Language and Time: A Cognitive Linguistics Approach

English Linguistics 32: 2 (2015) 442–453© 2015 by the English Linguistic Society of Japan

* I am grateful to two anonymous EL reviewers for helpful comments and suggestions on an earlier version of this paper. The usual disclaimer applies.

－442－

[Review]

Language and Time: A Cognitive Linguistics Approach

By Vyvyan Evans, Cambridge University Press, Cambridge, 2013, xviii+266pp.

Masuhiro Nomura

Hokkaido University*

Keywords: time metaphor, temporal frames of reference, transience, LCCM

1. Introduction

Time, one of the foundational domains of human experience, has been an eternal topic in a broad array of disciplines, including philosophy, physics, psychology and anthropology. Linguistics is no exception, where grammat-ical topics of tense and aspect have been intensively studied in theoretical as well as descriptive terms. Cognitive linguistics, ever since the seminal work by Clark (1973) and Traugott (1975, 1978), has distinguished itself from traditional linguistics in its concern with how time is conceptualized and expressed figuratively. Vyvyan Evans’ monograph entitled Language and Time is a new contribution to this line of research. While his earlier work (Evans (2004)) focuses on analyzing the polysemy of the word time, this book deals with the phenomenon of “temporal frames of reference” (henceforth, t-FoRs), defined as “the means that humans have available to them in order to fix events in time” (p. 3). More specifically, the study is concerned with (i) describing the cognitive representations for t-FoRs, and (ii) exploring the way linguistic expressions for t-FoRs are interpreted. Is-sues (i) and (ii) are respectively discussed in Part II “Temporal frames of reference” (Chs. 3–7) and Part III “Meaning construction and temporal ref-erence” (Chs. 8–11) of the book. This review is organized as follows: Section 2 presents a summary of each chapter. Section 3 critically examines Evans’ analysis of the three types of t-FoRs, arguing that they are not conceptually independent, as

443REVIEWS

Evans would hold. Section 4 is the conclusion.

2. Chapter Summaries

Chapter 1 “Introduction” provides an introduction to the nature of tem-poral frames of reference and elaborates on research questions (i) and (ii). Chapter 2 “Access semantics” offers an overview of the “Theory of Lexical Concepts and Cognitive Models” (LCCM), a semantic theory pro-posed by Evans (2009) and adopted as the theoretical framework for explor-ing issue (ii) in later chapters. In Chapter 3 “The nature of temporal reference,” Evans argues that temporal experiences are directly perceived and phenomenologically real, contrary to the commonly held view that time is an emergent concept, metaphorically understood in terms of sensory-motor experience of motion through space. He then considers how time is similar to and different from space: time is analogous to space by sharing three parameters, i.e. “directed-ness” (i.e. whether the substrate in a given domain is symmetric/isotropic or asymmetric/anisotropic), “dimensionality” (i.e. the constituent structure of matter) and “magnitude” (i.e. the quantifiability of a given substrate) (pp. 63–65). Time is distinguished from space by the notion of “transience,” defined as “the subjectively felt experience of (temporal) passage” (p. 66). There are three types of transience, namely, “anisotropicity,” “succes-sion” and “duration,” which correspond respectively to the three parameters mentioned above. These three types of transience correspond in turn to three types of t-FoRs, i.e. “deictic,” “sequential” and “extrinsic” t-FoRs, ex-emplified by (1a/a′), (1b) and (1c), respectively:1

(1) a. We’re approaching Christmas. (p. 73) a′. Christmas is approaching. (p. 88) b. Christmas precedes/is before New Year’s Eve. (p. 118) c. Time flows on (forever). (p. 61)

The deictic t-FoR and the sequential t-FoR designate a “future/past rela-tion” and an “earlier/later relation,” respectively. The extrinsic t-FoR, on the other hand, represents the conceptualization of time as a matrix, i.e. “the event within which all other events occur” (p. 5). The table below presents a summary of the three types of t-FoRs.

1 (1a) and (1a′, b) instantiate what Clark (1973) called “Moving Ego” and “Moving Time,” respectively.

ENGLISH LINGUISTICS, VOL. 32, NO. 2 (2015)444

(2) Three types of t-FoRs (based on Figure 3.1 and Table 3.6)

Deictic t-FoR Sequential t-FoR Extrinsic t-FoRParameter Directedness Dimensionality MagnitudeType of tran-sience

Anisotropic Succession Duration

Temporal rela-tion

Future/past Earlier/later Matrix

Example (1a, a′) (1b) (1c)

Evans posits the following four coordinates in order to identify an event with respect to a specific transience type and to describe a linguistically en-coded t-FoR (p. 78):

(3) Target event (TE): The event, in a temporal scene, that is identified with respect to transience.

Reference point (RP): The point which is deployed to fix the TE. Origo (O): The point that serves to ground the RP

to the transience type that defines the t-FoR.

Perspective point (PP): The perspective from which the tempo-ral scene is viewed; this can take the perspective of the TE or the RP.

To illustrate how these coordinates work, let us consider the examples in (1a) and (1a′). In (1a), the TE Christmas is fixed with respect to the RP We, which is also the O, “the ego-centric experience of now.” In (1a′), the TE is again Christmas and the RP and O remain inexplicit. (1a) and (1a′) are distinguished in that the PP is fixed at the RP in the former and at the TE in the latter.2 Chapters 4–6 present a detailed analysis of the three t-FoRs, describing their neurological bases, natures, and constructional types. Constructional types are captured in terms of “t-FoR constructions,” a subset of argument-struc-ture constructions, each composed of “vehicle” (form) and “lexical concept” (meaning). Lexical concepts are distinguished from each other by exhibit-ing a range of unique formal and semantic selectional tendencies (cf. (5)–

2 In Evans’ framework, PP is assigned to the subject of a sentence. I am not sure whether the subject of a sentence always corresponds to the perspective from which the temporal scene is viewed.

445REVIEWS

(7)). Evans offers a careful analysis of t-FoR constructions for each t-FoR. Deictic t-FoR lexical concepts, for example, are clustered as follows (p. 88):

(4)

To take one example, [future], the top right-hand classification in (4), is further classified into [imminence], [proximal imminence] and [fu-ture]. [imminence] is described as a pairing between the following lexi-cal concept and vehicle (p. 90):

(5) −lexical concept: [event x fixed as being future-based and imminent, with respect to egocen-tric experience of now, from perspective of event x]

−vehicle: NP VP e.g. Christmas is approaching/coming up/getting close.

Similarly, [proximal imminence] and [future] are described by (6) and (7), respectively (pp. 93–94):

(6) −lexical concept: [event x fixed as being future-based and proximally imminent, with respect to egocentric experience of now, from perspective of event x]

−vehicle: NP BE PrepP e.g. Christmas is near/close (by)/nigh/around the corner. (7) −lexical concept: [event x fixed as being future-based,

with respect to egocentric experience of now, from perspective of event x]

−vehicle: NP SITUATIONL VP P NP e.g. Christmas is/lies {ahead of / in front of} us.

[FUTURE]

Target-event PP [PRESENT]

[PAST]

[DEGREE OF REMOVE]

Deictic t-FoR lexical concept

[FUTURE]

[PRESENT]

Reference-point PP [PAST]

[DEGREE OF REMOVE]


In Chapter 7 “Time versus space,” Evans poses the question of why time is structured in terms of space, despite being directly perceived and phe-nomenologically real. His proposal is that we experience the world by per-ceiving events, which are by nature temporally structured, and that “as time is not the object of perception, but the manner in which it is facilitated, our representational systems re-utilise the perceptually correlated sensory-motor reflexes for purposes of re-presentation of time in the conceptual system” (p. 153). He then compares spatial and temporal references in terms of domain-general as well as domain-specific reference strategies. In Chapter 8 “Conceptual metaphors and lexical concepts,” Evans argues for a dissociation (and, for that matter, a division of labor) between concep-tual metaphors and lexical concepts. He claims that neither the different readings associated with time, indicated by square brackets in (8), nor the selectional tendencies exhibited by each lexical concept (e.g. the [matrix] lexical concept cannot co-occur with deictic motion, as in *Time is flowing toward us) can be fully accounted for by the conceptual metaphor time is motion of objects alone:

(8) a. The time for action has arrived. [moment] b. Time flies when you’re having fun. [duration] c. The young woman’s time [=labor/childbirth] approached.

[event] d. Time flows on forever. [matrix] (p. 181)

Chapter 9 “Figurative meaning construction in LCCM theory” argues that figurative language understanding arises from the way the cognitive models of a lexical concept are organized—the distinction between the cognitive models to which a lexical concept facilitates direct access (primary cognitive models) and indirect access (secondary cognitive models), degrees of en-trenchment of lexical concept(s), and the length of access route to the rele-vant cognitive model. For example, France rejected the EU constitution is judged figurative by virtue of a clash arising between the primary cognitive models associated with France, i.e. [geographical landmass], [nation state] and [holiday destination], and the information associated with the expression rejected the EU constitution. This clash triggers a search in the secondary cognitive models, which achieves a match with the cognitive model of [electorate] and creates a relevant figurative conception. Chapter 10 “Semantic affordances and temporal reference” maintains that figurative meaning construction involves activation of not only conceptual metaphors, but also “semantic affordances,” defined as inferences arising from the activation of a cognitive model to which a lexical concept facili-

447REVIEWS

tates access (p. 221). For example, while both Christmas is approaching and Christmas whizzed by (this year) are licensed by the conceptual meta-phor time is motion of objects, they differ with respect to the semantic affordance associated with each lexical concept, i.e. ‘relative imminence’ (associated with [approaching]) and ‘compressed durational elapse’ (as-sociated with [whizzed by]). It is also pointed out that some temporal reference expressions (e.g. Christmas has disappeared.) are interpreted by deploying semantic affordances alone. Chapter 11 “Universals and diversity in the cross-linguistic representation of time” addresses the issue of cross-linguistic variation in temporal refer-ence. It is proposed that the extrinsic t-FoR is likely to emerge diachronic-ally later than the deictic and sequential t-FoRs, that Time-RP conceptual metaphors (e.g. Christmas comes before New Year.) are presumably univer-sal, and that commodity metaphors for time (e.g. save/waste/lose time) are not universal. The chapter ends with a summary and implications of the book.

3. Evaluation: Are the Three Types of t-FoRs Independent of One Another?

As briefly summarized in the previous section, Evans’ new book covers a wide range of topics concerning temporal frames of reference. Here, I will limit myself to evaluating his analysis of the three types of t-FoRs by con-sidering whether they are independent of one another. I will argue that the sequential t-FoR and the extrinsic t-FoR are conceptually dependent on the deictic t-FoR, contrary to what Evans would have us believe.

3.1. Sequential Temporal Reference as Ego-based As we saw in Section 2, the sequential t-FoR, grounded in the transience type of succession, describes the temporal relation of earlier/later. For the sequential t-FoR, Evans equates the O with the RP (p. 115), claiming that “the RP and O do not have an egocentric basis, but inhere in the event se-quence itself” (p. 116).3 Thus, in (1b) (=Christmas precedes/is before New Year’s Eve.), Christmas (TE) is fixed and sequenced with respect to New Year’s Eve (RP/O). This account falls short of motivating the precedence

3 Likewise, Moore (2014: 66) assumes that “the field-based frame of reference [= Evans’ sequential t-FoR—M.N.] has to do with a relationship between times that holds regardless of ego’s perspective.”


relation between TE and RP/O; in other words, it remains unclear why pre-cede or before is used instead of follow or after in (1b). I suggest that the sequential t-FoR is egocentric in nature, designating relative motion of events with regard to Ego (observer): from the viewpoint of Moving Ego, who moves forward in the direction of the future, the future events fixed on the time axis appear to be moving toward him/her from front to back. This rel-ative motion makes it appear that the earlier event ‘precedes’ the later event. A strong piece of evidence for this alternative account comes from the West African language Hausa (Hill (1978)). In Hausa, Monday is viewed as ‘behind’ Tuesday, while Tuesday is viewed as ‘in front of’ Monday (Hill (1978: 528)). This seemingly peculiar conception is accounted for in terms of the way Hausa speakers perceive objects in space. For instance, when English speakers are asked to put a red block in front of a green one, they will arrange the blocks as if the blocks were facing them. Hausa speakers, by contrast, are likely to arrange the blocks as if the speaker and the blocks were in tandem, construing the object farther away from the speaker as ‘in front of’ the object closer to the speaker.

(9) Spatial perception by English (top) and Hausa (bottom) speakers

Hausa speakers take the same in-tandem perspective when they talk about temporal sequences, conceptualizing the event farther away from the speaker (e.g. Tuesday) as ‘in front of’ the event closer to the speaker (e.g. Monday).4

4 This suggests that Ego is not in motion when it views temporal sequences in Hausa; otherwise, events would have relative motion with regard to Ego. In English, by con-

green red

red greenrrededdd eneeenee

reddddgreen

449REVIEWS

In sum, the sequential t-FoR, which involves such lexemes as before, after, precede, follow and come, is conceptually dependent on the deictic t-FoR in the sense that the ascription of motion or the front/back orientation to events could not arise without Ego.5 Under this analysis, the O in the sequential t-FoR is identified with Ego, rather than with the RP.

3.2. Extrinsic Temporal Reference as Ego-based Previous work on time metaphor often confuses Moving Time (e.g. (1a′)) and “time as such” (e.g. (1c)). For example, Lakoff and Johnson (1999: 158) confuse the two when they write that “[i]n this metaphor [=the Flow-of-Time metaphor; M.N.], the present is the part of the river that is pass-ing us, the future is the part of the river flowing toward us, and the past is the part of the river that has already flowed past where we are.” Here, what they regard as flowing is actually a sequence of events, which, by virtue of relative motion, appears to go past us, from front (future) to back (past).6 Evans avoids this confusion by positing the extrinsic t-FoR as a distinct category.7 The extrinsic t-FoR is claimed to be “an intellectual achievement” (p. 61) made possible by the reification of duration; it is used to provide “a means of fixing an event in an ‘absolute’ way, without refer-ence to an observer” (p. 129). Now, does the extrinsic t-FoR really fix an event “without reference to an observer”? Consider (1c) (=Time flows on (forever)), which instantiates the Flow-of-Time metaphor. Why is it that time is believed to flow from the past to the future? I suggest this is because the past is conceptualized as located higher on a path than the future. This difference in height enables

trast, Ego is in motion when viewing temporal sequences; I am inclined to take the con-trast January is {before/?in front of} February (Evans (2004: 230)) as indicative of rela-tive motion, rather than factive motion. See also Núñez and Sweetser (2006: 408). 5 This type of analysis is adumbrated by Traugott (1975) when she wrote “It appears that the concept of serial ordering, as developed linguistically, is established with respect to the speaker only insofar as it is seen as an asymmetric relationship, the spatial image of which is founded on our visual and tactile perception” (p. 220). Evans correctly ana-lyzes Christmas arrives/comes before New Year’s Eve, which pertains to [immediacy of presence], as involving elements from both deictic and sequential t-FoRs (p. 123). My point is that the entire sequential t-FoR, rather than its subset, is egocentric in nature as well as in origin. 6 The time of an event is metonymically related to the event. 7 Likewise, Moore (2014: 60) distinguishes these two by positing time is a mover for the extrinsic t-FoR.


time to move in a downward motion from the past to the future:8

(10) a. The authentic history of Arabia scarcely ascends to the fifth century of our era. (James Bell, A System of Geography, Popular and Scientific, Or a Physical, Political, and Statisti-cal Account of the World and Its Various Divisions, Volume 4, p. 217)

b. ideas that descend from those of ancient philosophers (LDCE) c. from ancient times down to the present day (KDEC)9

d. The meeting was moved up from Thursday to Tuesday. (O-LEX)

e. Let’s move the meeting down to three o’clock. (O-LEX) How do we know that time flows? If the flow of time represented “the event in which all other events occur” and we were all embedded in the event, we would not know whether time is in motion or not (just like we, living on the earth, do not feel the earth moving); we need to step aside the flow of time to observe it. Linguistic evidence in (11) indicates that this is exactly the case: we recognize the flow of time because time and Moving Ego proceed separately, while sharing the same path:

(11) a. race against time b. {ahead of / behind} time c. Time has finally caught up with me.

This suggests that the extrinsic t-FoR is conceptually dependent on the deictic t-FoR. I would further venture to speculate that the very concept of the flow of time may result from the subjectification of the motion of Ego (in a way that is analogous to Langacker’s (1998) analysis of be going to). In sum, I would analyze (1c) such that the TE is time and the RP is Ego, which remains implicit.10

8 It is sometimes asserted that English does not use the vertical axis in temporal ex-pressions (e.g. Kövecses (2010: 41)), but this is not correct as the examples in (10) indi-cate. 9 The idiomatic expression down the road (e.g. Many authorities are concerned that long-term side effects will show up years down the road. (IDM)) is not incompatible with the past is higher, future is lower conceptualization, though down can be used in the sense of ‘along’ (e.g. There is a pleasant little cafe bar a hundred yards down the road (LDCE)). Also, I analyze the expression up to now, which is close in meaning to down to the present, as instantiating the more is up metaphor, whereby each day is added to a stack. 10 As an example of the extrinsic t-FoR, Evans analyzes Christmas 1914 saw a foot-ball match between British and German forces, such that the TE is Christmas, the RP

451REVIEWS

To summarize, I have argued in this section that both the sequential t-FoR and the extrinsic t-FoR are conceptually dependent on the deictic t-FoR.11 The three types of t-FoRs can be diagrammed in an integrated man-ner (adapted from Nomura (2005: 13)):

(12) The integration of the three types of t-FoRs

This figure depicts a single complex frame consisting of a path, Ego, time and events: the deictic t-FoR is represented by Moving Ego going down the sloped path from the past to the future. The sequential t-FoR is repre-sented by a succession of events which are fixed on the path and in relative motion (indicated by dashed arrows) with regard to Moving Ego. Finally, the extrinsic t-FoR is represented by time moving in the same direction as Moving Ego. The three types of t-FoRs can thus be represented as a single coherent whole, with the sequential and extrinsic t-FoRs being conceptually dependent on the deictic t-FoR. Given that the deictic t-FoR is grounded in human sensory-motor experience of moving about and perceiving objects in the world, sensory-motor experience deserves more importance than it re-

is 1914 and the O is the incarnation of Christ (pp. 133–134). This may be valid as an analysis of the Anno Domini dating system, but may not be so as an analysis of sentence-level constructions, which should be Evans’ main target of analysis (cf. p. 75). 11 Evans proposes an implicational hierarchy for t-FoR systems: Deictic > Sequential > Extrinsic (p. 161). This is in accordance with the present discussion in so far as the deictic t-FoR is regarded as primary.

Future

Past

time

event 2

event 1

event 3

Ego


ceives in this book.12

4. Conclusion

Evans’ monograph, despite criticisms made above, constitutes an impor-tant contribution to the linguistic study of time in general and temporal frames of reference in particular, laying the foundations for future research in this area: his detailed description of t-FoR constructions in English will serve as a model for analyzing t-FoR constructions in other languages, his proposed cross-linguistic universals are worth testing, and his account of figurative meaning construction deserves serious consideration and refine-ment. Furthermore, this work will provide a wealth of data and insights for neighboring disciplines concerned with the study of time. Undoubtedly, this is a must-read for those interested in the way time is experienced and expressed in language.

REFERENCES

Clark, Herbert H. (1973) “Space, Time, Semantics, and the Child,” Cognitive Devel-opment and the Acquisition of Language, ed. by Timothy Moore, 27–63, Aca-demic Press, New York.

Evans, Vyvyan (2004) The Structure of Time: Language, Meaning and Temporal Cognition, John Benjamins, Amsterdam.

Evans, Vyvyan (2009) How Words Mean: Lexical Concepts, Cognitive Models, and Meaning Construction, Oxford University Press, Oxford.

Hill, Clifford (1978) “Linguistic Representation of Spatial and Temporal Orienta-tion,” BLS 4, 524–538.

Honda, Akira (2011) “Zikuukan Metafaa no Keikenteki Kiban o megutte (A Note on the Experiential Basis of the Space-to-Time Metaphor,” Kobe Gaidai Ronsoo 62.2, 33–56.

Kövecses, Zoltan (2010) Metaphor: A Practical Introduction, 2nd ed., Oxford Uni-versity Press, Oxford.

Lakoff, George and Mark Johnson (1999) Philosophy in the Flesh: The Embodied Mind and Its Challenge to Western Thought, Basic Books, New York.

Langacker, Ronald W. (1998) “On Subjectification and Grammaticization,” Discourse and Cognition: Bridging the Gap, ed. by Jean-Pierre Koenig, 71–89, CSLI Pub-

12 See Honda (2011) for a cogent argument on the experiential basis of time meta-phors.

453REVIEWS

lications, Stanford.Moore, Kevin E. (2014) The Spatial Language of Time: Metaphor, Metonymy, and

Frames of Reference, John Benjamins, Amsterdam.Nomura, Masuhiro (2005) “Zikan no Gainenka ni kansuru Oboegaki (A Note on the

Conceptualization of Time),” Eigo Seinen 151.9, 10–13.Núñez, Rafael E. and Eve Sweetser (2006) “With the Future Behind Them: Conver-

gent Evidence from Aymara Language and Gesture in the Crosslinguistic Com-parison of Spatial Construals of Time,” Cognitive Science 30, 401–450.

Traugott, Elizabeth C. (1975) “Spatial Expressions of Tense and Temporal Sequenc-ing: A Contribution to the Study of Semantic Fields,” Semiotica 15, 207–230.

Traugott, Elizabeth C. (1978) “On the Expression of Spatio-Temporal Relations in Language,” Universals of Human Language, vol. 3, ed. by Joseph Greenberg, 369–400, Stanford University Press, Stanford.

AbbreviationsIDM = Collins COBUILD Dictionary of Idioms, 2nd ed.KDEC = The Kenkyusha Dictionary of English CollocationsLDCE = Longman Dictionary of Contemporary English, 5th ed.O-LEX = O-LEX Japanese-English Dictionary

[received December 20, 2014, revised and accepted May 8, 2015]

Faculty of Letters Hokkaido University Kita 10 Nishi 7, Kita-ku, Sapporo Hokkaido 060–0810 e-mail: [email protected]

Language and Time: A Cognitive Linguistics Approach

Documents