安 徽 大 学

硕 士 学 位 论 文

题          目 Cognitive Approach to Deixis

专          业   英 语 语 言 文 学     

研  究  方  向   英 语 语 言 学        

姓          名   孙 中 勤  届别 二00三

导师姓名、职称   朱 跃     教授        

二00三年五月八日

 

Cognitive Approach

to Deixis

MA Candidate: Sun Zhongqin

Supervisor:    Zhu Yue    

 

School of Foreign Studies

Anhui University         

 

May 8, 2003

 

 

Contents

Acknowledgements……………………………………………………...2

Abstract…...……………………………………………………………..3

Chapter 1    Introduction………………………………………….…6

Chapter 2    Traditional Accounts of Deixis………………………...8

Chapter 3    Limitations of Traditional Views on Deixis……….…19

Chapter 4    Cognitive Approach to Deixis………………………...24

4.1   The ICM of Deixis……………………………………..25

4.2   Mental Space Theory and Deixis……………………..29

4.3   Deictic Prototype Effect……………………………….37

4.4   Physical Space Lays the Foundation for deixis……...43

4.4.1    Spatial Conceptualization of Social Deixis……...…..46

4.4.2    Spatial Conceptualization of Time Deixis…………..48

        4.4.3  Spatial Conceptualization of Discourse Deixis ……50

Chapter 5    Conclusion……………………………………………..53

Bibliography……………………………………………………………54

Acknowledgments

  I should like to take this opportunity to express my heartfelt gratitude to all those who have given me much help in my pursuit of the master’s degree and in my thesis writing.

  First of all, I would like to express my special thanks to my supervisor Zhu Yue, who has offered me his insightful comments and enlightening suggestions in the whole course of my thesis writing. Without his patient instructions, I could hardly have completed my thesis.

  Likewise, I express my sincerest thanks to Professor Zhou Fangzhu. I was initiated to the subject of deixis under the influence of his lecture on deixis and translation. My deepest gratitude also extends to Professor Hong Zengliu, Professor Chen Zhenfa, Professor He Gongjie, Assistant Professor Xiao Shuhui, whose lectures have broadened my scholastic vision.

  Last but not least, grateful acknowledgment is made to my wife Cui Guizhen. Her encouragement and support have relieved me of a lot of anxieties and tensions that have accompanied the writing of this paper.

Abstract

Deixis is examined as a pragmatic phenomenon encoded by particular linguistic expressions which establish a relationship between the speaker and such various contextual parameters of the speech event  as the addressee (in particular, his social position with respect to the speaker), time, place, and discourse point in the speech event.

Traditionally, deixis is described according to its function and the contextual parameter it defines. The paper first gives a brief description of the traditional views on deixis, then it points out that there are some issues needed to be further clarified. Firstly, traditional accounts of deixis distinguish between deictic and non-deictic terms and usages of these terms on the basis of rigid criteria for membership in the deictic category, which obscures our understanding of deixis itself. Secondly, an essential characteristic of all categories of deixis appears to be their egocentricity and the speaker generally constitutes the deictic center of the speech event, but there are various exceptions to this, in which deictic expressions are used in ways that shift this deictic center to other participants, or indeed to protagonists in narratives. Traditional accounts of deixis cannot provide a convincing explanation for this phenomenon of deictic projection. Thirdly, some particular expressions are used to express place, time, and discourse deixis in different contexts. There must be some conceptual and social relatedness at least among the three basic deictic categories, and it remains unexplored.

Cognitive linguistics provides a new theoretical framework for the study of deixis. The paper follows this new approach to analyze deixis and handle the above problems. It claims that deixis is a pragmatic category which can be understood in terms of an idealized cognitive model (ICM); mental space theory may offer some reasonable explanations for the phenomenon of deictic projection; prototype structure can help us understand the deictic expressions and their usages and it clearly shows that deixis is a graded category. Moreover, the paper attempts an account of the common origin of different deictic categories. It argues that social, time and discourse deixis are metaphorically derived from place deixis. This analysis is consistent with and confirms the spatialization-of-form hypothesis, which requires a metaphorical mapping from physical space onto a conceptual space. It claims that social deixis is based on the metaphorical understanding of social space as physical space, time deixis is based on the metaphorical understanding of time as space, and discourse deixis is based on the metaphorical understanding of discourse as time and time as space.

Key Words:  deixis     mental space     image schema

    deictic prototype effect     idealized cognitive model

    metaphorical mapping      spatial conceptualization

摘 要

指示语作为一种语用现象，直接涉及语言结构和语境之间的关系。说话人与言语事件中各种语境参数（如听话人，以及时间、地点、话语）之间的联系就是通过指示语来表达的。对于传统研究而言，指示正是通过它在言语事件中充当的功能及它所表示的语境参数来定义的。

本文首先就传统的指示语研究作了扼要的介绍，然后指出在传统的研究中存在一些需要进一步澄清的问题。第一，传统的指示语研究严格区分指示词语和非指示词语以及它们的用法，这种刻板的分类反而在某种程度上影响了我们对于指示语的全面理解。第二，以自我为中心是所有指示类型的一个基本特征。在言语事件中，通常以说话者为中心角色。然而在实际运用中，指示语有时也会发生自我中心转移和语境变换。对这一指示投射现象, 传统的研究没有给予令人信服的解释。第三，一些指示代词和指示副词在不同的语境下可分别充当地点、时间以及篇章指示语。这一现象表明至少在这三种基本指示类型中，隐含一些共有的认知联系。

   本文综合了认知语言学的不同理论对以上现象加以分析。文章认为指示语作为一种语用范畴可以结合莱考夫的理想认知模型进行理解，心理空间理论能够对指示语的投射现象作出合理的解释，类典型理论会更加清楚地显现指示范畴中指示语的各种用法。在此基础上，文章进而探讨了基本指示类型的共同起源。文章认为社交，时间，篇章指示语分别是物理空间向社交空间，时间和篇章结构进行的隐喻投射。作为概念隐喻认知的结果，表示空间关系的地点指示语常被用来表达社交，时间，篇章指示语，这一分析也有力地验证了莱考夫形式空间化假设。

关键词： 指示语     心理空间    隐喻映射     意象图式  

指示原型效应    理想认知模式    空间概念化

Chapter 1    Introduction

Deixis in its traditional linguistic sense refers to the fact that certain linguistics forms have direct pragmatic interpretation depending on parameters of the speech situation, rather than a stable semantic value. In particular, their interpretation is contextually anchored to the identity of the speaker and addressee, their locations, and the time of the utterance. When A asks B on the phone, “Will you come here?” the linguistic expressions you, here, and will are interpreted as “addressee”, “location of the speaker”, and “after time of utterance”, respectively.

Fillmore(1971, 1975), Lyons(1968, 1977), Jarvella & Klein(1982), Levinson(1983), He Zi-ran(1988, 1997)，He Zhao-xiong(2000)， Chen Zhi-an, Peng Wei-xuan(1994), Zhang Quan(1990，1994) and other scholars have made a considerably detailed study on deixis, but no single and comprehensive theory of deixis has ever been proposed, and linguists have generally tended to describe deixis according to their function and contextual parameters they define. Consequently, there are always some problems needed to be further clarified. Firstly, traditional accounts of deixis distinguish between deictic and non-deictic terms and usages of these terms on the basis of rigid criteria for membership in the deictic category, which obscure our understanding of deixis itself. Secondly, an essential characteristic of all categories of deixis appears to be their egocentricity and the speaker generally constitutes the deictic center of the speech event, but there are various exceptions to this, in which deictic expressions are used in ways that shift this deictic center to other participants, or indeed to protagonists in narratives. Traditional accounts of deixis cannot provide a convincing explanation for this phenomenon of deictic projection. Thirdly, some particular expressions are used to express place, time, and discourse deixis in different contexts. There must be some conceptual and social relatedness at least among the three basic deictic categories, and it remains unexplored.

With the development of cognitive linguistics in recent years, it seems possible to propose an experiential framework for the analysis of deixis.  In contrast to other philosophical views that have affected linguistics, experiential realism views language as part of general cognition. This view is adopted in cognitive linguistics, which aims to explain how language is systematically grounded in human cognition. One of the basic principles of experiential realism and cognitive linguistics is that language is not a representation of objectively existing reality, but of reality as it is perceived and experienced by human beings.

My paper aims at proposing a cognitive structure for deixis, in terms of which deixis is understood and used to structure reality, so that reality is internally experienced and hence reproduced or changed. The approach to deixis adopted in the paper originates in cognitive semantics and comprises cognitive models, image schematic and metaphorical structures as in Lakoff (1987) and Lakoff and Johson (1999), mental spaces as in Fauconnier(1985, 1997), and prototype structure as in Rosch (1978) and Taylor (1995).

Chapter 2    Traditional Approaches to Deixis

Deixis is considered the most obvious and direct linguistic reflection of the relationship between language and context (see Levinson 1983:54). It relates the use of language with respect to a particular time, place, and person.

Karl Buhler, the Austrian psychologist and semiotician, was the first person to employ the term in its modern sense. Buhler (1982) identified three deictic components of a deictic field of language―temporal, spatial and person. At the center, or origo, are here, now, and I. Deixis is not limited to a few selected words such as I, you, and this. It is also, as defined by Peirce (Buchler, 1955:56) in his description of the same phenomenon, the function that connects all languages uses to situations.

 [A] symbol, in itself, is a mere dream; it does not show what it is talking about. It needs to be connected with its object. For that purpose, an index is indispensable. No other kind of sign will answer the purpose. That a word cannot in strictness of speech be an index is evident from this, that a word is general—it occurs often, and every time it occurs, it is the same word, and if it has any meaning as a word, it has the same meaning every time it occurs; while an index is essentially an affair of here and now, its office being to bring the thought to a particular experience, or series of experiences connected by dynamical relations.

  Deixis has long been associated with the gestural dimension of language. Peirce’s definition of the index as anything that focuses the attention or anything that startles us applies both to the spatial gesture of pointing and to the phonic gesture of shouting (Buchler, 1955). Revzin (1974:18), in an essay on the possible origins of language, calls deixis “the primordial function of gesture” and finds in deictic signals “ the concepts conveyed in modern language by words of type ‘here’, ‘there’, ‘near’, ‘distant’, and by different grammatical categories”. Calling attention to a situated particular is the most fundamental function of deixis. This function, which can be performed gesturally by pointing and shouting, can also be performed verbally by a “single deictic particle [whose] function is to draw attention to some feature of the situation, …normally accompanied by some paralinguistic movement of the head or hands…. We may think of the deictic as meaning ‘look!’ or ‘there!’”(Lyons, 1975:645).

The deictic function is not only a phylogenetic stage in the development of language, but the continuing prerequisite for all references. All language uses depend on some felt relevance to situation, on the attention of participants, and on their ability to lift out the topic. Russell thinks that all indexical expressions (or as he prefers, egocentric particulars) can be reduced to a single primary one by translating them into expressions containing this. The pronoun I will thus be rendered “the person who is experiencing this”(see Levinson, 1983:57). Gale (1967) points out that the efforts to produce a language devoid of deixis are in vain, because without it, “we could not know, for example, that some event is now happening, simultaneous with this token [instance of speech]”(1967:152). If such cues were removed from verbal articulation, they would have to be supplied by inference or other contextual cues. Like zero in mathematics and the dark space in the theater, deixis orients us within a situation without calling attention to itself.             

  Deictic expressions are so pervasive and systematic in human languages that deixis could be taken as part of semantics proper. However, in a truth-conditional account of meaning this would not be possible because “it directly concerns the relationship between the structure of languages and the contexts in which they are used”(Levinson, 1983:55) and truth-conditions cannot make reference to contextual parameters of meaning. Hence, deictics, which are often described as being on the borderline between semantics and pragmatics, appear to be best accounted for in pragmatics (see Levinson 1983:55, 94-96).

Traditionally, it is essential to distinguish different kinds of usage of deictic expression. Deictic expression refers to those linguistic units or morphemes that have a deictic usage as basic or central, for most such expressions have non-deictic usages. In addition to deictic vs. non-deictic usages of deictic expressions, traditional accounts also distinguish distinct kinds of deictic usage. Following Fillmore (1971), there are two kinds of deictic usage, namely gestural usage and symbolic usage. Terms used in a gestural deictic way can only be interpreted with reference to an audio-visual-tactile, and in general a physical monitoring of the speech event, while symbolic usages make reference only to contextual co-ordinates available to participant antecedent to the utterance. Within non-deictic usages, anaphoric and non-anaphoric usages are distinguished. An anaphoric usage is where some term picks out as referent the same entity (or class of objects) that some prior term in the discourse has picked out. The distinctions can be illustrated by the word “there”:

(1)   I want you to put it there.

(2)   Is Johnny there?

(3)   I drove the car to the parking lot and left it there.

“There” in the above examples has three usages. Its gestural usage can be seen in example (1). We have to know where the speaker is pointing in order to know what place he is indicating. The symbolic usage is exemplified in the telephoner’s utterance, such as example (2). This time we understand the word “there” as meaning “in the place where you are”. The anaphoric usage of “there” is in example (3). In that case, the word refers to a place which has been identified earlier in the discourse, namely the parking lot.

Generally speaking, traditional deictic categories include person, place, time, discourse and social deixis (Lyons, 1968, 1977, Fillmore, 1971, 1975, and Levinson, 1983). The following is a very brief account of these deictic categories.

2.1. Person Deixis 

Person deixis is deictic reference to the participant role of a referent, such as the speaker, the addressee, and referents, which are neither speaker, nor addressee. Person deixis clearly operates on a basic three-part division, exemplified by the pronouns for first person, second person and third person. Pronominal systems, which are the most obvious manifestations of person, generally exhibit this three-way distinction.

Person deixis primarily makes reference to the speaker as the deictic center of the speech event and addressee. It is to be expected that as interlocutors take turns in the speech event and interchangeably become speakers and addressees, the deictic center switches from the one interlocutor to the other. In this sense, person deixis is said to encode participant roles. In some languages, person deixis may also encode other participant roles such as those of the by-stander, or of the hearer, who is other than the addressee, or those of the source, who is other than the speaker. But third person is quite unlike first or second person, in that it does not correspond to any specific participant-role in the speech event.

  Here it is important to see that the traditional category of plural is not symmetrically applied to first person in the way it is to third person: we does not mean plural speakers in the same way that they means more than one third person entity (Lyons, 1968: 277). In addition, in many languages, there are two first person plural pronouns, corresponding to “we-inclusive-of-addressee” and “we-exclusive-of-addressee”. This distinction is not manifested in English directly, but it is perhaps indirectly: for the contraction from let us to let’s only seems felicitous if the us is understood inclusively, as illustrated below (Fillmore, 1971):

(4) Let’s go to the cinema.

(5) ? Let’s go to see you tomorrow.

  In Chinese, “咱们” is a pronoun used to denote the speaker and addressee, so it is an inclusive usage of a plural first person pronoun. In general, “we” in English corresponds to “我们” in Chinese. For example:

(6) Right, we make a pretty good team at this, don’t we?

对，我们(咱们)是一队好搭档，是不是？

2.2. Place Deixis

Place deixis is deictic reference to a location relative to the location of a participant in the speech event, typically the speaker. The most obvious place-deictic terms in English are the adverbs “here” and “there” and the demonstratives “this” and “that”, along with their plural forms.

Locations can be specified relative to other objects or fixed reference points, as in:

 (7) The school is three hundred yards from the church.

Alternatively, they can be deictically specified relative to the location of participants at the time of speaking. If I say,

(8) “Johnny lives upstairs,”

You will understand me as meaning upstairs of the place where I am at the time I say the sentence, unless the immediately preceding discourse has provided some other reference point. If I say

(9)  “Harry lives nearby.”

The same can be said. You will understand that Harry lives near to the place where I am when I say the sentence, again, except for the case where a reference point has been identified in the immediately preceding discourse.

  For words that can refer to areas or spaces, the word “this” followed by an appropriate noun locates an object as being in the same area as the speaker is at coding time. Thus we can talk something like “in this room”, “on this planet”, “in this city”, etc. The English adverb “here”, when used for locating objects, is paraphrased as “in this place”, “at this place,” etc., in either the gestural or the symbolic use of “this”.

  But the demonstrative determiners combine with non-deictic terms for spatial organization to yield complex deictic descriptions of location. The non-deictic conceptual organization of space includes all those distinctions between surfaces, spaces, enclosures, containers and so on, and between fronts, backs, tops, and sides of objects, not to mention widths, lengths, heights, etc. The way we understand these expressions, especially those connected with the horizontal dimensions, depend on how we have imputed an orientation in space to the reference object. These expressions have additionally a use that shows up in deictically anchored situations. In particular, the location of the speaker and his outlook on the world can determine the orientation of the objects around him. Thus:

(10)   This side of the box.

can mean “the surface of the box that can be called a side which is nearest to the location of the speaker at coding time (or CT)”, but:

(11)   This side of the tree.

simply means “that area of the tree visible from the point where the speaker is at CT (or the space between that area and that point)”. The difference between the glosses for (10) and (11) depends clearly on boxes, but not trees, having intrinsic sides. The ambiguity may arise when intrinsic properties of reference objects interfere with the deictic use of non-deictic expressions, as in the following example:

(12)   The ball is behind the truck.

On the deictic reading of “behind”, the truck is between the ball and the speaker. On the non-deictic intrinsic-property-of-object reading, the ball is at the rear end of the truck.

  So far, place deixis has been related to the ways in which physical objects are located in particular places in the physical world, containing the location of speaker and /or addressee. Place deixis can also be related to the path taken by a moving object in relation to its source or origin and to its goal or destination, as expressed in certain motion verbs such as “come” and “go”. These verbs encode assumptions about the goal of the motion and the relative positions of interlocutors in a particular speech event (see Fillmore 1971: 221).

2.3. Time Deixis

Time deixis is reference to time relative to a temporal reference point. Typically, this point is the moment of utterance.

Time deixis makes reference to the ways particular cultures conceptualize and measure time. Generally speaking, in most languages, time is measured in cycles that refer to the 24-hour day and the ways it is divided into sub-sections of sociocultural significance. As Fillmore (1971) observes, these units either serve as measures relative to the deictic center including utterance time, or they serve as absolute time indicators of events. These two uses are illustrated in the following two examples:

(13) She arrived last night.

(14) She arrived in March, 2000.

The utterance in (13) is interpreted in relation to speaker time, that is, the day on which this utterance is spoken, while the one in (14) is interpreted calendrically, in terms of a point in time which is specified on an absolute time measurement scale.

  The important distinction concerning time deixis relates to the time an utterance is actually produced as opposed to the time it is received by the addressee. In this case reference is made to coding time (CT) and receiving time (RT) respectively. In the canonical situation of utterance, with the assumption of the unmarked deictic center, RT can be assumed to be identical to CT (Lyons (1977: 685) calls this assumption deictic simultaneity). The most prominent expressions in encoding time deixis in English are the adverbs “now”, “then”, “soon” and “recently”. When time deixis interacts with calendrical unit of time, words like “today”, “tomorrow” and “yesterday” are used to locate an utterance relative to time.

Another basic type of time deixis in English is in the choice of verb tense. Whereas other languages have many different forms of the verb as different tenses, English has only two basic forms, the present as in (15), and the past as in (16).

(15) I live here now.

(16) I lived there then.

The present tense is the proximal form and the past tense is the distal form. Because something that took place in the past, as in (16), is typically treated as distant from the speaker’s current situation. Perhaps less obviously, something that is treated as extremely unlikely from the speaker’s current situation is also marked via the distal (past tense) form, as in (17)

(17) I could be in Hawaii (if I had a lot of money).

2.4. Discourse Deixis

  Discourse deixis is deictic reference to a portion of a discourse relative to the speaker's current “location” in the discourse. Therefore, discourse deixis is deixis in text. Discourse deixis has to do with the choice of lexical or grammatical elements which indicate or otherwise refer to some portion or aspect of the ongoing discourse―something like, for example, “the former”. Most commonly, the terms of discourse deixis are taken from systems of deictic and non-deictic time semantics, for the very good reason that any point in a discourse can be thought of as a point in time―the time at which that portion of the discourse is encoded or decoded―with preceding portions of the discourse conceived as occurring earlier in time, later portions thought of as occurring later in time. Expressions in discourse deixis taken directly from non-deictic time semantics are words like “earlier” and “later”, and phrases like “the preceding x” and “the following x”. So, a text, whether in its written or oral realization, is closely related to the concepts of space and time. Since discourse unfolds in time, it seems natural that time-deictic or space-deictic words can be used to refer to portions of the discourse as in the following examples:   

(18) I bet you haven’t heard this joke.

(19) That was the funniest story I’ve ever heard.

(20) There’s a nice point to discuss in class.

(21) Here’s a powerful argument.

An interesting point about the use of spatial deictic terms to express discourse deixis is that the proximal-distal distinction in space deixis acquires temporal status in relation to the unfolding of the text. Thus this can be used to refer to a forthcoming portion of the discourse, as in (18), and that to a preceding portion, as in (19). Moreover, the use of here and there to express discourse deixis is also tied to the temporal dimension of discourse. In (20) this comment is typically made in reference to a point already made, whereas the argument is about to follow the statement made in (21), which in fact serves to introduce the argument itself.

2.5. Social Deixis

  Social deixis is reference to the social characteristics of, or distinctions between, the participants or referents in a speech event. According to Levinson (1983:89), social deixis should set limits to those aspects of language structure that encode the social identities of participants, or the social relationship between them, or between one of them and persons and entities referred to. There are of course many aspects of language usage that depend on these relations, but these usages are only relevant to the topic of social deixis in so far as they are grammaticalized. Obvious examples of such grammaticalizations are polite pronouns and titles of address, but there are many other manifestations of social deixis such as kinship terms, names and honorifics.

  Social deixis gives an indication not only of where the speaker stands in time and place, but also of his/her status within the social structure, and of the status the speaker gives the addressee. There are two basic kinds of socially deictic information that seem to be encoded in languages around the world: relational and absolute. Relational social deixis is deictic reference to a social relationship between the speaker and addressee, bystander, or other referent in the extralinguistic context. For example, the use of vous or tu in French, Sie or du in German, 你 or您in Chinese can show either power or solidarity, distance or closeness. English used to have you for distance, thou for closeness; now English has only retained the you, but social deixis in English expresses social position by other forms of address like “Jack”, “Jack X”, “Mister X”, “Professor X” and the like. Absolute social deixis is deictic reference to some social characteristic of a referent (especially a person) apart from any relative ranking of referents. Often absolute social deixis is expressed in certain forms of address. The form of address will include no comparison of the ranking of the speaker and addressee; there will be only a simple reference to the absolute status of the addressee. There are, for instance, forms reserved for certain speakers, addressees or by-standers, which are what Fillmore has called “authorized speakers” and “authorized recipients”. For example, in Chinese there is a special form that the emperor uses for self-reference.

Chapter 3 Limitations of Traditional Views on Deixis

From the traditional accounts of deixis, we may find that the usages of deictic expressions have been analyzed in considerable detail within pragmatics, but no single and comprehensive theory of deixis has ever been proposed. Even Levinson (1983:94) himself admits:

“The lack of theoretical discussion reflects the present state of our understanding: we have, on the one hand, only the rather simple philosophical approaches to indexicals (covering just some aspects of person, time and place deixis), and, on the other hand, a mass of complicated linguistic facts, to which some preliminary order has been brought by the work of Fillmore and Lyons in particular”.

In the absence of significant theories of deixis, linguists have generally tended to establish categories of deictics according to their function and the contextual parameters they define. As a result, some problems arise from the traditional accounts of deixis.

(i) Traditional accounts of deixis distinguish between deictic and non-deictic terms and usages of these terms on the basis of rigid criteria for membership in the deictic category, which obscure our understanding of deixis itself. For example, within non-deictic usages traditional approach distinguishes anaphora from non-anaphora. By definition, deixis depends directly on extra-linguistic context for establishing referents, whereas anaphora depends on prior establishing referents by some other means. Apparently, deixis and anaphoricity are prototypically incompatible phenomena. But sometimes it is perfectly possible that a deictic term is used both anaphorically and deictically. Take there for instance,

   (22) I was born in London and have lived there ever since.

There refers back to whatever place London refers to, but simultaneously contrasts with here on the deictic dimension of space, locating the utterance outside London. It is also quite possible for the gestural usage to combine with the non-deictic anaphoric usage too:

(23) I cut a finger: this one.

Here this one refers to whatever a finger refers to, but simultaneously must be accompanied by a presentation of the relevant finger.

A traditional account of deixis is faced with further problems when participant roles are encoded in other lexical categories such as titles, kinship terms, proper names and any permitted combination of the three when used as vocatives. For example, to a certain person named ‘王兵’，different vocatives may show different social relationship between the speaker and addressee.

(24) 王主任，这是我的调查报告。

(25) 王教授，你去年指导的那个学生来看你来了。

(26) 小王，你到我的办公室来一下。

(27) 王兵，把衣服洗一下。

（28）爸，这道题怎么做？

(29) 三叔，我爸晚上请你吃饭。

   The different vocatives or third person referential noun phrases, when used to mark the addressee in the speech event, as from example (24) to example (29), constitute socially constructed person deixis. They locate the speaker and discourse in a social world. The choice of one form will

certainly communicate something (not directly said) about the speaker’s                                                                  view of his or her relationship with the addressee. In those social contexts vocatives or third person referential noun phrases typically mark distinctions of the social status about the speaker and addressee. So the speaker in the above examples may be the addressee’s subordinate, academic, leader, wife, son, or nephew respectively.

In traditional views of deixis, vocatives are usually discussed under both person and social deixis (see Levinson 1983:71 and 92), while the deictic use of third person reference is only mentioned in the discussions of person deixis in which its social origin is evidently obscured and the deictic interpretation of the vocatives is usually sacrificed. In fact, person deixis cannot be analyzed independently of the sociocultural construction of reality.

As a result, this account of deixis is too rigid and at the same time too simple and the analysis of deixis in that framework is fragmentary and does not provide a comprehensive account of the phenomenon, but only partial descriptions at best.

(ii) An essential characteristic of all categories of deixis appears to be their egocentricity and the speaker generally constitutes the deictic center of the speech event. The deictic center is a reference point in relation to which a deictic expression is to be interpreted. Deictic centers are typically assumed to be as follows: a) the center person is the speaker, b) the center time is the time at which the speaker produces the utterance, c) the central place is the speaker’s location at utterance time or coding time, d) the discourse center is the point which the speaker is currently at in the production of his utterance, e) the social center is the speaker’s social status and rank, to which the status or rank of addressees or referents is relative. In the following example, the speaker, the actual location and the actual time of the utterance are, respectively, the deictic centers for the interpretation of I, here, and now:

(30) I’m over here now.

But there are various exceptions to this, in which deictic expressions are used in ways that shift this deictic center to other participants, or indeed to protagonists in narratives. This kind of phenomenon may be called “deictic projection”, which, coined by Lyons (1977: 579), refers to the various derivative usages in deictic expressions. Fillmore, Lyons and Levinson once mentioned some deictic derivative usages, but these usages are only regarded as a kind of language phenomenon, and have not been systematically studied.

Deictic projection can be found in all the traditional categories of deixis, especially in place and person deixis. For instance, it frequently happens that “this” is selected rather than “that”, “here” rather than “there”, and “now” rather than “then”, when the speaker is personally involved with the entity, situation or place to which he is referring or is identifying himself with the attitude or viewpoint of the addressee. The conditions which determine this empathetic use of the marked member of these deictically opposed demonstratives and adverbs are difficult to specify with any degree of precision. But there is no doubt that the speaker’s subjective involvement and his appeal to shared experience are relevant factors in the selection of those demonstratives and adverbs which, in their normal deictic use, indicate proximity.

(iii) Some particular expressions are used to express place, time, and discourse deixis in different contexts. For example, time deixis is often expressed with adjectives such as “next” and “last” and the demonstratives “this” and “that” when attached to specific time cycles and their names, as in “next week”, “last Sunday”, “this year”, and “that March”. The interesting feature of the adjectives is that they are also place deictic, as in “this room”, “that room”, “next stop”, “last station”. It is more interesting that the same set of words is also discourse deictic, as for example in “in this\that\next\last paragraph”, where the deictic center is the paragraph containing the utterance itself. There must be some conceptual and social relatedness among the three basic deictic categories, and it remains unexplored in the traditional accounts of deixis.

Chapter 4  Cognitive Approach to Deixis

Some scholars have proposed that the deixis is based on human perception and image schematic structure as well as the interactive character of communication. Buhler, Fillmore, Clark, and others stress that there is a unified conceptual centering of events underlying the linguistic fact, an egocentric modeling of reality in which the here, now and I of the speaker (and secondarily, of the addressee) have priority over other elements of the speech situation. For example, as Innis(1988) observes, Buhler (1982[1934]) advocates the fusion and intersection of perceptual and linguistic processes in the framework of his theory of language. He claims that deictic expressions refer to a deictic field of language whose zero point – the origo – is fixed by the person who is speaking, the place of utterance and the time of utterance (I, here, and now, respectively).  When a person uses a deictic term,

He also feels his body in relation to his visual orientation, and uses it deictically. His (conscious, experienced) body feeling representation stands in relation to the visual space. Spatial orientation can never, either in man or animal, be simply a matter of conceptually isolated vision. …data from vision, audition, and feeling are perceived and processed together by that already mentioned registration device and … another group of contributions is registered there too, contributions of so-called kinesthesis derived from specific intrinsic movements of head and body.” (Buhler(1982[1934]):25)

  Relating to the above embodied conception of deixis, Hanks’ (1992) interactive approach makes reference to an image schematic figure-ground relation. He thinks that the identification of a referent (figure) in relation to an indexical origo (ground) need not necessarily be the speaker. He observes that deictic reference “ organizes the field of interaction into a foreground upon a background, as figure and ground organize the visual field”(Hanks 1992:61). Rubba (1996) observes that, in actual usage, deictic expressions are not always used to identify elements in the immediate utterance situation. Ariel (1998) also claims that, even though references to speaker and addressee are quite common, it is their overall cognitive accessibility rather than their physical salience that determines how they are referred to.

  A further issue in the study of deixis concerns an appraisal of its social functions. Zupnik (1994) finds that person deixis in political discourse has the power of achieving social goals. More recently, Hawkins (1997) views deixis as a process of ideological grounding based on culturally sustained systems of images.

Based on the above analysis of deixis, the paper attempts an account of deixis from a cognitive linguistic perspective. It aims at solving the problems arisen in the traditional accounts of deixis. The approach to deixis adopted in the paper originates in cognitive semantics and comprises cognitive models, mental spaces as in Fauconnier(1985, 1997),  prototype structure as in Rosch (1978) and Taylor (1995), image schematic and metaphorical structures as in Lakoff (1987) and Lakoff and Johson (1999). The paper will be further claimed that a cognitive approach to deixis provides the basis for a unified treatment of this phenomenon in terms of which both cognitive and societal aspects of language use may be better understood.

4.1 The ICM of deixis

An important principle of experiential realism is that human knowledge is organized in terms of idealized cognitive models (ICMs). These are complex conceptual structures, gestalts, any element of which can correspond to a conceptual category. Concepts, then, are characterized relative to ICMs. When linguistic elements are associated with conceptual elements in ICMs, the result is a symbolic ICM. Thus, the meaning of a lexical item is represented as an element in an ICM.

  In his discussion of there-constructions, Lakoff (1987) describes the experiential gestalt (the ICM) of the central deictic there-constructions in locational terms as follows:

It is assumed as a background that some entity exists and is present at some location in the speaker’s visual field, that the speaker is directing his attention to it, and that the hearer is interested in its whereabouts but does not have his attention focused on it… The speaker then directs the hearer’s attention to the location of the entity (perhaps accompanied by a pointing gesture) and brings it to the hearer’s attention that this entity is at the specified location. Additionally… if the entity is moving, the motion may be indicated. And the speaker may choose to describe the entity or its location. (Lakoff 1987:490)

 Lakoff calls the above the pointing-out ICM of there-constructions, and convincingly argues that it gives rise to the prototypical structure of this category and motivates its various uses. It is clear from the above that this description of deixis is based on the following inferences: firstly, there should be an existential presupposition of an entity in space; secondly, the speaker intends to direct the hearer’s attention to the entity, and thirdly, the speaker’s intention is fulfilled by the use of the particular construction. It is reasonable to assume that whatever definition is proposed for the deictic ICM, it should include the three essential aspects of the ICM of there-constructions.

 Fillmore (1982:35) once described the participating role of interactants in the communicative situation, that is, their interchanging roles as speakers and addressees, their location in space and utterance time. So, participants can also be viewed as specific entities in space, whereas their utterances are temporally defined.

In fact, every utterance is spatiotemporally unique, being spoken or written at a particular place and at a particular time. We can specify the actual spatiotemporal situation of any utterance-act by giving its spatiotemporal co-ordinates. We may say, for example, that a particular utterance was produced by X in Hefei at 12 noon on 1 January 2003; and we can be more or less precise than this in our specification of the spatiotemporal co-ordinates of the speech act. The spatiotemporal co-ordinates are, however, only one part of the actual situation of utterance. Other components can also be described. For example, each of the participants must know his role and status. Linguistically, relevant roles are of two kinds: deictic and social. Deictic roles derive from the fact that in normal language behavior the speaker addresses his utterance to another person (or other persons) who is present in the situation and may refer to himself, to the addressee(s) or to the other persons and objects (whether they are in a situation or not), not by means of name or description, but by means of a personal or demonstrative pronoun, whose reference is determined by the participation of the referent in the language event, at the time of the utterance. Deictic roles are grammaticalized in many languages in what is traditionally called the category of person. Social roles are culture-specific functions, established in a society and recognized by its members, for example, the function of being a parent, a teacher or a priest. In many languages there is a richly different set of terms of address which the speaker must control if he is to produce appropriate utterances in various situations. Social role may also determine the selection of personal pronouns and associated components of the grammatical structure of utterances. A clear instance of this is the special first person pronoun “朕” used by the emperor in China. Generally speaking, however, it would seem to be status, rather than role, which is the determining factor in the selection of pronouns. But role normally implies status. There are, however, many aspects of language behavior that are systematically determined by social role: the use of various characteristic expressions by a judge addressing the jury or a preacher addressing the congregation, by lovers in situations of intimacy, and so on.

By social status is meant the relative social standing of the participants. Each participant in the language event must know, or make assumptions about, his status in relation to the other. And the participants must know where they are in space and time, too. The speaker of a language must control and be able to correlate at least two different systems of spatiotemporal reference: one is the deictic system whose co-ordinates are created by the act of the utterance itself; the other is the culture-specific system for referring to time and place that is lexicalized in the language he is speaking and also reflected, at least partly, by the deixis in the utterance.

Apparently, then, deictic categories reveal our conceptualization of human beings as objects in space and of human language as an object in time. The center of this conceptualization, on which the deixis is based, is the human being, or in communication terms, the speaker. To be specific, the spatial dimension of deixis should be of central importance. Besides, time is represented as the fourth dimension of space, along which discourse unfolds. Moreover, a fifth dimension emerges, that of social rank, which is also implicitly defined in terms of space. Significantly, the deictic center moves (another spatially dominated concept, here) while communicative interaction is in progress. Figure 4.1 will clearly illustrate the time and social dimensions of a deictic expression in space.

                  ( Figure 4.1)

Based on the above analysis, the paper proposes that the ICM of deixis involves an authorized speaker, an unfocused addressee, and a linguistic act of pointing to an entity in space by the speaker. Accordingly, a deictic expression is one that builds a mental space in which the speaker and the addressee are co-present at a given point in time. In short, a mental space evoked by a deictic expression involves the conceptualization of the deictic center. It is clear from the above definition that the cognitive gestalt of deixis is based on linguistic representation of a physical act performed by a human being in the presence of another human being.

4.2 Mental Space

  An ICM structures a mental space. Mental spaces are “constructs distinct from linguistic structures but built up in any discourse according to guidelines provided by the linguistic expressions” (Fauconnier 1985:16). They concern the understanding of any fixed or ongoing states of affairs such as immediate reality, fictional or hypothetical situations, past or future situations, representations of situations as in pictures and photos, or abstract domains such as subject matters (economics, politics, linguistics, etc.). Linguistic expressions functioning as space-builders include prepositional phrases, such as “in the play”, adverbs (“really”, “probably”), connectives (“if…then”, “either…or”), and clauses of prepositional attitude (Mary hopes…, believes…,claims…, etc.). Mental spaces are represented as sets of elements with relations holding between them. Elements in a space may have counterparts in another space. An element in one space may trigger another element, the target, in another space on the basis of a pragmatic function holding between the two elements. For example, one pragmatic function links authors with their works. Thus, reference to the author may trigger reference to his works, and the connector between the two is the above pragmatic function. It is suggested that “connectors are part of ICMs, which are set up locally, culturally, or on general experiential or psychological grounds” (Fauconnier 1985:10).

The mental space which is structured by the deictic ICM relates to the linguistic expressions that express it. This mental space is domain that builds up as we talk. For example, the drama ICM introduces and structures the mental space of a play. In other words, because we understand the drama as a make-up piece of literary discourse, we can also understand that a murder in a play will not actually cause the physical death of the actor.

  Going back to the deictic ICM, the paper suggests that it introduces and structures a mental space, which is built in discourse by the use of specific linguistic expressions. For example, in using the adverbial “now”, a mental space is built in the discourse which is structured by the deictic ICM, so that within this space the speaker and an addressee is assumed to appear at coding time. Similarly, the expression “this house” builds a mental space whereby, the co-presence of speaker and addressee is assumed within a particular location that is specified as the speaker’s location at coding time. In both of these examples, the mental spaces which are built by the corresponding linguistic expressions are structured by all aspects of the deictic ICM. That is, the speaker, who is authorized by his own utterance, points to an existing and at least temporally and locationally definite entity in relation to himself, to direct the unfocused addressee’s attention to it.

With regard to place deixis, it has been observed that the expressions “here” and “there” prototypically encode the relationship between the speaker and a location in which an entity is found. In this case, since the speaker is also understood as an entity in space, the use of these expressions builds a mental space in which both entities, the speaker and the object, exist in the physical space.

“There” is primarily and prototypically deictic in that it builds a mental space structured by the deictic ICM. However, when its referent is not specified within this deictic mental space, a non-deictically structured mental space is also built. In this case, it is the anaphoric usage. Apparently, if a particular location is not identified in terms of the speaker or the addressee, it is identified independently in terms of other locative expressions. For example,

(31) We’re there.

Fillmore (1971:226) has observed that “there” in this example refers to the place we previously mentioned as our goal, whereas Lyons (1977: 672) has claimed that this odd use of “there” can be explained in terms of the situational salience of this referent. Both Fillmore and Lyons are right in making these observations, which, however, do not explain this phenomenon in themselves. Why does the speaker use “there” instead of “here” to refer to the location where the speaker is at utterance time? Fauconnier’s (1997) blending theory may offer a convincing explanation. It seems that in this sentence ‘there’ builds a blended space. Blended space inherits structural elements from another two mental spaces, thereby making it possible to inferentially understand the expressions that seem to appear odd. Consider a classical example:

(32) If Churchill had been Prime Minister in 1983 instead of Nevill Chamberlain, Hitler would have been deposed and World War II averted.

This example asks us to blend conceptual structure from different mental spaces to create a separate mental space. The input spaces include (a) Churchill in 1983 as outspoken opponent of Germany and (b) Neville Chamberlain in 1938 as prime minister facing the threat from Germany. To construct the blend, we project parts of each of these spaces to it, and develop emergent structure there.

?

(Source: Turner & Fauconnier 1998)

From mental space (a), the blend takes Churchill. From mental space (b), the blend takes the role prime minister. In the blend, Churchill is Prime minister by 1938. The blend is contrary-to-fact with respect to both of its input spaces. The antecedent and the consequent exist in the blended space; neither exists in either of the input spaces.

Now, come back to the deictic word “there” in example (30). It builds a blended space which inherits elements from two deictic spaces. The one deictic space contains the speaker and the addressee at coding time, encoded by “now”, and the speaker’s current location as “here” and the addressee’s current location as “there”. The other deictic space contains the addressee at coding time, encoded by “now”, and the addressee’s location as here and the speaker’s location as “there”. In the blend, the speaker’s coding time is inherited from either deictic space as “now” and the speaker’s location from the second deictic space as “there”. This blend has two important consequences. Firstly, it leads to the understanding that the speaker is taking the addressee’s point of view. Secondly, it “moves” the speaker to a location other than the one he now occupies, namely the one that the addressee currently occupies and one that the speaker himself occupied at some point in the past. Inferentially, then, “there” makes reference to past time, that is, it is also time deictic. Taking the addressee’s present point of view and moving to past time lead to the further inference that the speaker is where he was in the past (in the past) expected (by the addressee and himself) to be.

Now, let’s come to two examples about the deictic projections.

(33) Four score and seven years ago our fathers brought forth on this continent, a new nation, conceived in Liberty, and dedicated to the proposition that all men are created equal. Now we are engaged in a great civil war, testing whether that nation or any nation so conceived and so dedicated, can long endure…. We have come to dedicate a portion of the field, as a final resting place for those who here gave their lives that that nation might live. It is altogether fitting and proper that we should do this. (A. Lincoln’s Gettysburg Address)

                                                ( source: 刘宇红，2002 )

  In example (33), “those” refers to the national heroes in American history, which is the conventional deictic usage, because something that took place in the past, is typically treated as distant from the speaker’s current situation, so the distal demonstrative pronoun “there” is used here. But, Abraham Lincoln made a speech in Gettysburg, which was a city of America. Why did he address America “that nation”? Furthermore, the motion verb “come” usually means “moving towards the speaker’s location at CT”, that is to say, we should “come here” instead of “come there”. Generally speaking, “that nation” should be corresponding to the adverb “there”. Traditional theory of deixis cannot give a clear explanation for this phenomenon. From the cognitive linguistic perspective, the space builder “four score and seven years ago” structures a mental space, which includes two elements “our fathers” and “a nation cherished liberty”; then the other space builder “now” structures a new mental space. As far as the new mental space is concerned, the two elements “our fathers” and “a nation cherished liberty” are outside the space, they are distal, so the deictic expressions “those” and “that” are used here.

?

               Mental space 1  (SB: four score and seven years ago)                       

               a: our fathers       b: a new nation                                      

?

                         New space  (SB: now)

                             c: Gettysburg

(34) 到四川来， 觉得此地人建造房屋最经济。火烧过的砖， 常常用来做柱子，孤零零的砌成四根砖柱， 上面盖上一个木头架子，看上去瘦骨嶙嶙， 单薄的可怜；……我现在住的“雅舍”正是这样一座房子。不消说，这房子有砖柱，有竹篱，一切特点都应有尽有。（梁实秋《雅舍菁华-雅舍》）

                                                   ( source: 刘宇红，2002 )

In example (34), Mr. Liang Shiqiu addresses his house as “这房子”, which satisfies the egocentric principle, and it coincides with the context. He lives in Taiwan,  which  is far away from Sichuan, but when he mentioned Sichuan, he used the words “此地”, and called the characteristics of the houses there “这样” . This cannot be explained within the scope of traditional theory of deixis, too. But mental space theory may offer a convincing explanation to this phenomenon. The space builder “到四川来” structures a mental space, in which the location of Sichuan is certainly addressed by “此地”. The houses in Sichuan and “雅舍”in Taiwan described by the author have similar structures and characteristics, and this analogy gives rise to blending. In blending, structures from two input spaces are projected to a separate space, the “blending”. The blend inherits partial structure from the input spaces, and has emergent structure of its own. The first input space is built by the space builder “到四川来”, in which the characteristics of the houses in Sichuan exists. In the other separate but related input mental space, which is structured by the space builder “现在”, the characteristics of “雅舍”occurs. These two input spaces share frame structure: there is a speaker, who describes the characteristics of the house (in input mental space 1, the speaker is there by default). This shared frame structure constitutes a third space, a generic space, connected both input spaces. There is a fourth space, the blend, which inherits “雅舍”from the second input space. Furthermore, the characteristics of “雅舍” is a fusing element from the both input spaces. So in the blend space, the speaker may address the characteristics of “雅舍” as “这样”.

 

 Mental space 1 (SB: 到四川来)

                           a: 此地 (四川)    b: 房屋

                           c: 火烧过的砖，砖柱， 木头架子……

   

4.3 Deictic Prototype Effects

  Basically, there are two ways to understand the notion of prototype. It can be deduced from categorization experiments. For instance, some members of a category first come to mind in association experiments and are recognized more rapidly as category members in verification tasks. If one takes these members as prototypes of the respective categories, this leads to definitions like “best example of a category”, “salient example”, “clearest cases of category membership”, “more representative of things  in a class” or “central and typical examples”.

But in cognitive linguistics the prototype is taken as a mental representation, as some sort of cognitive reference point. This definition may range from the more concrete notion of “image” or “schema” to the more abstract “representation of a category” or “ideal” according to the categories to which they are applied (see Lakoff 1986).

Unlike the homogeneous categories postulated by the logical hypothesis, cognitive prototype categories always consist of good and bad members and include marginal examples whose category membership is doubtful. Prototypical (or good) members of cognitive categories have the largest number of attributes in common with other members of the category and the smallest number of attributes which also occur with members of neighboring categories. Bad examples (or marginal category members) share only a small number of attributes with other members of the category, but have several attributes which belong to other categories as well. So, the category boundaries are fuzzy.

As far as deixis is concerned, traditional approaches to this phenomenon set a sharp boundary between deictic terms and deictic uses of terms. But deixis is sometimes clearly and strongly expressed, and sometimes weakly expressed. That suggests that deixis is not an absolute category but a graded one. Lakoff (1987:70) observes that if the ICM of a category fits a particular instance of this category perfectly, then this is a prototypical member, if the ICM fits less well, then the member is less prototypical, and if not at all, then the entity is not a member of this category. Given a candidate member of deictic category, if a situation fits perfectly the deictic ICM in terms of which the category is characterized, then the member is prototypical. If the situation fits less well, the member is less prototypical, and so on. In what follows the paper attempts to illustrate the prototype effects of deixis in terms of deictic ICM in person deixis.  

Apparently, person deixis operates on a basic three-part division, exemplified by the pronouns for first person (“我”), second person (“你”), third person (“他”, “她”, or “它”) and their plural forms. By definition, deixis depends directly on extra-linguistic context for establishing referents. “我” is used by the speaker to point to himself in the presence of the addressee, so it is prototypical person deixis. Similarly, the personal pronoun “我们” is used to encode the roles of speaker and possibly the addressee as entities in space. By using this term, the speaker identifies himself not as a single entity but rather as a member of group. If the group contains the speaker and specific addressee, then the use of “我们” or “咱们” is prototypical. For examples:

(35) 我们是兄弟。

  (36) 咱们吃饭去。

If the addressee(s) or even the speaker is excluded from reference, then the use of “我们” is less prototypical. When a general manager says to the villagers in a poor mountainous area,

(37) “我们要在这儿捐资建立一所希望小学”.

In this utterance, the addressees are excluded from reference, but some specific group is intended and conceived of as a set of entities in a mental space containing them. “我们” in this example has a representative function, and so the space is created in terms of the speaker and those she represents.

  Along the same lines, it is possible to analyze examples of the so-called editorial “我们”, or academic “我们”, as in the following examples:

(38) 本文只对转换生成语法作一概述，有关它的新发展我们准备另文介绍。

    (39) 学点语言学——这就是我们对语言教师，特别是外语教师提出的要求。

  These uses of the personal pronoun can also be considered marginally deictic in that the pragmatics of the discourse and academic writing violates the presupposition of the existence of a group of people containing the speaker as member. Moreover, “我们” in example (38) is an entity in discourse space. Apparently, if physical space is the prototypical space for deixis, discourse space is less prototypical and the entity in non-prototypical space is itself non-prototypical. Therefore, in this case, the deictic ICM only partly structures the mental space built by the personal pronoun, which is less prototypically deictic.

  The second person pronoun “你” or “你们” is used by the speaker to refer to the addressee(s). In this case, it is a prototypical deictic usage. However, it is also noticed that “你” or “你们”may take a more general reading, whereby the addressee is not precisely specified as in the following examples:

   (40) 你想在事业上取得成功，你就得好好学习，下一番苦功。

   (41) 他那刻苦钻研的精神你不能不佩服。

“你” in the above two examples presents varying degrees of specificity of the entities. Though “你” does refer to an addressee, the addressee does not carry the presupposition of definiteness. Hence this use is marginally deictic.

  In general, third person reference is not regarded as deictic in that the entity referred to is not identified in terms of the speaker pointing to it. For example,

(42) 我打了小弟一耳光，他没有还手。

  The pronoun “他” in (42) does not belong to the deictic category. Because it does not make reference to an entity relative to the speaker  but refers back to the referent picked out by “小弟” in the preceding clause, hence it is an anaphoric usage of the pronoun. From this example, we may conclude that deixis and anaphoricity are prototypically incompatible phenomena.

  However, sometimes, the third person referential expressions may have a deictic reading because they may encode the socio-personal role of addressee or speaker. When third person reference is used deictically to identify the addressee, or the speaker, the expression used for third person reference builds a mental space other than the reality deictic space including speaker and addressee. The reality space is subjective in the sense that it corresponds to the experienced co-presence of speaker and addressee. Given the interlocutor’s understanding of the co-presence reality space, third person reference builds a “mirroring” space in which interlocutors can observe themselves objectively and hence talk about themselves in third person. This can be shown when a teacher in the kindergarten says to a child:

(43) “有什么问题告诉老师, 好吗?”

The connectors linking the two spaces and the elements within them, such as speakers and addressees with their mirror images, make reference to socio-pragmatic knowledge and cultural practices invoking social institutions and corresponding social roles. A mirror space is triggered by an instance of non-prototypical deixis, whereby the co-presence of speaker and addressee in the real, deictic space is shifted by one of the interlocutor’s building a “mirror” space when using third person reference for himself or for his addressee.

It is generally assumed that third person reference to speaker or addressee serves to create some kind of social distance between them, often associated with power, authority, or solidarity and endearment . This seems to be the case when a mother told her son “涵涵，睡觉去”.

In view of such diversity of social meaning, I would like to claim that ‘mirror’ spaces and third person reference to speaker or addressee cannot be strictly associated with any one social variable. In ‘mirror’ spaces, social relationships may be created, maintained, reinforced or challenged. Because it is cognitively possible to imagine different worlds and different realities, language here is used to structure such realities in a way that is experientially and socially meaningful.

Moreover, the referential expression used to identify a party appears to encode the social relationship between interlocutors or the relationship each towards the referent. For example , a person named “王天一” may be addressed in different forms:

(44) 王局长在开会吗?

(45) 天一在家吗?

This is made possible in so far as forms can function as vocatives, such as proper names for instance, can never be socially unmarked, as Zwicky (1974) has observed. Reference to a third party (i.e.王天一) is made in the real deictic space, which, therefore, does not include the third party as one of the interlocutors at coding time. However, the socially marked use of name builds an imagined deictic mental spare including the speaker and the third party in the role of addressee. The use of a particular form of a address(王局长)to refer to addressee in this imagined deictic space is then inherited to the real deictic space of(44). Alternatively, the speaker may derive or “copy” a form of address from an imagined deictic space including the third party as addressee and the real space current addressee as speaker in the imagined space. So ,if I am a friend of “王天一”, I may use “王局长”to his surbordinate (because this is the term of address that his surbordinate would use towards him), but ‘天一’to this wife (because this is the term of address used in couples). Apparently, this process of setting up imagined, deictic and ‘copying’ their elements in the real, deictic space is a subjectification process. In short, where in (43) interlocuters are transferred to an objective, non-deictic space, in (44) and (45) non-interlocutors become part of the subjective, deictic space.

  So far the person deixis has been analyzed in terms of deictics as a category structured around an ICM and the deictic ICM gives rise to prototypical effects. From the analysis we may draw the following conclusions:

Firstly, like the other prototypical categories, deixis exhibits degree of typicality; not every member is equally representative for the category. In the deictic category, there are prototypically deictic usages, less prototypically deictic usages and marginally deictic usages. The deictic ICM can clearly show the prototype effect of deictic expressions.

Secondly, deictic category is blurred at the edge. Anaphoric identification of an entity very poorly fits the deictic ICM.

4.4 Physical Space Lays the Foundation for Deictic Categories

The importance of space in utterances is made obvious by the ways in which languages refer to objects: either by naming or describing them, or by locating them in space (Lyons 1977:648). This is no trivial matter and it reflects the human experience of perceiving entities in space. Looked at from one point of view, man is merely a middle-sized physical object. But in man’s world—the world as man sees it and describes it in everyday language —he is, in the most literal sense, the measure of all things. Anthropocentrism and anthropomorphism are woven into the very fabric of his language: it reflects his biological make-up, his natural terrestrial habitat, his mode of locomotion, and even the shape and properties of his body. We live and move normally, on the surface of the earth, and we do so, again normally, in an upright position. This gives us a fixed zero- point at ground-level (see Lyons, 1977:690). Given the egocentricity of deixis in general, a speaker is an entity in space, and his utterances are produced in that space. Thus participant roles, their social identification and their construction in and through discourse are inscribed in space. So the significance of place deixis in utterances cannot be over-emphasized.

The hypothesis that spatial expressions are more basic, grammatically and semantically, than various kinds of non-spatial expressions has been termed localism, and has been ordinarily supported by cognitive psychologists (Miller and Johnson-Laird, 1976). Spatial expressions are linguistically more basic, according to the localists, in that they serve as basis for other expressions, and the reason why this should be so, is that spatial organization is of central importance in human cognition. More specifically, it has been systematically observed that temporal expressions, in many unrelated languages, are patently derived from locative expressions. For example, as Traugott (1975) has argued, nearly every preposition or particle that is locative in English is also temporal, while the prepositions “for”, “since” and “till”, which are temporal rather than spatial in modern English, derive historically from locatives. What is true of prepositions and particles is true also of very many verbs, adverbs, adjectives and conjunctions, not only in English, but also in several other languages. Localism treats verb tense and aspect as instances of a kind of spatialization. Tense expresses time dexis, and there is an obvious parallelism between time deixis and place deixis. As “here” and “there” can be analyzed as meaning “at this place” and “at that place”, respectively, so “now” and “then” can be analyzed as meaning “at this time” and “at that time”. Moreover, there is an interdependence of time and distance, given that, in human experience, to cover a long distance requires more time.

  Aspectual distinctions are even more obviously spatial than tense distinctions. As a point is contained within a line, so an event can be located within the duration of a state or process, as the following examples illustrate:

(46) Our first child was born (at a time) when we were very hard up.

(47) He was run over (at a time) when he was crossing the road.

On the localism hypothesis, these structures give rise to corresponding structures in other, non-spatial domains, by means of analogy or metaphor, as in (48) and (49) below:

(48) John is in a state of blissful ignorance;

(49) John is in the process of cleaning his teeth.

In (48) a state is viewed as a spatial domain, where as in (49) a process is similarly understood.

  Localists usually treat temporal location as being less concrete than spatial location, but more concrete than various kinds of so-called abstract location. For example, looked at from this point of view, “in despair” can be said to be less concrete than “in London”. Much of what is commonly thought of as being metaphorical in the use of language can be brought within the scope of the thesis of localism. The use of particular prepositions, verbs or adverbs in English provides plenty of evidence for this view: prices are said to go up or come down, according to whether they increase or decrease; if A is judged to be better than B, A is said to be above or in front of B; and so on. The process whereby someone or something passes from one state to another may be accounted for in terms of the localistic notion of a journey. By journey metaphor, an entity is moving from a source towards a goal. For example,

(50) John became a teacher.

The utterance could be paraphrased as follows: “John was in a state of not being a teacher, which then changed so that he moved into the state of being one”. Analyzed in this way, all verbs denoting a change of state may be regarded as verb-of-motion. Even if-clauses can be analyzed along the same lines so that “if p then q” may be localistically analyzed as “q comes from p”.

  Given the importance of spatial organization in human cognition, the paper now turns to deictic categories in order to examine to what extent spatial conceptualization accounts for them.

 

4.4.1 Spatial Conceptualization of Social Deixis

A conceptual metaphor involves two experiential domains; one is understood in term of the other.  Usually, an abstract domain, the source domain cognitively structures the target domain while preserving its conceptual topology. In these terms the experiential domain of physical space is mapped onto the abstract domain of social reality and structure it in a way that is consistent with our understanding of this reality. People are “close” socially if they have an important social relationship, “distant” if they have none or only a peripheral one. To be specific, difference in social rank between speaker and addressee is understood as distance between them, in particular social distance. Social solidarity is understood as spatial proximity. As a consequence, social deixis is understood in terms of place deixis. This means that speaker is socially constructed in the same way that she is spatially constructed in deictic discourse by the use of deictic constructions.

The proposed analysis of social deixis in terms of place deixis is consistent with and confirms the spatializaion-of-form hypothesis (Lakoff 1987:283). This hypothesis requires a metaphorical mapping from physical space onto a conceptual space. According to this hypothesis, cognitive structure is understood in terms of image schemas plus a metaphorical mapping. Here, the paper will illustrate how conceptual, social space is understood through image schemas and metaphorical mapping.

Schema 1:  CENTRE vs. PERIPHERY

This schema is based on our experience of our bodies as having centers (e.g. the trunk) and peripheries (e.g. fingers, toes, etc.). Centers are viewed as more important than peripheries, which depend on the centers, while centers do not depend on peripheries. The structural elements of this schema, then, are in a spatial domain, a center and a periphery. The center is obviously the speaker, whereas the periphery carries the object of deixis as an entity in space. It is here proposed that conceptual, social relationship is understood in terms of this image schema and a metaphorical mapping of properties of physical space on properties of experienced social reality. For examples,

(51) He spoke of the social life of the ant. He warned against seeing ant life in terms of human life —though the language he spoke was colored by anthromophism. We have named the ants for ourselves: queen, worker, soldier, parasite, slave. We name their social behavior for ours: we talk of classes and casts. (A·S·Byatt, Still Life :195)

People are “close” socially if they have an important social relationship, “distant” if they have none or only a peripheral one. For instance,

(52)“咱们兄弟之间的话不足与外人道也。”

The implied meaning between “咱们”,“兄弟”and “外人” illustrates this image schema.

Schema 2:  UP vs. DOWN

It has been claimed (Lakoff 1987:283) that hierarchical structure is understood in terms of the UP-DOWN schema. Apparently, this schema is realized in space and based on the experience of the vertical orientation of our bodies. A social hierarchy is spatially understood in terms of the UP-DOWN schema, as shown in ‘climbing up the social ladder’ or ‘downward mobility’. Another example is derived from the basic locative experience UP vs. DOWN.

(53) The government was overthrown.

Lakoff argues that what lies behind the example is the metaphor “CONTROL IS UP”. “Before the event takes place, the government is in control (metaphorically upright), and afterwards it is not in control (metaphorically, it has fallen over)”( Lakoff 1987:439). In China, we usually say,

(54) “在旧社会，军阀及政客高高在上，劳苦大众处于社会的底层”。

4.4.2 Spatial Conceptualization of Time deixis

       It has been claimed that notions of time are expressed in terms of space in various languages; and that this kind of metaphorical expression actually reflects the general process of human cognition where time is conceptualized in terms of space via cross-domain mappings. The general conceptual metaphors of time are: TIME IS A MOVING OBJECT,  TIME PASSING IS MOTION. Specifically, time is understood in terms of things (i.e. entities and locations) and motion. The present time is at the same location as a canonical observer, with future times being in front of the speaker and past times behind the speaker. Either time or the speaker is moving while the other is stationary.

      When time is constructed as a moving object, it is also moving towards or away from the speaker. For examples:

(55) Thanksgiving is coming up on us.

      (56) 现在，新的“太平洋世纪”正在向我们走来。

The object is moving with the speaker as the goal or the destination. It is moving along a line, or a path, towards the deictic center. Conceivably, when the moving object reaches its goal, the time will be ‘now’, central time or, in traditional terminology, coding time. Similarly, the object is moving away from the speaker, towards the past in the following examples: 

(57) The time has passed when… 

(58) 流失的岁月冲淡了人们的记忆。

In the above cases, the speaker is the starting point of the motion. Time is moving away from the deictic center, away from now into the past.

  In understanding the passing of time as motion, we view one thing as moving and another thing as stationary; the stationary thing is the deictic center. Since motion is continuous and one-dimensional, the passage of time is also understood as continuous and one-dimensional. All the above examples entail that the time passing the observer is the present time.

However, sometimes, time is regarded as a fixed location with respect to which the observer is moving. In this case, times are conceptualized as bounded spaces, with units of time comparable to the units of length or distance. The observer, a traveler in this case, is now in motion. He has come out of the past times, is in the present time, and will enter the future times, as shown in the following example:

(59) 他们告别了阴沉的昨天，走向光辉的明天。

So far, it has been seen how time is conceptualized as entity in space. On the basis of time metaphors described above, the claim now can be made that time deixis is understood as place deixis in terms of the LINEAR image schema and these time metaphors. English expresses a linear representation of time in terms of which time deixis is specified through tense. Linear representation is essentially a spatial concept, since lines are two-dimensional objects in space. In the conceptualization of time as a moving object in space, the object is linearly moving towards the observer (the speaker or deictic center). When the observer is moving towards the time span, she will eventually arrive at the time span which will thus become “now” for the observer. In both cases, when the observer and time-object meet, it is “now” for the observer at arrival time. Before and after, the arrival time is “then”.

4.4.3    Spatial Conceptualization of Discourse Deixis

  Discourse may be understood in terms of the PART-WHOLE schema. It structures the metaphorical mapping of place onto discourse. This image schema arises from bodily experience, in that we experience our bodies as wholes with part. Lakoff (1978:273) claims that the PART-WHOLE schema is asymmetric in that if A is part of B, B cannot be part of A. Moreover, the whole exists only if all the parts exist in the configuration. If the whole is at place P, then all the parts are there, too. A typical property of the schema is that the parts may be contiguous to one another. In these terms, discourse is understood as spatial entity, which is the whole consisting of contiguous parts. The part of the discourse at which the speaker stands is the deictic, discourse center. The parts are also organized according to the ILNEAR ORDER schema which motivates their contiguity. Significantly, the speaker may move towards a part of the discourse, or the part may move in relation to him. For examples:

(60)  I shall now move to the next point.

(61) In the following paragraph, we’ll talk about long-term memory.

In (60) the speaker is moving from one part of the whole to the next. In (61), it refers to the part which is moving after the speaker and the speaker’s present discoursal location.

On either directionality of motion, when the part and the speaker meet at the same location, this location is the deictic center. This deictic center contains all of the elements of the here and now, or the phenomenon present for the user of the deictic terms.  For example, in fictional narrative, readers and authors shift their deictic center from the real-world situation to an image of themselves at a location within the story world. This location is represented as a cognitive structure often containing the elements of a particular time and place within the fictional world, or even within the subjective space of a fictional character. If a deictic center has been established, the reader can correctly localize those story aspects where they belong. The deictic center does not remain static within the story, but shifts as the story unfolds. The story tends to be constructed locally. Although there are many temporal and spatial shifts in the presentation of a text, the reader tends to witness most events as they seem to happen. The events tend to occur within the mental model at the active space-time location to which the reader has been directed by the syntax and semantics of the text. Once the existents and events of the story world are created by the text, they exist in relation to one another. Deictic terms, proper names, pronouns, definite descriptions, and other referring expressions almost invariably refer to these existents and events.

The understanding of discourse as motion in space in this double sense, as well as the linearity of motion, is probably responsible also for conceiving of discourse as temporal entity. The expression “unfolding the discourse in time” is characteristic of this complex way of conceptualizing discourse. Also, expressions such as “the beginning of this paragraph” or “the end of this paragraph” seem to point to a temporal understanding of discourse.

Chapter 5    Conclusion

Traditionally, deixis is described according to its function and the contextual parameters it defines. Though it provides a lot of valuable views, it also causes some problems. Firstly, traditional accounts of deixis distinguish between deictic and non-deictic terms and usages of these terms on the basis of rigid criteria for membership in the deictic category, which obscures our understanding of deixis itself. Secondly, traditional accounts of deixis cannot provide a convincing explanation for the phenomenon of deictic projection. Thirdly, such expressions as “this”, “that” are used to express place, time, and discourse deixis in different contexts, the traditional approach did not explore the underlying cause.

The paper approaches these problems from the perspective of cognitive linguistics. It proposes that deixis constitutes a pragmatic category which is structured around a deictic ICM. In this analysis, the deictic category exhibits prototype effects along a scale of prototypicality where the various deictic usages are distributed. Furthermore, it claims that the mental space which is built by the use of specific deictic terms can offer some reasonable explanations for the phenomenon of deictic projection. The paper also argues that social, time and discourse deixis are metaphorically derived from place deixis. This analysis is consistent with and confirms the spatialization-of-form hypothesis, which requires a metaphorical mapping from physical space onto a conceptual space. To be specific, social deixis is based on the metaphorical understanding of social space as physical space, time deixis is based on the metaphorical understanding of time as space, and discourse deixis is based on the metaphorical understanding of discourse as time and time as space.

Bibliography

Ariel, M. 1998. The linguistic status of the here and now. In Cognition linguistics 9: 189-238.

Buchler, J. 1955. Philosophical Writings of Peirce. New York: Dover.

Buhler, C. 1982. The Deictic Field of Language and Deictic Words. In R. Jarvella & W. Klein (Eds.), Speech, Place and Action: Studies on Deixis and Related Topics. New York: Wiley.

Coulson, S. & G. Fauconnier. 1999. Fake Guns and Stone Lions: Conceptual Blending and Private Adjectives. In Cognition and Function in Language, edited by Barbara A. Fox, Dan Inrafsky, Laura A. Michaelis, CSLI Publication. P143-158.

Fauconnier, G. 1985/1994. Mental Spaces. Cambridge, MA:MIT Press.

Fauconnier, G. & Turner, M. 1996. Blending as a Central Prrocess of Grammar. In Conceptual Structure, Discourse, and Language, edited by Adele E. Goldberg. Standford: Center for the Study of Language and Information.

Fauconnier,G. 1997. Mappings in Thought and Language. Cambridge:CUP.

Fillmore,C.J. 1971. Towards a theory of deixis. The PCCLLU Papers (Department of Linguistics, University of Hawaii).

Fillmore.C.J. 1975. Santa Cruz Lectures on Deixis. Mimeo, Indianna University Linguistics Club.

Fillmore, C. C. 1982. Towards a descriptive framework for spatial deixis. In Jarvella & Klein(eds.) Speech, Place and Action. John Wiley & Sons Ltd.

Gale, R. 1967. Indexical signs, egocentric particulars, and token-reflexive words. In P. Edwards (Ed.), The Encyclopedia of Philosophy: Vol.4 New York: Macmillan.

Goddard C. 1998. Semantic Analysis. Oxford University Press.

Hanks, William F. 1992. The indexical ground of deictic reference. In A. Duranti and Ch. Goodwin (eds), Rethinking Context: Language as an Interactive Phenomenon. Cambridge: Cambridge University Press, 43-76.

Haverkate, H. 1992. Deictic categories as mitigating devices. In Pragmatics 2: 505-522.

Hawkins, B. 1997. The social dimension of cognitive grammar. In W. A. Liebert, G. Redeker and L. Waugh (eds), Discourse and Perspective in Cognitive Linguistics. Amsterdam and Philadelphia: John Benjamins, 21-36.

Innis, Robert E. 1988. The thread of subjectivity: Philosophical remarks on Buhler’s language theory. In A. Eschbach (ed.), Karl Buhler’s Theory of Language. Amsterdam and Philadelphia: John Benjamins, 77-106.

Jarvella, R. J., & Klein, W. 1982. Speech, Place and Action: Studies on Deixis and Related Topics. Chichester: John Wiley & Sons Ltd.

Lakoff, G.&M. Johnson. 1980. Metaphors We Live By. Chicicago: University of Chicago Press.

Lakoff, G.& Brugman, C. 1986. Argument forms in lexical semantics. BLS 12: 442-54.

Lakoff, G. 1987. Women, Fire, and Dangerous things: What Categories Reveal about the Mind. University of Chicago Press.

Lakoff,G.& M. Johson, 1999. Philosophy in the Flesh- The Embodied Mind and Its Challenge to Western Thought. New York: Basic Books.

Levinson, S.C. 1983. Pragmatics. Cambridge: Cambridge University Press. 54-96.

Lyons, J. 1968. An Introduction to Theoretical Linguistics. Cambridge: Cambridge University Press.

Lyons, J. 1975. Deixis as the sourse of reference. In E.L. Deenan (Ed.) Formal semantics of natural language. Cambridge: Cambridge University Press.

Lyons. J. 1977. Sematics, Vols 1 & 2. Cambridge: Cambridge University Press.

Revzin, I. 1974. From Animal communication to Human Speech. In C. Cherry (eds.), Pragmatic Aspects of Human Communication. Boston: Reidel.

Rosh, Eleanor, 1978. Principles of categorization. In e. Rosch and B. B. Lloyd (eds), Cognition and Categorization. Hillsdale, NJ: Lawrence Erlbaum, 27-48.

Rubba, Jo, 1996. Alternate grounds in the interpretation of deictic expressions. In G. Fauconnier and E. Sweester (eds), Spaces, Worlds and Grammar. Chicago and London: The University of Chicago Press, 227-261.

Svorous, S. 1994. The Grammer of Space. Amsterdam: John Benjamins.

Sweetser, E. and G. Fauconnier. 1996. Cognitive links and domains: Basic aspects of mental space theory. In Fauconnier, G. and Sweetser, E. Spaces, worlds, and grammar. Chicago, IL: Yhe University of Chicago Press.

Taylor, J.R. 1995. Linguistic Categorization: Prototypes in linguistic theory. Oxford: Clarendon.

Ungerer, F.et al. 1996. An Introduction to Cognitive Linguistics. London: Longman.

Zupnik, Yael-Janette. 1994. A pragmatic analysis of the use of person deixis in political discourse. Journal of Pragmatics 21: 339-383.

陈治安, 彭宣维(Chen zhi-an, Peng wei-xuan) 1994. 人称指示语研究. 《外国语》第3期。

何兆熊(He Zhao-xiong)，2000.  《新编语用学概要》. 上海：上海外语教育出版社。

何自然(He Zi-ran)，1988. 《语用学概论》. 湖南教育出版社。

何自然(He Zi-ran)，1997. 《语用学与英语学习》. 上海：上海外语教育出版社。

刘宇红(Liu Yu-hong)，2002. 指示语的多元认知研究. 《外语学刊》第4期。

张辉(Zhang Hui)，1998. 论空间概念在语言知识建构中的作用. 《解放军外语学院学报》第1期。

张权(Zhang Quan)，1990. 常规语境中的指示现象. 《外语研究》第2期。

张权(Zhang Quan)，1994. 试论指示词语的先用现象. 《现代外语》第2期。

赵艳芳(Zhao Yan-fang)，2001. 《认知语言学概论》. 上海：上海外语教育出版社。