LANGUAGE AND LINGUISTICS 3.3:491-519, 2002
2002-0-003-003-000019-1
TIME PASSING IS MOTION*
Kathleen Ahrens Chu-Ren Huang
National Taiwan University Academia Sinica
In this paper, we focus on a specific instantiation of the conceptual metaphor
TIME IS SPACE, which is argued by Yu (1998) to be a universal metaphor on the
basis of data from Mandarin and English. We examine the instantiation TIME
PASSING IS MOTION. We point out that the source domains of MOTION and
SPACE must be clearly differentiated. Once these two domains are differentiated,
the inconsistencies in what is traditionally called the TIME IS MOTION metaphor
disappear for both English and Chinese. We also demonstrate that where TIME IS
A MOVING POINT OVER A LANDSCAPE, the ego is attached to this point and
must be plural when the verb is an achievement verb of movement, but may be
singular when the verb is an activity verb of movement. Lastly, we conclude that
when TIME IS A MOVING ENTITY the orientation of the ego is a conceptual
subsidiary that can be parameterized by different languages, as well as by
language change and variation.
Key words: time, conceptual metaphor, linguistic metaphor, ego
1. Introduction
The main point of the cognitive linguistic paradigm is that we use our bodily based,
concrete experiences to interpret and encode non-bodily based, abstract phenomena
(Lakoff and Johnson 1980). When we do so, we are using conceptual metaphors.
Conceptual metaphors are often so embedded in our language and ways of thinking about
the world, that we do not consciously realize we are structuring the world in a particular
way. For example, when speaking of the concept of time, people often refer to the concept
of money, as in the following English and Chinese examples:
(1) a. He spent a lot of time in meetings.
b. She invested a lot of time in her students.
*
This work was supported by NSC grants #88-2420-H-002-005 and #89-2411-H-002-063-MC.
We would like to thank Hui-Ru Xiong for providing examples and comments on an earlier
version of this work, and to thank Yung-O Biq, Vicky Lai, Dora Lu, Lily Yi-wen Su, Jim Tai
and other IsCLL-7 participants for their comments on an earlier version as well.
Kathleen Ahrens & Chu-Ren Huang
492
(2) a. Ta weile jiesheng shijian, zhu yici fan jiu chi qi tian.
s/he because save time cook once rice then eat seven days
‘In order to save time, s/he cooked once a week and ate (the same thing)
for seven days.’
b. Ta wei haizi fuchu shijian tai shao.
s/he for children pay-out time too little
‘S/he spends too little time with his/her children.’
Money is a physical object, which allows humans to obtain other concrete objects that
they desire. Money can be saved, spent, invested, etc., In addition, examples (1)-(2) also
demonstrate that English and Chinese speakers can structure and understand the concept
of time in terms of money. When time is structured in this way, the emphasis is on the
preciousness of time.1
Moreover, because the TIME IS MONEY metaphor maps
similar concepts in Chinese and English, it can be said to be a cross-linguistically valid
conceptual metaphor.
In what follows, we examine a conceptual metaphor that is argued by Yu (1998) to
be not only a cross-linguistically valid metaphor, but a universal metaphor on the basis
of data from Mandarin and English; namely, the TIME IS SPACE metaphor. We focus
on a specific aspect of the TIME IS SPACE metaphor, namely the TIME IS MOTION
THROUGH SPACE metaphor in English and then we present our analysis for the
Mandarin Chinese data. We then compare previous analyses of the metaphor with our
own, as well as contrast our analysis for Chinese with the English data.
Our analysis leads to the following theoretically significant generalizations: First,
the source domains of MOTION and SPACE must be clearly differentiated. Once these
two domains are differentiated, the inconsistencies in the TIME IS MOTION metaphor
disappear. Second, within the domain of motion, there are two subcases that map to the
target domain of time. In the first case, where TIME IS A MOVING ENTITY, we argue
that the ego involved is facing the past, and not the future in Mandarin as Yu (1998)
suggests. In the second case, we argue against the moving ego account adopted in
previous studies of metaphoric mapping (e.g., Lakoff 1993, 1994, Lakoff and Johnson
1999, and Yu 1998). Instead we argue that the mapping is anchored as a point moving
over a landscape. The ego only moves as a consequence of being necessarily attached to
this moving reference point. This attached ego analysis offers a logical account of the
following generalization that we discovered: that the ego must be plural when the event
is an accomplishment, but may be singular when the event is an activity. We conclude
that one universal metaphor of time has moving time as its primary conceptualization,
1
See Ahrens (2001) for a discussion on how to analyze the significance and emphasis of the
source domain for the interpretation of the target domain.
TIME PASSING IS MOTION
493
but that the orientation of the ego is a conceptual subsidiary of this metaphor that can be
parameterized by different languages, as well as by language change and variation.
2. TIME IS MOTION metaphor in English
Lakoff (1993, 1994) AND Lakoff and Johnson (1999) argue that there are two
conceptualizations of time in English within the metaphor TIME PASSING IS
MOTION. First, there is the TIME PASSING IS A MOVING OBJECT metaphor.
Examples of this metaphor are given in (3):
(3) a. The time will come when you will see the error of your ways.
b. Christmas will be here before you know it.
c. I can’t wait until the New Year comes.
In this metaphor, time moves towards the ego, the person who is the point of reference
in relation to the event, as schematized in Figure 1:2
Time
( .
}
[]
|
0
<------------------|------------------->
Past P Future
Figure 1: Time passing is motion of an object
There is another metaphor in English, the TIME PASSING IS MOTION OVER A
LANDSCAPE metaphor, as shown in (4):
(4) a. As we go forward into the new millennium, let us not forget our past
travails.
b. We’re approaching the end of the fiscal year.
c. Sara always rushes headfirst into a new year, without stopping first to
make any new year’s resolutions.
2
Upper-case ‘P’ indicates the present point in time.
Kathleen Ahrens & Chu-Ren Huang
494
In this metaphor, the time passing is understood as the ego itself moving over a
stationary landscape, passing units of time along the way, as illustrated in Figure 2:
( .
}
|
| \
0 0
<-|-|-|-|-|-|-|-|-|-|-|-|-|-|-|-|-|-|-|-|-|-|->
Past P Future
Figure 2: Time passing is motion over a landscape
In both cases, the orientation of the ego is towards the future, as the examples in (5)
show:
(5) a. We’re looking forward to the arrival of Christmas.
b. We’re looking forward to the end of the semester.
Yu (1998:92) argues that “the two special cases proposed by Lakoff (1990, 1993, 1994)
for English ... account for the spatialization of time in Chinese.” However, in what
follows, we show that these two subcases of the TIME (PASSING) IS MOTION
metaphor are not enough to account for the data in Mandarin Chinese.
Before we do so, however, we must first clarify the source domains in the English
cases. According to Lakoff (1993) the conceptual metaphor of time has the following
properties and inferences in English:3
The General Metaphor: TIME PASSING IS MOTION
Ontology: Time is understood in terms of things (that is, entities and locations)
and motion.
Background condition: The present time is the same location as the
canonical observer.4
Mapping: Times are things. The passing of time is motion. Future times are
in front of the observer; past times are behind the observer. One thing is
moving, the other is stationary; the stationary entity is the deictic center.
Entailment: Since motion is continuous and one-dimensional, the passage of
time is continuous and one-dimensional.
3
We have added the boldface.
4
Note that Lakoff’s ‘canonical observer’ is what we are referring to as the ‘ego’.
TIME PASSING IS MOTION
495
Special Case 1: TIME PASSING IS MOTION OF AN OBJECT
The observer is fixed; times are entities moving with respect to the observer.
Times are oriented with their fronts in the direction of motion.
Entailments: If time 2 follows time 1, then time 2 is in the future relative to
time 1. The time passing the observer is the present time. Time has a velocity
relative to the observer.
Special Case 2: TIME PASSING IS MOTION OVER A LANDSCAPE
Times are fixed locations; the observer is moving with respect to time.
Entailments: Time has extension, and can be measured. An extended time,
like a spatial area, may be conceived of as a bounded region.
(Lakoff 1993:216-218)
This analysis raises several issues. First, although we agree that the general metaphor is
TIME IS MOTION, the ontology contradicts the metaphor because it includes things
along with motion, as if they are separate cases, i.e., TIME IS AN ENTITY, and TIME
IS MOTION. It is possible that time is also understood as an entity,5
but if so, that
would then be a different source domain from the cases discussed here where time is a
moving entity (i.e., example (3)).
Second, in the special cases, what is being moved is different.6
In the first case,
time itself is moving. In the second case, the ego is moving; however, the source
domain ‘motion over a landscape’ does not entail or necessarily imply an ego. In fact, it
is ambiguous. This phrase could also mean that time is understood as motion over a
landscape, in contrast to the first case where time passing is motion, but not necessarily
over a landscape.
Clausner and Croft (1997) argue that metaphors that are not at correct level at
schematicity (i.e., too general or too narrow) will have gaps in their mappings. For
example, they point out that the metaphor IDEAS ARE BUILDINGS does not map
many concepts from the source domain to the target domain, such as plumbing and
wiring. They suggest that narrowing the source domain to THE STRUCTURAL
INTEGRITY OF THE BUILDING and having it map to the target domain of THE
CONVINCINGNESS OF THE ARGUMENT will have the desired effect of ruling out
concepts (such as plumbing) that cannot be mapped.
5
But see the discussion on schematicity below.
6
Lakoff and Johnson (1980) and Lakoff (1994) have pointed out that the mappings between the
two cases are inconsistent. However, as Ahrens (1995) points out, inconsistencies in the
metaphor system should be reanalyzed to see if the source domain has not been correctly
specified.
Kathleen Ahrens & Chu-Ren Huang
496
Narrowing the source domain in the second special case to a moving ego limits the
scope of possible mappings to the ego itself and excludes the possibility of rain, or other
objects moving over the landscape. Shinohara (1999) notes this fact in his discussion of
which motion verbs may be used in the TIME IS MOTION metaphor, suggesting that
only time or the ‘observer’ (what we are referring to here as the ego) can move. This is,
in fact, what we are suggesting as well. In the first subcase, time (understood as an
object) is moving; and, in the second case, time (understood as an ego) is moving. In the
former case, the object must be moving towards an ego; in the latter case, the ego must
be moving over a landscape.
However, our formulation of the subcases also differs from Shinohara (1999b)
because his two postulated submappings TIME IS A MOVING OBJECT and TIME IS
A LINE ALONG WHICH OBSERVERS MOVE again run into the problem that there
is no way to unify the notion of lines and moving objects in a general metaphor. The
subcases that we have suggested for TIME PASSING IS MOTION, on the other hand,
allow for the motion concept to be unified into moving entities or moving egos. This is
not to say that time is not understood as a line along which observers move, but that if
the target domain is limited to the concept of TIME PASSING as opposed to TIME
itself, the similarities in the source domains of the two subcases become apparent. In
addition, we shall also re-interpret Lakoff’s background condition as a presupposition.
We shall show below how this presupposition of the location of the ego plays a crucial
role in the metaphorical mapping.
With the reformulation of the special subcases, a more precise ontology becomes
apparent, as given in (6) for English:
(6) The General Metaphor: TIME PASSING IS MOTION
Ontology: The passage of time is understood in terms of entities
(objects or egos) that move.
Presupposition: The ego is situated at the present time.
Special Case 1: Time passing is an object that moves towards an ego.
Special Case 2: Time passing is an ego that moves across a landscape.
The mappings between the source and target domains for each of the special cases are
given below in Table 1. We shall revise this table after discussing the Mandarin data,
but it can serve as a starting point for comparison between the two languages.
TIME PASSING IS MOTION
497
Table 1: Mappings between source and target domains for TIME PASSING IS
MOTION (English)
Source Domain Target Domain (Case 1) Target Domain (Case 2)
moving entity time ego
orientation of moving entity towards ego towards future
orientation of ego towards future towards future
Table 1 sets out what is being mapped in each case. In case 1, the moving entity in the
source domain is mapped to time in the target domain. The orientation of time is
towards the ego. For example (3) above, all the instances relate to a point in time for a
particular ego (i.e., ‘you’ in (3a) and (3b) and ‘I’ in (3c)). The orientation of the ego is
towards the future (example (5a)). In case 2, the moving entity is the ego who is
oriented and moves towards a point in the future (examples (4) and (5b)).
In this section we have laid out a modified analysis of the TIME IS MOTION
metaphor in English. The main advantages to our reanalysis are 1) that it more clearly
defines the scope of the target domain, and 2) that we are thus able to specify what
concepts can be mapped between the source and target domains. However, it does have
a similar disadvantage to the proposal by Lakoff (1993, 1994) since the mappings
between the subcases are inconsistent; that is, the moving entity is being mapped to
time in the first case, but is mapped to the ego in the second case. The second case, as it
stands now, only has the notion of time passing because the ego moves over the
landscape.
In what follows, we shall analyze the same metaphor in Chinese to see if there is a
universal conceptualization of time as motion, as Yu (1998) claims. Before we leave the
discussion on English, however, we should like to mention two examples that do not fit
into either the Lakovian model or our revised model of TIME PASSING IS MOTION,
namely (7) and (8) below:7
(7) In the preceding weeks, we have failed to meet the minimum sales goals that
we set for ourselves.
(8) In the following weeks, we need to make a sustained effort to reach our
quarterly sales figures.
In (7), ‘preceding’ refers to the time prior to the time up to this point for the
7
This anomaly was first pointed out in Lakoff and Johnson 1980 (who noted that the issue was
raised by C. Fillmore (p.c.)), but was not resolved or discussed in later work (i.e., Lakoff 1993,
1994).
Kathleen Ahrens & Chu-Ren Huang
498
speaker and listeners. Note that this concept cannot map onto either of the TIME IS
MOTION subcases that we have mentioned above, because if X precedes Y, it means
that X starts before Y, which means that X must be in the future, given the orientation of
the ego is towards the future Figure 3(a). But the intended meaning is that the events
under discussion happened prior to this point in time Figure 3(b).
( .
}
[]
|
0
<-------------------|-------------------->
Past P Future
(a) Y X (following the definition of ‘precede’)
(b) X Y (intended meaning — weeks prior)
Figure 3: Anomaly with ‘precede’
In (8), ‘following’ refers to the time after to this point in time for the speaker and
listeners. Note that this concept also cannot map onto either of the TIME PASSING IS
MOTION subcases that we have mentioned above, because if X follows Y, it means that
X is behind Y, which means that X should be an event in the past, given the orientation
of the ego is towards the future Figure 4(a). But the intended meaning is that the events
under discussion will happen in the future relative to the present point in time Figure
4(b).
( .
}
[]
|
0
<-------------------|-------------------->
Past P Future
(a) X Y (following the definition of ‘follow’)
(b) Y X (intended meaning — weeks ahead)
Figure 4: Anomaly with ‘following’
TIME PASSING IS MOTION
499
We shall leave this puzzle and come back to it in the discussion section after we have
analyzed the TIME PASSING IS MOTION metaphor in Mandarin. To foreshadow our
findings, we shall argue that the orientation of the ego is a parameter that is
language-specific, while the concept of time passing as a moving entity towards an ego
is universal.
3. TIME PASSING IS MOTION metaphor in Chinese
In this section, we shall look at two subcases of the TIME PASSING IS MOTION
metaphor in Mandarin Chinese that are similar in several respects to the English TIME
PASSING IS MOTION metaphor.8
We shall first propose our analysis for Mandarin
and compare our proposals with previous analyses. In particular, we shall argue that the
orientation of the ego is the TIME PASSING IS A MOVING ENTITY is facing towards
the past, supporting C. Huang’s (1981) and Alverson’s (1994) analysis, but contrary to
Yu’s (1998) analysis. We shall also argue that TIME PASSING IS AN EGO MOVING
OVER A LANDSCAPE is not accurate for Mandarin, and instead suggest that the
source domain must be narrowed to a MOVING POINT that has an ego attached.9
3.1 TIME PASSING IS A MOVING ENTITY
One salient difference between English and Chinese concepts of time in the TIME
PASSING IS MOTION metaphor is that in Chinese the ego is oriented towards the past
(examples (9a-c)):
(9) a. qian-nian wo dao meiguo qu le.
front-year I to America go ASP
‘I went to America two years ago (the year before last).’
b. wo dasuan hou-nian jiehun.
I plan back-year marry
‘I plan to marry two years from now (the year after next).’
8
The Mandarin Chinese discussed in this paper refers to the Mandarin that is currently spoken
in Taiwan. Some classical Chinese examples are cited because they are more condensed in
form and meaning, and they also represent the same set of expressions that are used today.
9
Tai (1989) also proposes that Mandarin has both subcases of the TIME IS MOTION metaphor.
However, he does not delve into the orientation issue.
Kathleen Ahrens & Chu-Ren Huang
500
c. qian bu jian gu ren, hou bu jian lai zhe
front not see ancient people back not see come those
‘Before me, I can’t see any predecessors; behind me I can’t see any
followers.’
(Chen Zi-ang ‘A Ballad written while ascending You-zhou platform’
(cited in C. Huang 1981))
In the examples in (9), what is in front of the speaker is in the past, while what is in
back of the speaker is in the future. However, even though the ego’s orientation is
different, time still approaches the ego from the future as it does in English (examples
(10)-(13)).
Example (10) gives the definition of time as something that comes to the present
and goes to the past. Thus, time is approaching from the future.
(10) wang gu lai jin weizhi zhou.
go past come present call time
‘Time is what comes to the present and goes to the past.’
(Huai Nan Zi, Han Dynasty (cited in C. Huang 1981))
In (11), the speaker wants to enjoy the present, s/he doesn’t want to wait for the future:
(11) wei le dang ji-shi, he neng dai lai zi.
make merry should immediate how can wait coming pronoun
‘One should enjoy the present; there is no cause to wait for the time yet to
come.’
(Nineteen Ancient Poems (cited in C. Huang 1981))
In (12), a particular point in time (i.e., Christmas) is moving towards the speaker:10
(12) shengdanjie kuai dao le.
Christmas quick arrive ASP
‘Christmas is almost here.’
In (13) time is again coming towards the speaker. Note that the expressions in (13)
are highly lexicalized.
10
Discussion of a possible ambiguous reading of this example can be found in Section 3.3.
TIME PASSING IS MOTION
501
(13) a. jiang-lai
immediate come
‘future’
b. wei-lai
yet come
‘future’
c. lai-ri
coming day
‘future days’
d. lai-nian
coming year
‘future years’
(cf. Yu 1998:104, ex. (61a-d))
The conceptualization of time as a moving entity is schematized in Figure 5:
Time
{.
)
[]
|
0
<------------------|-------------------->
Past P Future
Figure 5: Time is a moving entity, ego is oriented towards the past
Thus, we propose that the first subcase of the TIME IS MOTION metaphor in Mandarin
is where time is understood as a moving entity that approaches an ego who is facing the
past.
3.2 TIME PASSING IS A MOVING POINT ON A LANDSCAPE
Can time passing in Mandarin also be conceived of as an ego moving and facing
towards the future, as in English? Example (14) is often cited in support of the moving
ego metaphor (cf. Huang 1977, Yu 1998):
Kathleen Ahrens & Chu-Ren Huang
502
(14) women yijing jinru le ershiyi shiji.
we already enter LE twenty-first century
‘We have already entered the twenty-first century.’
In addition to example (14), there are other examples such as (15):11
(15) women kuai dao qimokao le.
We soon arrive term-final-exam LE
‘We’re fast approaching finals.’
In example (14), the speakers have already attained a particular milestone on the
landscape. In (15), the speakers have yet to attain a particular milestone. These examples
indicate that milestones must be mapped from the source domain of landscape to the
target domain of time. In the target domain, these milestones indicate a temporal landmark
(i.e., a particular point in time). Figure 6 gives a preliminary conceptualization:
( .
}
|
| \
0 0
<-|-|-|-|-|-|-|-|-|-|-|-|-|-|-|-|-|-|-|-|-|-|->
Past P Future
Figure 6: Time passing is an ego moving over a landscape
(preliminary conceptualization)
In Figure 6, the landscape over which the ego is moving is dotted with landmarks of
time. The ego moves forward over the landscape, and in doing so passes these markers.
However, this explanation of the second subcase has the following problems. First,
it still involves primarily an ego moving over a landscape. It is the movement of the ego
that indicates the passage of time. As mentioned in Section 2, this means that this
subcase is inconsistent with the first subcase where TIME IS A MOVING ENTITY.
Second, it is interesting to note that in almost all the instances given in Yu (1998),
the ego can never be an individual.12
The ego may refer to a country (16a), mankind
11
The acceptability of this sentence varies with speakers’ ages. The majority of college-age
speakers found this sentence acceptable, while speakers a generation older varied more as to
its acceptability.
12
Yu’s (1998) data is based on data from the Mandarin Chinese that is presently spoken in
Mainland China.
TIME PASSING IS MOTION
503
(16b), or a group of people (i.e., (14), (15), and (16c)).
(16) a. Gongheguo zou guo le sishiwu nian bu pingfan de
republic walk over PRT forty-five years not ordinary MOD
licheng.
journey
‘The Republic has walked over an extraordinary journey of forty-five
years.’ (Yu 1998:117, ex. (91))
b. Renlei jijiang kua ru xin shiji.
Mankind soon stride into new century
‘Mankind will soon stride into the new century.’
(Yu 1998:119, ex. (96b))
c. Tamen zheng mai zhe da bu, ben xiang
they PRT stride PRT big step run-quickly toward
canlan de mingtian.
splendid MOD tomorrow
‘With big strides, they are running toward the splendid tomorrow.’
(Yu 1998:119, ex. (96d))
In addition, it is not possible to replace the plural pronoun form with a singular form, as
(17a-b) shows (cf. (14)-(15)):
(17) a. *wo yijing jinru le ershiyi shiji.
I already enter ASP twenty-first century
b. *wo kuai dao qimokao le.
I soon arrive term-final-exam LE
There is an apparent counter-example to the restriction on the ego as being a plural
entity (18).
(18) Ta zhi yong le shige yue jiu wancheng le
he only use PRT ten months already accomplish PERF
quannian renwu, tiqian jinru xiayi nian.
whole-year task ahead-of-time enter next year
‘He only used ten months to accomplish the whole year’s task, having
entered next year ahead of time.’ (Yu 1998:120, ex. (97))13
13
One reviewer pointed out that the nominalizer DE should be inserted between quannian and
renwu. While we agree with a reviewer on this point, there was no DE in the cited text.
Kathleen Ahrens & Chu-Ren Huang
504
In (18), what the speaker has entered is next year’s workload, not the next year itself.
Xiayi nian is a metonymic extension referring to the workload for next year.14
Thus ta
is not entering a new year; he is beginning his tasks for the next year earlier than
expected.
However, there is a case where a singular form can be used as in (19) below:
(19) a. wo bu zhidao ta zeme guo (tade) rizi.
I NEG know s/he how pass his/her day
‘I don’t know how s/he passes his/her days.’
b. ta mimihuhu de guo (tade) rizi.
s/he daze DE pass his/her day
‘S/he passed his/her days in a dazed manner.’
*ta guo (tade) rizi
s/he pass his/her day
In (19) the ego is a single entity that is moving through time in an indeterminate
manner.
Third, although it is the ego that must be moving over the landscape, the ego can
never stop, as example (20a) demonstrates. Moreover, it is possible for someone to wish
that time, when it is being understood as a moving entity, can stop, as in (20b).
(20) a. *tamen xiwang tamen bu hui jinru/daoda mingtian.
they wish they NEG can enter/reach tomorrow
b. Tamen xiwang mingtian yongyuan bu hui lai.
they wish tomorrow forever NEG can come
‘They wish tomorrow would never come.
Thus we must ask ourselves: Why would the target domain not allow an ego to
stop when it clearly could do so in the source domain? That is, a person walking over a
landscape could stop along the way. Moreover, in the first subcase, the source to target
domain mapping is straightforward. A moving entity can stop moving in the source
domain and when the moving entity is mapped to time, it can also stop.
These three questions－why inconsistencies between the subcases exist, why the
ego must be plural in all cases except for those involving the verb guo ‘to pass’, and
why the ego cannot stop and be detached from the moving time point－compel us to
14
Another possible explanation is that the phrase de renwu ‘MOD task’ has been ellipsed from
the latter part of the sentence (as in ‘the next year of work’).
TIME PASSING IS MOTION
505
reexamine the source domain. First, we observe that guo is an activity verb, while verbs
such as jinru ‘enter’ (example (14)) and dao ‘arrive’ (example (15)) are achievement
verbs. Moreover, examples such as (16)-(18), while not using achievement verbs, all
have a time phrase that indicates a point in time when a certain state will be achieved.
The achievement is demarcated by a temporal landmark which is a milestone in the
landscape.
The above two facts lead us to suspect that the contrasts have something to do with
how a temporal milestone is defined or perceived. This in turn leads us back to our
presupposition that the ego must always be located in the present time. Given that the
location of the ego must be at the present time-point, there is an alternative
conceptualization available. We could conceptualize that it is the reference time-point
that is moving while the ego moves by virtue of its location. This conceptualization will
account for all the data that were originally thought to support a moving ego metaphor.
It is also strongly supported by the fact that the ego cannot run ahead or fall behind a
particular point in time in Mandarin. As the reference point of the present moves
forward over the time landscape, the attached group ego moves with it. Example (21)
demonstrates that the ego must be attached, and that it must move forward with time, as
dictated by the presupposition and the metaphorical mapping of movement:
(21) Shijian bu deng ren.
time not wait person
‘Time will wait for no person.’ (cf. Yu 1998:126, ex. (106b))
Under this conceptualization of moving time-point with attached ego, we observe
that since the landscape is part of the mapping, the observer must also be part of the
mapping (because of the location presupposition). Hence, there cannot be conceptual
incoherence between the landscape and the ego. In addition, temporal milestones such
as the twenty-first century are conventionalized and shared by our culture(s). In other
words, when the temporal milestone ‘the twenty-first century’ is used, it assumes a
generic perceiver. Conceptual incoherence arises when the ego and the temporal
landscape entail different perspectives. This accounts for the unacceptable individual
subject with achievement events marked by generic endpoints (i.e., (17a & b)).
Our account predicts that when a specific interpretation of a typically general
temporal landscape is available, then an individual is allowed. Such an example is
found in both English and Chinese as supporting evidence of our account. Given a
science fiction context of time-travel, the following sentence (in both English and
Chinese) is perfectly acceptable.
Kathleen Ahrens & Chu-Ren Huang
506
(22) Wo/Zhangsan jijiang jinru shiwu shiji.
I/John will enter fifteen century
‘I/John will be entering the fifteenth century.’
Note that the time-travel scenario alters the temporal landscape, because the fifteenth
century (or other absolute time points) are now observed and defined for this particular
individual.
This analysis also accounts for the collocation of activity verbs (typically ‘pass’ in
English or guo ‘to pass’ in Chinese) with individual subjects. Notice that in these
sentences the timescape is defined relative to the ego and the sentence focuses on the
manner of the motion. For example, in (23) the definite article indicates a definite and
specific timescape linked to the subject:
(23) He passed the time happily.
Moreover, the sentence is incomplete without the adjunct. One can substitute other
adjuncts such ‘in a drunken stupor’ and ‘in limbo’, and so forth; but the adjunct can
never be omitted. Similarly, when a detailed description of the time is given, it is
usually anaphorically linked to the subject, as in (24a-b):
(24) a. S/he passed his/her time reading.
b. S/he passed his/her high school years dating underachievers.
Example (19b) and (19c) show that the same observations hold for Chinese (excluding
the definite article observation, because Chinese does not have a definite article). But,
as in English, the sentence is incomplete without the adjunct (19c).15
Moreover, the
description of time may be anaphorically linked to the subject (19a-b).
To sum up the discussion above, we propose that the TIME PASSING IS A
MOVING EGO metaphor be revised to TIME PASSING IS A MOVING POINT ON A
LANDSCAPE as in Figure 7.
15
There is a possible reading in certain contrastive contexts with stress on tade ‘his/her’.
TIME PASSING IS MOTION
507
( .
}
|
| \
0 0
• Time
<-|-|-|-|-|-|-|-|-|-|-|-|-|-|-|-|-|-|-|-|-|-|->
Past P Future
Figure 7: Time passing is a moving point on a landscape
The moving point in the source domain maps to the reference-time point in the target
domain. Given the presupposition of the location of the ego, this metaphor entails that
an ego moves with the time point. The landscape is a series of landmarks which are
interpreted as temporal eventive milestones over which the reference point and attached
ego move. This explanation has several advantages. First, the inconsistencies in the
mappings between the subcases of TIME IS MOTION disappear. Now both subcases
involve a moving entity. In the first case, the moving entity approaches an ego. In the
second case, the moving entity (or point) has the ego attached. Second, it can account
for why the ego is unable to stop as it moves forward in time. Since the ego is attached
to the reference point, and this point must continually move forward, the ego must
move forward with it.
3.3 A seemingly ambiguous interpretation
What is our basis for claiming that a point in time (i.e., ‘Christmas’) is moving
towards an ego-referent as in (12) above (repeated here below as (25) for convenience)
when in a straightforward spatial example like (26) ‘Taipei’ is not moving?16
(25) shengdanjie kuai dao le.
Christmas quick arrive ASP
‘Christmas is almost here.’
(26) Taibei kuai dao le.
Taipei quick arrive ASP
‘Taipei is coming up quickly.’
That is, why is (25) not treated as an instance of the TIME PASSING IS A MOVING
POINT ON A LANDSCAPE metaphor where an unspecified agent would be moving
16
Thanks to a reviewer for bringing up this point.
Kathleen Ahrens & Chu-Ren Huang
508
across a landscape that included particular points in time?
We chose instead to analyze (25) as an example of TIME PASSING IS A
MOVING ENTITY. Example (27) is an example in the spatial domain with a similar
surface grammatical structure, where ‘teacher’ is a moving entity:
(27) Laoshi kuai dao le.
Teacher quick arrive ASP
‘The teacher is almost here.’
We interpret (25) along the lines of (27), with ‘Christmas’ understood as a moving
entity. In order to explain our analysis, we focus on the conceptualization of the arrival
event.
In terms of conceptualization, an ‘arriving’ event involves a trajector with an
extended path of motion towards a landmark, with the profiling on the trajector entering
the vicinity of the relevant destination landmark (Langacker 1987:246). In (26), then,
‘Taipei’ cannot be the trajector because it does not involve a path of motion. It is
instead a landmark. The trajector in the case of (26) is the speaker. The
trajector/landmark distinction is clearly seen in (28) below, where either ‘Zhangsan’ or
‘the train’ is the profiled trajector moving towards the landmark of Taipei.
(28) Zhangsan/Huoche kuai dao Taibei le.
Zhangsan/train quick arrive Taipei ASP
‘Zhangsan/The train is approaching Taipei.’
Turning to the metaphorical instance, however, it is impossible to have profiled
trajector such as ‘Zhangsan’ or ‘Huoche’ as shown in (29):
(29) *Zhangsan/Huoche kuai dao shengdanjie le.
Zhangsan/train quick arrive Christmas ASP
‘Zhangsan/The train is approaching Christmas.’
As we discussed above, the TIME PASSING IS A MOVING POINT ON A
LANDSCAPE metaphor involves only a non-referential generic ego that is attached to a
reference point. It does not involve an agentive individual, such as Zhangsan, nor does
it involve a movable, profilable entity such as a train.
Thus, the seemingly ambiguous interpretation of (25) is no longer ambiguous once
it is shown that the trajector cannot be profiled or agentive. Instead, the only
linguistically available interpretation is the one in which TIME PASSING IS
TIME PASSING IS MOTION
509
UNDERSTOOD AS A MOVING ENTITY. Moreover, the fact that this question of
ambiguity can be answered also lends credence to our proposal, since in the original
discussion of the two subcases there would be no way to distinguish between the
ambiguous readings.
4. Discussion
We now turn to examining our proposal with other analyses and to comparing our
two subcases with their English counterparts.
4.1 TIME PASSING IS A MOVING ENTITY
In section 3.1, we proposed that TIME PASSING IS A MOVING ENTITY for
Mandarin. This metaphor entails time moving towards an ego that is fixed on the
landscape (representing the present). This ego faces towards the past.
However, just as in the English case, where ‘precede’ and ‘follow’ entail that the
ego would be facing in the ‘wrong’ direction (i.e., towards the past for English), there
are a few cases in Chinese where time is a moving entity and the observer is facing the
‘wrong’ direction (i.e., the front) as in (30), and schematized in Figure 8:
(30) a. zhan-wang qian cheng
look-out front route
‘look forwards to the future’
b. zhan-wang wei lai
look-up yet come
‘look forwards to the future’ (Yu 1998:100, ex. (53a-b))
Time
( .
}
[]
|
0
<------------------|-------------------->
Past P Future
Figure 8: Time is a moving entity, ego is oriented towards the future
We propose that TIME PASSING IS A MOVING ENTITY metaphor is universal in its
Kathleen Ahrens & Chu-Ren Huang
510
mapping of the entity (i.e., from a moving entity to time passing) and in the orientation
of the moving entity (i.e., towards the ego). However, the orientation of the ego is
language-specific with Chinese preferring an orientation towards the past, and English
preferring an orientation towards the future (Table 2).
Table 2: Universal mappings between source and target domains for TIME PASSING IS
A MOVING ENTITY (Chinese and English)
Source Domain Target Domain (Case 1)
moving entity time
orientation of moving entity towards ego
Preferred orientation of ego towards past (Chinese)
towards future (English)
In both languages, moreover, there can be exceptions to the preferred orientation. Thus,
from both a cross-linguistic and intra-linguistic point of view, the orientation of the ego
is not crucial to understanding the motivation underlying the TIME PASSING IS A
MOVING ENTITY metaphor. However, from a cross-linguistic and intra-linguistic
point of view, there is evidence to support the salient mappings of this metaphor;
namely that time passing is an entity that moves towards an ego.
Previous work on metaphors of time in Mandarin, such as C. Huang (1981) and
Alverson (1994), agrees with our assessment that the orientation of the ego is towards
the past.17
Indeed, these two authors do not mention the existence of examples such as
(13), where the ego is facing towards the future.
However, Yu (1998) argues that in Mandarin, as in English, the ego faces the future,
and time approaches the static ego from the future. Yu’s (1998) account involves an
idiosyncratic interpretation of the words qian ‘front’ and hou ‘back’ that are not
supported by the evidence. In order to argue against the past to the front orientation, he
postulates that time consists of a moving train with a series of cars, with the middle car
being the present and the cars following the middle car (i.e., the back cars) representing
the future, and the cars preceding the middle car (i.e., the front cars) representing the
past. Yu concludes that qian and hou “respectively refer to car 1 and car 5 relative to car
3, which are all conceived of as entities composing a long moving object—the train,
17
S. Huang (1977) also looked at TIME IS MOTION metaphor in Mandarin. However, he
conflated the two sub-cases into one metaphor. The Invariance Principle no longer allows such
conflation since it requires that “Metaphorical mappings preserve the cognitive topology (that
is, the image-schema structure) of the source domain, in a way consistent with the inherent
structure of the target domain” (Lakoff 1990).
TIME PASSING IS MOTION
511
which has an intrinsic front and back of its own (p.107).” Several problems occur as a
result of this explanation. First, bodily experience is rendered secondary to heuristical
experience, which is counterintuitive to the foundation of conceptual metaphor
research—that there is a bodily basis for metaphors. Second, the postulation of a train,
or a long moving object divided into sections is anachronistic. Third, the coherence of
the system breaks down because the point of reference is no longer the ego; it is the ego
watching a particular point on a moving object. Fourth, the explanation of the train will
not work for English, because the back cars will represent the past in English. That is,
the train must be turned around in order for his explanation to work in English, which is
similar to the parameter we have suggested for the orientation of the ego.18
In addition, from a historical point of view, data from the Archaic and Ancient
Corpus of Chinese suggests that the moving time metaphor with the static ego facing
the past is the traditional and primary conceptualization of time in Mandarin and the
only one that has been diachronically consistent for over two thousand years.
Furthermore, the time metaphors with the static ego facing the future are modern
conceptualizations of time that have only recently entered the language (i.e., example
(30a-b)). We also surmise that historical layering is involved in the English cases, as
‘precede’ is derived from the Latin praecedere [pre+cede (to go)] and ‘follow’ is
derived from Old English folgian ‘to follow.19
Latin is another language where time is
understood as a moving entity that approaches an ego that is facing towards the past
(Alverson 1994).
Thus, we postulate that the reason no previous explanation has been adequate is
because the historical layering of the metaphorical system has been ignored. We further
postulate that inconsistent conceptualizations within a particular metaphor system can
only exist if they are borrowed/created at different time periods. Under a historical
interpretation it becomes apparent the reason Yu (1998) argues for a future-facing ego in
Mandarin. He relies too heavily on a small subset of data taken from modern Mandarin
and uses generalizations from this small subset to try to re-interpret and unify a more
established generalization. Once data is taken from different time periods, the
dichotomy between the ancient and modern orientation of the ego can be explained.
4.2 TIME PASSING IS A MOVING POINT OVER A LANDSCAPE
metaphor
We postulated for the second subcase that passing time is being understood as a
18
Yu (1998) does not attempt to relate his train metaphor to the English data.
19
Webster’s Ninth New Collegiate Dictionary, 1987, Springfield, Merriam Webster Inc. Note
that the conceptual mappings of time have not been examined for Old English.
Kathleen Ahrens & Chu-Ren Huang
512
moving point with an attached ego over a landscape. Our proposal has the following
two advantages. First, the two subcases of TIME (PASSING) IS MOTION can have a
unified entailment. In both TIME PASSING IS A MOVING ENTITY and TIME
PASSING IS A MOVING POINT both involve the concept of movement. In the first
case, the moving entity is moving towards a fixed ego that is facing the past. In the
second case, a group ego is moving with a reference point over a landscape that is
marked with time points. Thus, the main concepts associated with time passing is a
moving entity’s relationship to an ego, or a moving entity’s relationship to the
landscape. This entity was initially analyzed as an ego, but our revised analysis
proposes that it is the present point in time which entails an attached ego. In both
cases the passage of time requires motion of an entity.
Second, conceptualizing TIME PASSING IS A MOVING POINT as the second
subcase has the advantage of unifying what Yu (1998) previously divided into two cases
(i.e., Cases 2 and 3). Case 3 for Yu (1998) is TIME AS BOTH A MOVING OBJECT
AND A STATIONARY LANDSCAPE. That is, he analyzes examples such as (21) as
part of Case 3. This analysis, however, has the disadvantage of having a conceptually
disparate source domain, which is to be avoided. In addition, the target domain is not
simply TIME. It involves, in fact, two separate notions: TIME PASSING is the
MOVING OBJECT, while POINTS IN TIME are part of the landscape. Once it is
recognized that the examples of the plural ego moving through time must move with the
present point in time, the two subcases can be viewed as one.
How does our analysis for the second subcase compare with English? First, we
note that Lakoff and Turner (1989) also argue that the cases where time moves with the
ego are a special subcase of their Case 2 (TIME IS MOTION OVER A LANDSCAPE).
Yu writes:
This is probably because this special case shares all the features of Case 2.
That is, there exists a time-landscape over which the Observer travels from
the past to the future. However, this case also has features that Case 2 does
not have. That is, there is also a time-object moving with the traveling
Observer in the same direction. This time-object determines the present time:
wherever it is the present time. (Yu 1998:125)
Our point for the Mandarin cases is that in all instances where an ego is moving
forward, it must be attached, or go with, the ‘time object’ or time point, and that this
time point reflects the present time. Is it possible, then, that the English cases can also
be re-analyzed as a time-object moving with the ego? Moreover, will it still involve a
plural ego when there is the concept of achievement involved? We think it does and
TIME PASSING IS MOTION
513
we shall sketch out our reasoning below to consider whether English has similar
restrictions on its subcase 2.
First, Lakoff (1994:57) gives ten examples of TIME IS MOTION OVER A
LANDSCAPE (repeated here in example (31), with letters added):
(31) a. There’s going to be trouble down the road.
b. He stayed there for ten years.
c. He stayed there for a long time.
d. His stay in Russia extended over many years.
e. He passed the time happily.
f. He arrived on time.
g. We’re coming up on Christmas.
h. We’re getting close to Christmas.
i. He’ll have his degree within 2 years.
j. I’ll be there in a minute.
Examples (g) and (h) follow the entailments that we have laid out for TIME IS A
MOVING POINT OVER A LANDSCAPE. The ego is plural, and must be attached to
the time point. In addition, both involve an achievement point (i.e., ‘Christmas’). This
point can be viewed more clearly with the examples in (32) and (33):
(32) *I’m coming up/getting close to Christmas.
Example (32) demonstrates that the ego cannot be a singular individual. Example (33a),
below, is an example of TIME PASSING IS A MOVING ENTITY and this entity is
moving towards an ego, while (33b) is an example of TIME PASSING IS MOVING
POINT OVER A LANDSCAPE.
(33) a. They wish Christmas would never come.
b. ??They wish they would never get to/come up on Christmas.
In (33a), it is possible for the speakers to wish the delay of Christmas. S/he does so by
expressing the hope that Christmas will not arrive (i.e., that point in time will stop
moving towards her/him.) But it is impossible for the speakers to conceptualize that
they will stop moving towards Christmas (33b). Thus, in English, as in Chinese, the
plural ego is attached to a time point that continually moves forward without stopping.
Example (31e) is an interesting case, because it is similar to the Mandarin example
of guo ‘to pass’ (cf. (19)). In addition, it has a similar mapping in that the ego does not
Kathleen Ahrens & Chu-Ren Huang
514
have to be a generic plural since there is no particular point where an achievement is
reached. English can also use other motion verbs, as in (34), but a particular manner
must be described as we discussed in Section 3.2:
(34) After the divorce, he moved through his days in a daze.
The other examples in (31), moreover, are not part of the TIME PASSING IS
MOTION metaphor. Example (a) can be viewed as the LIFE IS A JOURNEY
metaphor. Examples (b-d, f, i, j) are examples of TIME IS A LANDSCAPE. This
metaphor involves a piece of a landscape being mapped to a particular period of time in
a person’s life. But it does not involve movement in either the source or target domains.
It merely requires the notion of boundedness in the landscape.
This brings us back to the point that the source and target domains must be clearly
defined in order to understand the conceptual mappings that take place. Yu (1998) and
Chiu (1998) for example, start out with the concept that TIME IS UNDERSTOOD AS
SPACE and from that starting point, then discuss instances involving movement.20
Time is, in fact, understood as space, and both Yu (1998) and Chiu (1998) provide
analyses regarding the use of ‘up’ and ‘down’, ‘left’ and ‘right’ and their relationship to
understanding time. Moreover, time is also understood as a landscape, as we have just
discussed above. But the passage of time is a different metaphor, one involving
primarily motion. This motion may take place over a landscape (and it makes sense that
it would, since time itself is understood as a landscape), but the salient mapping in the
TIME PASSING IS MOTION cases is the presence of movement. The spatial notion, if
it exists, as it must in Case 2, is secondary to the concept of movement. Thus, we
propose that English and Chinese have similar mappings for subcase number two once
other conceptual mappings (such as TIME IS A LANDSCAPE) are recognized as not
being a part of the TIME PASSING IS MOTION metaphor.21
Lastly, a moving time conceptualization is preferred over a moving ego
conceptualization because the target domain is the passage of time not the passage of an
ego. Even if the two conceptualizations have the same descriptive predictions, the
moving time conceptualization should be the null hypothesis. However, we have also
20
Chiu (1998) argues that ‘time and self are moving’ and ‘time isn’t moving and self is moving’
is a better analysis of the data.
21
In addition, we would like to point out that the mapping is not TIME PASSING IS A
JOURNEY. First, the mapping is incomplete because the ego has no control over the journey.
S/he cannot stop or change direction as it can do in the source domain of journey. Also having
the mapping of time understood as a journey itself would be tautological because journeys
take time.
TIME PASSING IS MOTION
515
shown in this paper that a moving time conceptualization is more powerful because it
explains a wider range of data.
5. Conclusion
In sum, we have argued that the concept of time passing should be understood as
motion. This analysis can account for both the data in Mandarin and English, with the
only difference being that the orientation of the ego in the MOVING ENTITY metaphor
is parameterized depending on the language. We have revised the general metaphor as
in (35) below:
(35) The General Metaphor: TIME PASSING IS MOTION
Ontology: The passage of time is understood in terms of entities that move.
Presupposition: The ego is situated at the present time.
Special Case 1: Time passing is an object that moves towards an ego.
Special Case 2: Time passing is a point (that a plural ego is attached to)
that moves across a landscape.
The mappings between the source and target domains for these two cases are given
below in Table 3.
Table 3: Mappings between source and target domains for TIME PASSING IS MOTION
(English and Chinese)
Source Domain Target Domain (Case 1) Target Domain (Case 2)
moving entity Time-object Time-point with attached ego
orientation of moving entity Towards ego towards future
orientation of ego Towards future (English)
Towards past (Chinese)
towards future
Thus, in arguing for a universal conceptualization of time, TIME PASSING IS
MOTION qualifies for both English and Chinese. The only variation involves the
orientation of the ego in the first case, with Chinese preferring an orientation towards
the past, and English preferring an orientation towards the future.
In arriving at this conclusion, we have argued against Yu’s position that the TIME
IS MOTION metaphor has an ego facing the future as a universal conceptualization of
time, and demonstrated that the orientation is towards the past for Chinese and towards
the future for English. Thus, the orientation of the ego is a conceptual subsidiary of the
Kathleen Ahrens & Chu-Ren Huang
516
TIME PASSING IS MOTION metaphor that can be parameterized by language change
and variation.
We have also argued that the target domain of passing time should not be confused
with the target domain of time itself. In addition, the source domain must be a unified
and clear concept. When this fact is recognized, it becomes apparent that the
entailments of the MOVING POINT OVER A LANDSCAPE source domain are that
the ego is an attached entity that must move with the present time-point. Moreover, we
propose that these entailments hold for both English and Chinese.
Recent psycholinguistic work on the TIME IS MOTION metaphor in English has
demonstrated that the two cases in English occur with equal frequency (McGlone and
Harding 1998, Gentner and Imai 1992, and Boroditsky and Ramscar, in press).
However, the diachronic and synchronic data examined here for Mandarin suggest
that Case 1 is the primary conceptualization of time for Mandarin speakers.
Experiments are underway to test this prediction (Lai 2002).
Understanding time is not limited to motion, of course. Time is also understood as
space, as money, as a commodity, among other mappings. Moreover, we use these
different source domains in order to map different concepts about time. We suggest that
focusing on narrow source domains will facilitate the understanding of these mappings
(Ahrens 2001). We disagree with the idea that inconsistency is allowable within the
mappings from a single source domain to a single target domain as Lakoff (1994:58)
argues is necessary to understanding the TIME IS MOTION metaphor.
Of course, inconsistencies may exist among mappings from different source
domains to the same target domain because different aspects of the target domain need
to be mapped (Ahrens 1995). But inconsistencies should not exist within the same
source domain (barring diachronic considerations as we discussed above). Conceptual
metaphor analysis will only be able to contribute to a unified theory of knowledge (e.g.,
Wilson 1998) when it is recognized that inconsistencies in mappings from the same
source domain reflect a gap in understanding. These gaps can be filled, we suggest, by
narrowing the source and target domains under consideration so that there are clear
conceptual distinctions among them, and of course, by carrying out further data
collection and analysis.
TIME PASSING IS MOTION
517
References
Ahrens, Kathleen. 1995. Metaphorical paradoxes: A window on the conceptual system.
Proceedings from the 10th Pacific Asia Conference on Language Information and
Computation, 139-148. Hong Kong: City University of Hong Kong.
Ahrens, Kathleen. 2001. Metaphor mapping and knowledge representation: Principles
and processing II. Proceedings of the Symposium from Selected NSC Projects in
General Linguistics from 1998-2000. Taipei: National Taiwan University.
Alverson, Hoyt. 1994. Semantics and Experience: Universal Metaphors of Time English,
Mandarin, Hindi, and Sesotho. Baltimore: Johns Hopkins University Press.
Boroditsky, Lera, and Michael Ramscar. (In press). The roles of body and mind in
abstract thought and language. Psychological Science.
Clausner, Timothy, and William Croft. 1997. Productivity and schematicity in
metaphors. Cognitive Science 21.3:247-282.
Chiu, Hsiou-Hua. 1998. The Metaphor Time-As-Space in Mandarin. Chiayi: National
Chung Cheng University MA thesis.
Gentner, D., and M. Imai. 1992. Is the future always ahead? Evidence for
system-mappings in understanding space-time metaphors. Proceedings of the
Fourteenth Annual Conference of the Cognitive Science Society, 510-515.
Hillsdale, NJ: Lawrence Erlbaum Associates.
Huang, Chu-Ren. 1981. Time as flowing water: A discussion of the Chinese concepts of
time from the point of view of the ancient classics. Chung-Wai Literary Monthly
9.11:70-88. (In Chinese)
Huang, Shuanfan. 1977. Time and the semantics of LAI and QU. Proceedings of the
Symposium on Chinese Linguistics, 1977 Linguistic Institute of the Linguistic
Society of America, ed. by Robert Cheng, Ying-che Li, Ting-chi Tang, 53-66.
Taipei: Student Book Company.
Lakoff, George. 1990. The invariance hypothesis: Is abstract reason based on
image-schemas? Cognitive Linguistics 1:39-74.
Lakoff, George. 1993. The contemporary theory of metaphor. Metaphor and Thought
(2nd edition), ed. by Andrew Ortony, 202-251. Cambridge: Cambridge University
Press.
Lakoff, George. 1994. What is a conceptual system? The Nature and Ontogenesis of
Meaning, ed. by Willis F. Overton and David S. Palermo, 41-90. Hillsdale, NJ:
Lawrence Erlbaum Associates.
Lakoff, George, and Mark Johnson. 1980. Metaphors We Live By. Chicago: University
of Chicago Press.
Lakoff, George, and Mark Johnson. 1999. Philosophy in the Flesh: The Embodied Mind
Kathleen Ahrens & Chu-Ren Huang
518
and Its Challenge To Western Thought. New York: Basic Books.
Lakoff, George, and Mark Turner. 1989. More Than Cool Reason: A Field Guide to
Poetic Metaphor. Chicago: University of Chicago Press.
Lai, Tzuyin V. 2002. Tests of Time in Mandarin Chinese. Taipei: National Taiwan
University MA thesis.
Langacker, Ronald W. 1987. Foundations of Cognitive Grammar, Vol. 1. Stanford:
Stanford University Press.
McGlone, M.S., and J.L Harding. 1998. Back (or forward?) to the future: The role of
perspective in temporal language comprehension. Journal of Experimental
Psychology: Learning, Memory, and Cognition 24.5:1211-1223.
Shinohara, Kazuko. 1999a. Constraints of motion verbs in the TIME IS MOTION
metaphor. Paper presented at the 25th Annual Meeting of the Berkeley Linguistics
Society.
Shinohara, Kazuko. 1999b. Epistemology of Space and Time: Analysis of Conceptual
Metaphors in English and Japanese. Tokyo: Kwansei Gakuin University Press.
Tai, James H-Y. 1989. Toward a cognition-based functional grammar of Chinese.
Functionalism and Chinese Grammar, ed. by H-Y. Tai and Frank Hsueh, 187-226.
South Orange: Chinese Language Teachers Association.
Wilson, Edward. 1998. Consilience: The Unity of Knowledge. New York: Alfred A.
Knopf.
Yu, Ning. 1998. The Contemporary Theory of Metaphor: A Perspective from Chinese.
Amsterdam: John Benjamins Publishing Company.
[Received 9 May 2001; revised 4 November 2001; accepted 15 November 2001]
Kathleen Ahrens
Graduate Institute of Linguistics
National Taiwan University
1, Sec. 4, Roosevelt Road
Taipei 106, Taiwan
ahrens66@hotmail.com
Chu-Ren Huang
Institute of Linguistics, Preparatory Office
Academia Sinica
130, Sec. 2, Academia Road
Nankang, Taipei 115, Taiwan
churen@gate.sinica.edu.tw
TIME PASSING IS MOTION
519
時間的經過是一種運動
安可思 黃居仁
國立台灣大學 中央研究院
在此篇文章中我們將檢驗 “時間是空間” 這個隱喻。余 (1998) 在中英文
語料的基礎上主張這是各語言通用的隱喻。我們研究側重於隱喻的一個特定
層面：即 “時間的經過是一種運動”。我們主張來源領域中的 “運動” 及空間
兩個領域必須清楚區分。這個區分可以解決過去 “時間是運動” 隱喻的傳統
分析在中文和英文中不可避免的矛盾現象，我們同時也指出在 “時間是內風
景中的動點” 這個特例中，說話者自我必須隨動點移動，且若動詞為完成位
移動詞時，自我必須是複數，而動詞為動作動詞時則無此限制。最後，在 “時
間是動體” 的特例中，自我的面向是一個從屬的概念，在不同語言中和語言
變異時會有所不同。
關鍵詞：時間，概念隱喻，語言隱喻，自我