise information on things that were abstract or not present, such as the size of the prey and the plan on how to trap it, not to mention potential ideas or plans of action. Then, it was time to have language and use it.
Another theory on the origin of language is the hunted theory. According to this theory, human beings are considered the prey of beasts, either tigers, lions or any other carnivores. In order to survive and reproduce, they had to find a safe place to live in and be able to fight against the carnivores attacking them. Again, a single or a couple of ancient human beings was not able to accomplish the task. Then, a crowd was to be formed to look for safe and fruitful places and to be organized in order to fight in dangerous situations and defend themselves. In order to show the route to a newly found place, one needed to tell other members in the group how far it was, in which direction they should go and how long it might take to get there. These pieces of information are hard to show only with facial movements or gestures, because they are very complicated. It was even harder to cooperate in a defense against a beast. Pointing to a beast may mean “throw stones to it” or “light up with a torch” or “there is a beast over there, watch out.” But there was limited chance to figure out the appropriate meaning right at the time seeing the finger that was pointing to the beast. If the person receiving orders needed plenty of time to be sure what the person making gestures was talking about, there was great chance that he might have already been devoured by the beast. This is very inefficient communication.
A third theory is called the grooming theory. The theory is about the needs of social-networking. Being an animal or human being, one cannot live disconnected from the outside world. In ancient times, human beings did not have the technology to be able to fight against the beast alone. Either in size, speed or strength, ancient human beings were disadvantaged compared to larger-sized carnivores. There is no doubt that human beings would have died away if they did not unite and defend themselves together. When gathering together, they had to be aware whether the other people joining their society would be beneficial to the society, or harmful to it. Therefore, communication was an indispensable part of the admission process. In addition to the needs of survival, according to Maslow's hierarchy of needs theory (see Maslow, 1943), human beings have social needs, such as love, affection, care, belongingness and friendship.It is hard to build up any kind of relationships without communicating with each other and being familiar with each other's preferences and dislikes. Admittedly, human beings were able to dance, make gestures and some noises to express themselves, but how much information were they able to share with others and to which extent were they able to make themselves understood? Take bees for example, they have all kinds of dances to show where the flowers are and how far away they are and how difficult it is to get there. They can dance in a circle to show the direction of the flower, and they can also dance slowly to indicate how long it takes to get to the flowers. But they are not able to let other bees know about their feelings: are they in love? Or are they experiencing heartbreaking moments? One may argue that bees do not have the needs to be emotionally attached to each other, but it is not deniable that human beings need more than the dance or intervals of the dance steps to show their most “intrinsic thoughts” (Heim, 2011), the abstract feelings that cannot be measured nor displayed merely by time, number or space. When human beings were forced to let others know how they felt about each other and tried to form cliques among them, they had to use language, which is capable of making clear things beyond time and space. This is known as displacement, a design feature of language (Hockett, 1960).
For any reason why human beings began to use language to communicate, they had specific meanings that were to be conveyed by either words, phrases or sentences. It is understandable that at the very beginning of language, there were only sounds but no written languages, and human beings were not able to tell word boundaries, and of course, morphemes. The sound waves (Diver, 1995) the human beings, a.k.a. the speakers, produced were the only things other human beings, a.k.a. the hearers, could receive to collect the information they were supposed to get. It is logical to make the statement that meanings are associated with sounds produced by the speakers. To be able to correctly get the information signaled by the sounds, the hearers have to have the sounds as assets or inventory, i.e. they have to be exposed to the sounds before puberty and be able to decode the sounds respectively. Among the thirteen design features (Hockett, 1960), language has the characteristics of “creativity,” which means human beings are able to make unlimited meanings from limited sounds and comprehend sound combinations to which they have never been exposed before.
It is experimented by the scientists that any new born is able to pick up any language if exposed to it after birth and kept in the language environment until puberty. There is no wonder that a Chinese baby can speak English perfectly, if the baby is adopted right after birth and transported to the US to stay with his/her American stepparents thereafter. Plenty of facts have proved that human beings have difficulties in acquiring language if they are kept away from the human society until puberty, for example, a child who has lived isolated from human contact from a very young age where he has little or no experience of human care, behavior, or, most importantly in this case, human language. Like a child that was brought up by the wolves can most likely communicate with wolves, but he is unable to understand human beings or make himself understood by them. It is not surprising, because human sounds are new and foreign to the wolf kid and are therefore not in his inventory, and thus do not make any sense to him.
If human beings use language to communicate with each other, can anyone talk to anyone? Many people may have the experience of successfully communicating with foreigners without knowing each other's language, such as my mom who came to the US in 2009 for my master's graduation ceremony without any English education but managed to switch seats with an American gentleman in the plane without any help.She was very proud of her ability of merely using Chinese to communicate with him by pointing at their seats respectively. It is true that people sometimes do not have to know each other's language in a conversation. But once again, how well are they able to share information with each other? Or, to which extent can they express themselves? We have to admit that in this case, the conversation cannot get further than that of animal talks: they are not able to exchange information that contains abstract thoughts, displacement or history and so forth. The reason is obvious: different languages may have different sounds. Even if two languages have similar or the same sound systems, meanings associated with the sounds are not the same. When communicating, the hearer is not able to decode the sounds correctly. Similarly, a Chinese native speaker is unable to talk to an English native speaker, if none of each has ever known the other's language. It is easy to think that the sound “tree” does not mean tree to the Chinese, because the sound “tree” is not associated with the meaning, which is: a plant with a brown chuck and green leaves. Similarly, an American cannot understand the Chinese word shu ‘tree’ without any knowledge of the Chinese language.
We should not be disappointed with the fact that there are hundreds of languages spoken in the world. Like what is conveyed by the Babel story (Genesis, XI, pp.1—9): the God creates various languages for the salvation of human beings, not for the chastisement, because human beings are saved from the tyranny of a single “totalitarian” language and ideology. This at least points out two clear facts: first, there are various languages and they are different from each other; second, one should approach to the ideology the language represents when studying a language. Ideology can be simplified to the form of culture, to some extent. An example is that there is a saying in the German language that he is not yet as tall as three pieces of cheese ‘er ist noch nicht drei Käse hoch,’ but in Chinese, since there was no cheese until it was imported from the West, the saying is he is not yet as tall as three pieces of tofu ‘ta hai mei san kuai'r doufu gao.’ Meanwhile, this example adds more evidence to Wholf's hypothesis of linguistic relativity, in which he states that a language is defined/shaped by its culture and gives the example of the Eskimo language, which has more than 70 ways to name snow flakes.
Though Whorf's hypothesis of linguistic relativity (and its strong version: linguistic determinism) is not supported by a great number of linguists, there is still some evidence that a language may be influenced by the culture it is spoken in. For example, there are three genders (here: grammatical gender) in the German noun system: male, female and neuter. When a German says that he talked to his neighbor yesterday night, he unconsciously releases the information of the neighbor's gender to the hearer, c.f. Ich habe gestern abend mit meinem Nachbarn gesprochen ‘I talked to my male neighbor yesterday night’ and Ich habe gestern abend mit meiner Nachbarin gesprochen ‘I talked to my female neighbor yesterday night.’ But it will not be the case when it is translated in the Chinese or English language, c.f. in Chinese wo zuotian wanshang gen wode linju shuohua le ‘I talked to my neighbor yesterday night.’ And in either translation, the gender of the neighbor is not mentioned by the speaker nor marked by the language. It does not mean that Germans pay more attention to the gender of human beings, because they assign genders to existing or none-existing subjects too: a table is male, a blackboard is female, a book is neuter, and an idea is female, too. They are just more used to having either natural gender or assigned gender for everything in the world. They do not have to take more time to think about what is male and who is female—there is no evidence showing that Germans speak slower than any other non-gendered language speakers in general. Nor is there evidence that Germans learn the gender of each subject just as they learn how to spell and read (see Pinker, 1994). All the facts prove that the gender is genetically programmed in the German language. There is no need to ask why there is gender in the language or when the gender dates back to. Similar to language arbitrariness, where no one would ask why the sound of “tree” means the plant that has brown trunk and green leaves, the gender system is in the German language at the very beginning of it. The genders of some nouns have changed over time or differ themselves from region to region (such as Butter ‘butter’ can be female or male, or even neuter in some southern areas of Germany), but it is not a problem for native speakers: one doesn't need to know how to speak a language in the Middle Ages, and native speakers are perfect at the current sound and meaning system.
Native speakers, defined as people with a specific mother tongue, are able to perfectly use their native language to express their most intrinsic feelings and abstract thoughts. The native language, or mother tongue, is programmed in the genes of its native speakers (see Chomsky, 1969). That means, native speakers may unconsciously reveal their innermost thoughts by their language choices: either by word order, word choice or grammar choice and so forth. For example, in German, one may either use the active voice or the passive voice to signal the meaning “there are people working here.” If the passive voice is used, Hier wird gearbeitet , it signals the meaning that the speaker is not one among the people working here. To the contrary, if the active voice is utilized, Viele arbeiten hier , then, it does not exclude the speaker from the people working here.
A language learner is not expected to use the target language as perfectly as its native speakers do for two reasons. First, because the sound-meaning association is not programmed in language learners' genes, it is impossible for them to study all the sound-meaning associations of the target language: there are too many. Not to mention that lexicon is an open system, i.e. there are new words joining the lexicon with time, and there are old words that become obsolete over time. Native speakers may have encountered situations, in which they have misunderstanding with non-native speakers due to their imperfection of the language. Also, because of the insufficient skill in formulating thought in the non-native language, non-native speakers may have fuzzy formulations, which may be understood as fuzzy thoughts (Heim, 2011). In either conversations or literatures of non-native speakers, there may be miscommunication or misunderstanding arising from the imperfect expressions or descriptions. Therefore, I will limit my study to conversations and literatures by native speakers.
According to Stevens (unpublished manuscript; see Hockett, 1960), language is a system of signs designed for the purpose of communication. In a communication, there need to be a speaker, a hearer, and the information to be shared. In order to successfully deliver the information, the speaker has communicative goals, i.e. what the speaker needs to deliver to the hearer, and communicative strategies, i.e. how the information should be delivered to the hearer (see Kirsner, 2014). As soon as the speaker knows what to deliver, he/she, in the mental process, will encode thoughts and information into language, which includes structure and word, or grammar and lexicon. The speaker makes specific choices of structures and words in order to make the hearer comprehend the utterance and recognize what is supposed to be received in the conversation. Here the speaker needs to consider how to make the utterance comprehensible especially to the hearer. In order to achieve the goal, the speaker needs to take the hearer's background knowledge into consideration and then leads his/her mental process to the joint attention. In Chafe's famous pear story (Chafe, 1979, p.160), he mentions that if a story is watched by people from different cultures, their centers of focus are different. A person from a culture, where there is no pear planted or imported, pays more attention to the pear and uses more effort in how to verbalize the fruit, which he/she has never seen before. But, to the contrast, a Californian does not have any problems identifying pear in the video clip and thus pays less attention to the pear than the person, who does not know pear.
In the process of communication, the speaker evaluates the feedback from the hearer in the interaction, and adds more hints or gestures if needed. In Levy (1979, p.187), a girl is asked what classes she is taking and how her schedule looks like. Levy postulates four steps in which the girl mentally processes as the speaker in the communication with feedback from the hearer. First, she mentally lists the points that need to be included in her utterance, i.e. the information she delivers. Second, the girl thinks about how to describe the course she takes and give further specification. Third, the girl realizes that there may be a potential ambiguity in her utterance and tries to reduce it by establishing a subsidiary goal and satisfying it. Fourth, the girl provides further elaboration to minimize the potential ambiguity at best. In this experiment, the girl clearly has the communicative goal in her mind, which is to make the hearer familiar with her academic schedule, and changes communicative strategy according to her mastery of the information and the feedback from the hearer.
In Li and Thompson (1979, pp.311—312), they admit that structure and word choice is vital, but there is more beyond the words:
Since the inception of linguistics as an empirical science, a justifiably primary concern of grammarians has been the discovery of structural regularities in language. There is no doubt that statements of such regularities are vital to our understanding of the nature of language. It is equally true, of course, that not all aspects of sentence formation can be described by rules stated in terms of grammatical or even semantic properties. Rather, there are a number of facts about sentences that can only be understood in terms of speakers' and hearers' abilities to make inferences beyond what sentences actually say. Further more, certain rules that are pragmatically based are conditioned by the perception of the speaker at the time of the utterance. The speaker's perception of the world and his interpretation of the pragmatic factors may change from instance to instance, making such rules difficult to formulate.
There is no doubt that the speaker has to take the hearer's background into consideration. Howling at the moon may still work after a long time to a certain degree, but it is not an efficient and successful communication, because first, the communicative goal is not completely achieved, and second, taking the same time and effort, the speaker should have a huge chance to provide the hearer with all the details the speaker wants the hearer to be familiar with and make the hearer completely comprehend the utterance. For example, if a country girl, say Mary, from Alabama talks to a local Los Angeles city girl, say Connie, and supposes that Connie understands that people in Alabama air dry laundry on a rack in the open air, and uses this fact as an analogy, Connie would have difficulty in understanding what Mary actually means, since there is barely anyone air drying laundry in Los Angeles: people would use a dryer to dry it, because the rent is high and no one has a yard to air dry laundry.
Similar to what has been discussed before, in Wholf's theory of linguistic determinism, a culture may have influence on how people perceive the world and how people use language to express themselves, in the case that it is possible that the German unconscientiously notice the gender of people in sight, which is most probably not in the focus of Chinese people. Besides, culture also has influence on to which part of the speech one pays more attention to. A traditional Chinese is educated not to exaggerate individual achievements and thus is interested in the praise of individual performance, which is most likely in Western people's main focus. Therefore, the speaker should take the cultural background into consideration and changes communicative strategy according to the hearer.
In addition to culture, Hopper (1979, pp.214—215) discovers that it is the sequentiality that tells foreground sentences (the “main line” events) from the background sentences (the “shunted” events):
The foregrounded events succeed one another in the narrative in the same order as their succession in the real world; it is in other words an iconic order. The backgrounded events, on the other hand, are not in sequence to the foregrounded events, but are concurrent with them. Because of this feature of simultaneity, backgrounded events usually amplify or comment on the events of the main narrative...Strictly speaking, only foregrounded clauses are actually NARRATED. Backgrounded clauses do not themselves narrate, but instead they support, amplify, or COMMENT ON the narration.
If a hearer is unable to differentiate important information (here: foregrounded events), from peripheral information (here: backgrounded events), he/she may be confused as to which information is worth giving more effort to go into details and which information can be left out. If a student misses the part in a book, which he needs to study for the final, but studies the parts that are not quite related to the test, he/she may fail the course. In a worse and more serious case, a military officer's failure to attack the right city may be detrimental to a battle. In all, if the hearer is not able to get the most important part of the conversation, the communication is bearing a failure. Ochs (1979, p.51) quotes earlier literatures that an infant at his “one word stage” in the language development at first “deletes certain highly predictable information in the utterance, which will be expressed by itself at a later stage” (Bates, 1976; Greenfield and Smith, 1976). This too adds more evidence that human beings unconscientiously tend to put unpredictable information to the focus of the utterance.
In addition to structural signs that show different levels of importance in meaning, word choice may also result in change in meaning. Taking “it” and “that” for example. Both can be used to refer to something or someone mentioned before. Say, you meet your neighbor in front of your house. And your neighbor tells you that his son is admitted to the best school in the States. You say “I know it,” for example, and mean “let's talk about it; I have more about it to share with you.”
In opposition, you say “I know that,” and you mean “that's the end of this topic, I do not want to talk about it.” This is confirmed by Linde (1979, pp.347—349). In her study, she finds out that the speaker chooses “that” to move from the event, to which “that” makes reference to, to the next event. Besides, she (1979, pp.344—348) also points out that “that” is used to accomplish the task to refer to a preceding statement taken as a statement, while “it” is used to refer to a statement that is discussed and emphatically affirmed. In addition, “that” also involves contrasting with a previous evaluation and typically refers to an item that is “not within the node of focus.” This, however, contrasts with the characteristics of the German definite articles
der
,
die
and
das
(and their inflected forms) and personal pronouns, which will be further discussed in Chapter 3.
Diessel (1999) postulates that there are three features demonstratives should possess. First, demonstratives are deictic and have specific syntactic functions. Second, they serve pragmatic functions. And third, they have semantic features. This study will be limited to the semantic aspect.
Previous studies also show that the meaning of demonstratives comprises or contains two components or features: deictic and qualitative (Lyons, 1977; Fillmore, 1982). Deictically, demonstratives can indicate a stationary referent as well as a moving referent. Qualitatively, they may indicate speaker's attitude towards the referent or the quantity of the referent (i.e. whether the referent is a single entity or a set of identical entities) (Diessel, 1999).
Most reference grammars focus only on the deictic feature of demonstratives. For these traditional grammars ( Xiandai Hanyu Cidian , 2004; Duden, 2002), deixis means to point to the referent without marking the degree of attention the speaker urges the hearer to put onto the referent. In contrast, in the Columbia School linguistic theory, there is a hierarchy of the differing levels of attention given when different demonstratives are used (Garcia, 1975). Within the school's linguistic theory, the demonstratives are recognized as instructive and urge the hearer to pay attention to the referent (Kirsner, 1993). If the demonstrative means HIGH DEIXIS, it will make the hearer pay more attention to the referent. And if the demonstrative has the meaning of LOW DEIXIS, it will tell the speaker to pay less attention to the referent. Finally, in Cognitive Grammar, demonstratives “represent a conventionalized, grammaticalized means of connecting nominal referents to the subjects of conception” and, therefore, function to provide a platform which is shared by the speaker and the hearer (Langacker, 1997). Similar to the cognitive view, Diessel (1999, 2006) argues that demonstratives do not necessarily show relative location, but rather direct the speaker's and hearer's joint attention.
The term which will be the center of the study is deixis , which is used in different ways by different people. It is a device through which every language incorporates contextual information (Weissenborn & Klein, 1982). Since deixis is the place where language and reality meet, deictic meaning shifts according to the contextual information, such as where, when, by whom, etc. Denny (1978) calls it a “relativity of lexical semantics” that refer to a “particular man-environment relations found for particular human groups” (p.72). Some other factors such as relative distance and visibility can affect the meaning as well.
There are many studies on the German demonstratives in light of their deictic force. Ehrlich (1982) examines German demonstrative adverbs, and states that there is an opposition between hier ‘here’ and da ‘there’ as well as hier and dort ‘there,’ clearly differentiating the speaker's place, reference place, and denotation place. Hartmann (1982) has written about the differences between Rhineland German and standard German. He shows that there are two definite articles in the Rhineland German with two functionally distinct paradigms; one is deictic, and the other is non-deictic. He points out further that the non-deictic mode indicates that the noun-phrase is common or known information to both the speaker and the hearer. Pasierbsky (1982) studies the development of person deixis in Chinese and posits that the original personal pronoun system has been taken over by a more socially oriented system with an account of speech participants.
As noted above, the term deixis, when used in Columbia School analyses to describe the dedicated, invariant meaning indicated by a specific signal (morphology, word order, etc.), has had a different, more technical interpretation than is given in general linguistic literature. Rather than being a description of where the referent is with respect to speech act participants, it is an instruction to the hearer to seek out and attend to the referent, and may exist in different degrees. There are analogs to this in other areas of grammar. Reid (1977) examines two tenses in French and finds out that passê simple means HIGH FOCUS, whereas imparfait means LOW FOCUS. Here the meaning FOCUS is an earlier term for DEIXIS in the Columbia School. It should be noted, however, that there has been a recent tendency among some Columbia School linguists to dispense with DEIXIS as a meaning or component of a meaning. Working on Serbo-Croatian and Spanish pronouns, Gorup (2002) and Garcia (2009) have analyzed the deictic effect attributed to these elements as a pragmatic consequence of the amount of information (nature and number of different dedicated meanings which they do signal).
Several analyses of demonstratives in various languages have been made within the Columbia School framework covering such languages as Afrikaans, Dutch, and Swahili, but little has been done on the German and Chinese demonstratives, especially beyond the syntactic level. Kirsner (1979, 1993) was the first one who studied Dutch demonstratives deze ‘this’ and die ‘that’ in the Columbia School framework. His hypothesis is that by using deze (with the allomorph dit ), the speaker necessarily urges the hearer to exert more attention to the referent, and the opposite by using die (with its allomorph dat ). Thus, the employment of demonstratives is rather a consideration of interaction.
Following Garcia (1975), Kirsner (1979) defined deixis as “the force with which the hearer is instructed to find the referent” (p.359). He further postulates three aspects of deixis: first, “...deixis, urging the hearer to find the referent, communicate[s] that the referent's identity cannot yet be taken for granted”; second, “...[deixis] provides a ‘tighter link’ with its first mention... insisting that the hearer search for the world, alerts him to it more, induces him to pay more attention”; and third, “the command to search for the referent favors the inference that effort is required to find it and, hence, that a specific referent exists—that is, that it is a localizable entity rather than a disembodied general concept” (Kirsner, 1979, pp.358—359). He anticipates that deze ‘this’ “will be used when the hearer's task is more difficult (i.e., when it is harder to select the referent in question)” and “will suggest more forcefully than die that a specific referent exists” (Kirsner, 1979, p.369).
Moreover, Kirsner (1979) gives three strategies that are consistent with the definition of deixis: first, “NOTEWORTHINESS. The speaker will direct attention strongest to entities that he, the speaker, is most interested in talking about”; second, “GIVENNESS. The speaker will direct the hearer's attention strongest to entities that are not given, ‘in the hearer's consciousness’ (Chafe, 1976)”; and third, “FOREGROUNDING. The speaker will use more than one means of drawing the appropriate amount of attention to the noun's referent, so that strong urging of the hearer to find it will be coupled with devices for foregrounding the noun in question and weak urging will be coupled with devices for backgrounding” (p.360).
Now Kirsner (1993), comparing a Cognitive Grammar and Columbia School analysis of the Dutch demonstratives, shows that it is possible to derive deictic-like effects from purely locative meanings, such as NEAR and NON-NEAR. For example, by claiming that a referent is near the speaker with deze , one can suggest to the hearer that the object is more vivid, seen in more detail, and is more zeroed in upon and perceived more precisely in the way that near objects are when compared with far objects, which in turn entices the hearer to pay more attention to deze 's referent. [1] Thus, his original (1979) Columbia School analysis could be replaced by one less radical, not postulating purely instructional meanings such as HIGH DEIXIS and LOW DEIXIS, making it actually unnecessary to postulate meanings such as HIGH DEIXIS and LOW DEIXIS. However, Kirsner (2011) argues that Afrikaans differs from Dutch in having an unmarked demonstrative dié contrasting with proximate hierdie and distal daardie and the definite article die . Since one cannot derive the deictic effect of unmarked dié from a locative meaning (since it specifies neither near nor far), it forces the analyst to postulate DEIXIS as a meaning in the Afrikaans demonstrative system. We should also note that Diessel (1999, 2006) includes the German demonstrative dies ‘this’ in his examples to support his argument that demonstratives do not always show relative location.
In addition to Diessel's analysis on dies , Hopkins and Jones (1972) review German grammar books and summarize that jener ‘that’ is mostly omitted for the sake of reducing learners' confusion, because jener is rarely used. They summarize conditions where jener, instead of article plus locative, is used. For example, the referent is mentioned before or jener is followed by relative clauses or adjectives. They also point out that jener , which is commonly thought to be the translation for ‘that,’ is not a demonstrative or deictic pronoun.
In the case of Chinese demonstratives, as stated in the traditional grammar book Xiandai Hanyu Babai Ci (Lü et al.; 1981), zhe refers to people or things that are close to the speaker, whereas na refers to people or things that are close to the hearer. Tao (1999) comments on this idea, saying that “this may be true for isolated sentences, but is not always true with discourse data” (p.72).
Since Teng's (1981) idea of freeing the referent from a spatial distant view and expanding the data to speech act, the referred distance has been expanded to a contextual or psychological level (Lü, 1985; Tao, 1999; Fang, 2002; Biq, 2007; Yang, 2007; Xu, 2008), where demonstratives are not limited to spatial distance, but rather to metaphorical distance, such as the speaker's positive or negative attitude towards the referent. Biq (2007) argues that demonstratives are going through stages of lexicalization in spoken Mandarin in Taiwan. She specifically mentions the distal demonstrative na , along with its associated phrase nazhong ‘that kind,’ which has the meaning of “approximation, vague identification and speaker's uncertainty” (p.130), while Fang (2002) describes that the proximal demonstrative zhe is in the process of lexicalization. Xu (2008) considers the distal demonstrative na as a discourse marker in spoken Chinese. He scrutinized 8.22 hours of spoken Chinese data and proposes that na enables the inter-subjectification of both participants of the discourse and thereby induces discursive and social interaction. Tao (1999) puts forth more factors which will influence the use of demonstratives. Factors such as the shift of discourse structures, discourse properties of focused referent, the building of the text, the speaker's assumption of the hearer, and the speaker's attitude towards the referent.
Other usages of the Chinese demonstratives are studied in Hayashi and Yoon's cross-linguistic research on the use of demonstratives in discourse (2006). They see demonstratives as “filler words” when the speaker fails to find the appropriate word immediately in an on-going discourse (pp.485—486). In their paper, they mention three types of functions: the placeholder use, the avoidance use, and the interjective hesitator use. This discourse phenomenon exists mostly in East Asian languages such as Chinese, Japanese, and Korean, but it can also be found in Finnish. The existence shows that the use of demonstratives as “filler words” is extremely culture-related and thus, to some extent, is limited to being applied within East Asian languages in general.
[1] The following examples and explanations belong to Robert Kirsner: Compare the English expression Look closely , in which claiming the relevance to the communication of nearness tells the hearer to concentrate his or her attention on the intended referent . Furthermore, there is no converse term. (While two members of a large family can be either closely related or distantly related, there is no command *Look distantly. ) That communicating precision of location, in turn, suggests nearness and hence that the two concepts are related is shown by the sentences The pole is near you, The pole is far from you, The pole is right near you versus *The pole is right far from you.