3.1. deictic expressions

The observation that Moving Ego and Ego-centered Moving Time expressions are canonically deictic plays a central role in this paper. A discussion of temporal deixis will lay the groundwork for further discussion. One temporal issue people care about is the question of when things happen. The commonness of tense systems in languages attests to the importance of this question. Tense systems are essentially deictic. Deixis (adjective: deictic) is defined as a built-in dependence of a morpheme, word, or phrase on its situation of utterance for its meaning and particularly for its reference. For example, now is a deictic word in that it refers to the time when it is uttered. This subsection explains how Moving Ego and Ego-centered Moving Time expressions are canonically deictic. The next subsection will examine a contrasting kind of expression that is not deictic.

The structure of Moving Ego and Ego-centered Moving Time expressions is such that the simplest way to understand them is as deictic expressions. For example, (8a–8b) mean that Christmas is expected soon relative to the time of utterance. The reader will of course imagine the examples in a context in which they would be felicitous. It is in such a context that they are claimed to be deictic.

1. (8)

   1. a. We are approaching Christmas.

   2. b. Christmas is approaching.

In order to understand past-tense versions of these sentences, as in (9), it is necessary to imagine someone in the past relative to whom Christmas was imminent. This illustrates the claim that Moving Ego and Ego-centered Moving Time sentences canonically express a relationship between a time (such as Christmas), and the ‘Now’ of a real or imagined person (‘Ego’).

1. (9)

   1. a. We were approaching Christmas.

   2. b. Christmas was coming.

The sentences in (9) are examples of what Karl Bühler (1990 [1934], p. 140) has called Deixis am Phantasma or “imagination-oriented deixis”. Such sentences are not interpreted relative to the ‘here and now’ of their utterance. Rather, they are understood relative to an imaginary ‘here and now’, but the same principles of interpretation apply whether the situation of utterance is real or imagined. This phenomenon of imagination-oriented temporal deixis has been called transferring a temporal deictic element to ‘another center’ (Fillmore, Reference Fillmore, Jarvella and Klein1982, pp. 38–39) or shifted deictic center, and transposed indexical ground or decentering (Hanks, Reference Hanks1990; Haviland, Reference Haviland, Gumperz and Levinson1996). Technically, decentering consists of deictic words indexing (presupposing) something other than the ‘here-and-now’ of the utterance, which is what they canonically index (Hanks, Reference Hanks1990, pp. 208ff.), or indexing a ‘here-and-now’ that is altered in imagination.

Fillmore (1982, p. 38) illustrates the phenomenon of shifted deictic center (i.e., imagination-oriented deixis) with the sentences in (10).

1. (10)

   1. a. Several years ago, he lived near the beach.

   2. b. Several years ago, he had lived near the beach.

In (10a), since ago is a deictic word, the time when ‘he’ lived near the beach is understood to be several years previous to the time of utterance. Example (10a) does not involve decentering. In (10b), because of the pluperfect he had lived near the beach, the understanding of the sentence involves imagination-oriented deixis. That is, the time at which he lived near the beach is understood to be several years previous to some imagined deictic center (i.e., an imagined ‘here and now’), which in actual discourse would be supplied by context.

The claim that Moving Ego and Ego-centered Moving Time are canonically deictic means that there are good deictic examples of expressions that employ these metaphors, and also a range of examples on a continuum from deictic to non-deictic. Treating the deictic uses as basic allows us to make sense of the range of data.

Near the deictic end of the continuum, we have examples in which ego’s experience of the present is imagined in detail. Near the other end, few aspects of deixis remain, but we can still understand this type of sentence as being semantically related to the deictic ones. The continuum is illustrated with Ego-centered Moving Time expressions in (11) (cf. Moore, Reference Moore2014a, p. 26).

1. (11)

   1. a. Summer has come. (The Ground is the moment of utterance.)

   2. b. When summer comes, you’ll know it. (Ground = a future Time with a specified experiencer. This Time is specific in the sense that it is roughly determinable relative to a calendar.)

   3. c. The end of the world is bound to come eventually. (Ground = a future Time with an unspecified experiencer. This Time does not have any specific relation to an ordinary, e.g., Gregorian, calendar.)

   4. d. A future time is one that hasn’t come yet. (Ground = any imagined present with any imagined experiencer, on the condition that sentence (d) is understood as a definition of a future time.)

Thus, although a sentence like (11d) does not depend on its time of utterance for its meaning, its analysis requires deictic principles. That is, in order for the sentence to mean what it does, there has to be some Ego who is experiencing a present moment relative to which a future Time has not yet occurred.

The process of imagination-oriented deixis, or decentering as I shall call it, can proceed to the point that an utterance may hardly seem deictic at all. This can be illustrated with an example of a conventional spatial use of come. ^{Footnote 4} At this point the deictic semantics of come have become quite subjectified (in the sense of Langacker, Reference Langacker2008, p. 537).

1. (12) If you keep on walking, you will eventually come to a bathroom.

In (12), the Ground of come is the unspecified Location of the bathroom. All that seems to remain of the deictic semantics of come is the idea of an expected endpoint or Goal. The highly subjectified spatial deictic semantics of come in (12) are similar to the temporal semantics of come in (13).

1. (13) Tuesday comes after Monday.

This use of come prompts the conceptualizer to imagine an experiencer (Ego) at whose Location the weekdays metaphorically arrive in order (cf. Dewell, Reference Dewell2007). The conceptualizer is defined as whoever is understanding the sentence or utterance in question (cf. Langacker, Reference Langacker1987).

Example (13) gives a rough idea of where the limits of the deictic continuum are, and suggests the possibility that come in (13) evokes in the conceptualizer a notion of metaphorical movement relative to Ego. In other words, there may be two frames of reference involved in (13), one evoked by come and another evoked by after. The semantic involvement of Ego in (13), if it exists, would be part of the background relative to which the after relation between Tuesday and Monday in (13) is understood.

This discussion has brought out the contrast between the terms conceptualizer and Ego. Whereas the conceptualizer is whoever is understanding the sentence, Ego functions as part of the semantics that I am analyzing. Ego is the (possibly fictive) person that the concepts ‘here’ and ‘now’ presuppose. The conceptualizer is a real person who is using language. Claims that I make about what goes on in the conceptualizer’s mind are claims about what I think is logically necessary for a linguistic form (word or phrase) to mean what it does. It goes without saying that psychological methods are also necessary to support hypotheses about what goes on in people’s minds.

It is of course possible (and maybe typical) for the conceptualizer and the role of Ego to have the same viewpoint, as in a case where I might say Ramadan is coming, and I expect Ramadan soon. But we have seen that the conceptualizer and the role of Ego do not necessarily have the same viewpoint, as in a case where I might say or read ‘Ramadan was coming’. Returning to the main thread of this subsection, we have seen that Moving Ego and Ego-centered Moving Time expressions are canonically deictic, and that in order to appreciate this observation, it is necessary to recognize the phenomenon of decentering (i.e., imagination-oriented deixis).

3.2. deictic neutrality

The canonically deictic nature of Moving Ego and Ego-centered Moving Time expressions contrasts with the nature of expressions that instantiate sequence is relative position on a path (Also called ‘front/back Moving Time’), such as those in (14).

1. (14)

   1. a. They hoped that the bombing would weaken law enforcement morale ahead of a full-scale uprising against the government. [i.e., before an uprising. KPFA radio 30 March 2010.]

   2. b. It is interesting to note that a Dellinger fade out is often followed by a magnetic field change around 48 hours later. [BNC A19 573]

In the source frame of sequence is relative position on a path (sequence is position for short), two or more entities move in the same direction on a Path such that one is ahead of another. In the metaphor, the entity that is ahead maps onto an earlier Time and the one that is behind maps onto a later Time. The spatial source frame of ordered motion and the counterpart temporal target frame of succession each apply to a potentially infinite set of entities. The name ‘sequence is relative position on a path’ follows the convention in metaphor theory of naming metaphors with the convention ‘source-frame is/are target-frame’, for example states are Locations .

Sequence is position has generally been analyzed as a submapping of Ego-centered Moving Time, and indeed there is a level of generality at which the two share properties (see Moore, Reference Moore2014b). The reason for analyzing the two as distinct is that while instantiations of Ego-centered Moving Time are canonically deictic, instantiations of sequence is position have no tendency to be deictic, although they are in principle perfectly compatible with deixis. The property of neither favoring nor disfavoring deixis is termed deictic neutrality.

The claim is that instantiations of the metaphor sequence is position, such as those in (14), do not require the conceptualizer to include ‘Ego’s Now’ in the temporal relationship that is depicted by the metaphor. In other words, the concepts Past, Present, and Future are not relevant to sequence is position. A simple demonstration can help make this point. As we have seen, a shift in tense with Ego-centered Moving Time requires the conceptualizer to construct an imaginary deictic field to understand the metaphorical motion relative to Ego. With sequence is position, by contrast, a shift in tense as in (15) induces the conceptualizer to take a different perspective based on the tense, but it does not induce any modification involving the metaphor structure (i.e., Figure–Ground relations).

1. (15)

   1. a. New Year’s followed Christmas.

   2. b. New Year’s follows Christmas.

   3. c. New Year’s will follow Christmas.

This perspectival contrast between sequence is relative position on a path (in 14–15) on the one hand, and Moving-Ego/Ego-centered Moving-Time (in 8–9) on the other, can be analyzed as a contrast in the kind of frame of reference the conceptualizer uses.

3.3. frames of reference

A spatial frame of reference is a strategy for using a Ground to say where a Figure is located. A number of recent works treat the topic of space-derived temporal frames of reference, including Bender and Beller (2014, a review of the relevant research), Bender, Rothe-Wulf, Hüther, and Beller (Reference Bender, Rothe-Wulf, Hüther and Beller2012), Evans (Reference Evans2013), Moore (Reference Moore, Achard and Kemmer2004, Reference Moore2014a) Núñez and Cooperrider (Reference Núñez and Cooperrider2013), Tenbrink (Reference Tenbrink2011), and Zinken (Reference Zinken, Evans and Chilton2010). Unfortunately we will not be able to discuss this wide variety of analyses here.

I propose two frames of reference for analyzing sequence is position and Moving-Ego/Ego-centered Moving Time. These frames of reference have to do with how it is possible to use the relevant spatial notions for temporal determination. They do not fully account for the temporal determination, because there are factors involved that are not necessarily directly tied to the spatial strategies that this paper analyzes; examples of such factors are clocks and calendars.

My idea of a spatial frame of reference is the same as that in Levinson’s (Reference Levinson2003) cross-cultural taxonomy of intrinsic, relative, and absolute. The idea that spatial frames of reference should be relevant to temporal metaphor makes sense because temporal metaphor is a matter of mapping spatial relations onto temporal relations.

Since motion involves time, motion metaphors of time are not strictly mappings from space to time. They are mappings from certain frames that involve space and time to other frames that involve time but not space. In a given source frame, spatial and temporal concepts come together in particular ways that make the frame useful for organizing temporal concepts. The fact that the source frames of temporal metaphor are specialized for temporal concepts means that the taxonomy of frames of reference that are relevant to temporal metaphor will not necessarily be the same as the taxonomy of frames of reference that are relevant to spatial concepts in general.

3.4. summary and conclusions for ‘Frames of reference and imagination-oriented deixis’

Motion metaphors of time depict temporal phenomena in terms of source-frame motion scenarios. The construals of these motion scenarios presuppose frames of reference that structure how the metaphors depict temporal concepts. I have proposed two contrasting frames of reference: ego-perspective and field-based, each of which structures semantic relationships among entities on a Path. These frames of reference are based on particular motion events that involve the temporal concepts depicted by the metaphors. The field-based frame of reference is based on a scenario in which two Movers are ordered in an ahead/following relation. The ego-perspective frame of reference is based on a scenario in which a Mover approaches, arrives at, and passes a Location. (In later sections we will consider mappings that involve space without motion.)

The ego-perspective frame of reference can be decentered/transposed. That is, the elements of the frame of reference can be instantiated by conceptual content that is not present in the ‘here and now’, or the situation of utterance can be otherwise altered in imagination. ^{Footnote 7} An analysis of temporal metaphor must take decentering into account because a decentered deictic utterance may be used in a way that does not actually depend on the situation of utterance but still depends on deictic principles, as in (11d) above (A future time is one that hasn’t come yet). Now that we have a clear view of some of the functions of motion metaphors of time, we are ready to have a closer look at some of the spatial and temporal concepts involved.

4.1. experiential motivations of the ego-perspective metaphors

One way of approaching experiential motivations is to talk about grounding scenarios. These are scenarios in which source-frame and target-frame concepts correlate saliently, similar to Grady and Johnson’s (Reference Grady, Johnson, Dirven and Pörings2002) primary scenes. ^{Footnote 8} The idea of a grounding scenario or a primary scene is that the scenario/scene highlights the source-frame and target-frame concepts of a given metaphor in a single experience of a type that is recurrent in everyday life. Studying grounding scenarios – which include more structure than is actually mapped – is a way of exploring the spatial and temporal concepts that are involved in the metaphors. In Sections 6 and 9, we will discuss primary metaphors, which arise from primary scenes. Now let us look at the grounding scenario for Moving Ego.

I use occur as a technical term for talking about Times as events. To say that a Time occurs is equivalent to saying that it ‘comes’. Here are some sample sentences that could be said in the context of the Moving Ego grounding scenario.

1. (20)

   1. a. The station is ahead.

   2. b. We are approaching the station.

   3. c. We have arrived at the station.

   4. d. We have passed the station.

In the grounding scenario for Moving Ego, we see analogues to all of the temporal concepts that the metaphor depicts. For example, corresponding to the Past being metaphorically behind Ego, we have Ego remembering Locations behind her where she has been. Ego’s expectations of her future arrival at a Location corresponds to the Future being metaphorically ahead of her. Now let us look at the grounding scenario for Ego-centered Moving Time.

Here are some sample sentences that could be said in the context of the grounding scenario.

1. (21)

   1. a. The bus is coming.

   2. b. The bus has arrived.

   3. c. The bus is going by

   4. d. The bus has gone by.

Parallel to the case of Moving Ego, the grounding scenario for Ego-centered Moving Time has analogues to all of the temporal concepts that the metaphor depicts.

4.2. earlier-toward-later metaphorical motion not involving Ego

In addition to the earlier-toward-later metaphorical forward motion that is associated with Ego, this subsection examines such motion in the absence of Ego. The observation in Section 4.1 about the correlation between Ego’s forward motion and the present moment getting later also applies as a default to any entity that moves relative to some background (Kranjec, Reference Kranjec2006). That is, for any moving entity, a region that is more advanced on its Path correlates in experience with a later Time. And this kind of motion can be thought of as forward motion unless there is some motivation to the contrary. This idea is illustrated in (24). The context is that a small ball is rolled along a horizontal track in an experiment.

1. (24) There is a scale on the track so we can measure the forward motion of the ball.

Consistent with the observation that the (forward) motion of any moving entity correlates with Now getting later, we have three metaphors in addition to Moving Ego in which a moving entity maps onto a Time and an entity that is more advanced on the Path maps onto a later Time.

• Now is a Mover as in It’s getting close to midnight. [Cf. Lakoff & Turner, Reference Lakoff and Turner1989, pp. 44ff.]

This metaphor seems to be similar to Moving Ego except that the present moment is characterized with the word it as an aspect of the setting of the utterance (Moore, Reference Moore2014a, p. 43).

• Time is a Mover as in Time marches on, or Time keeps moving forward.

Expressions of this metaphor can be paraphrased as ‘The present moment always gets later’, so it may be possible to analyze time is a Mover as a decentered perspectival metaphor. This is speculative, but it may be possible to analyze the Ground relative to which ‘time’ moves as the previous Locations occupied by the Mover. This is consistent with the idea that English speakers are comfortable with the metaphorical construal of ‘Now getting later’ as forward motion.

If we assume that this construction refers to the present moment in general rather than the Now of the speech act, it seems that the subject slot is restricted to the words time and hour, and perhaps other words that refer to the ‘matrix’ sense of time (cf. Evans, Reference Evans2003, chapter 11). Note that Time is a Mover is not the same as a Time is a Mover, the latter being entailed by Moving Time metaphors (Ego-centered Moving Time and sequence is relative position on a path).

• A situation is a Mover

This one is different from the other ‘forward motion’ metaphors in that it profiles the internal constituency of a situation, and does not necessarily designate when the situation occurs. Moore, Reference Moore2014a, p. 44). ^{Footnote 11} That is, the metaphor depicts the aspectual progress of a situation. For example, (25b) depicts the entire duration of a situation as occupying an extent rather than depicting a discrete Time as an entity that changes Location relative to a Ground. In fact, (25c), which is similar to (25b), could be analyzed as involving extent rather than motion, loosely analogous to (25d). While I am comfortable claiming that (25a) is a motion metaphor of time, we might want to analyze (25c) in some other way, noting that a starting time is marked with from and an ending time is marked with to.

1. (25)

   1. a. The rainy season went on for a long time without any rain.

   2. b. The party went from dawn to dusk.

   3. c. The beans boiled from dawn to dusk.

   4. d. The clothesline hung from the house to the shed. [But note that (d) is imperfective while the others are perfective.]

In addition to the correlation between forward motion and the current time getting later, another motivation for structures like (25a–25c) is that earlier-to-later is presumably the most reasonable order in which to scan a process, ^{Footnote 12} since that is the order in which processes are understood to evolve (Section 4.1.1). Example (26a) from Langacker (2008, p. 502) is evidence for an earlier-to-later bias because the order of mention of the opening and closing times iconically represents the actual order, and it would not normally be felicitous to reverse the order and say that the stores are open ‘between 10 and 7 in the evening’ to convey the meaning in (26a).

1. (26)

   1. a. In the evening, stores are open between 7 and 10.

Further suggestive evidence that English speakers presuppose an earlier-to-later order when talking about temporal phenomena comes from the temporal meaning of next as in (26b) (cf. Langacker, Reference Langacker, Pecher and Zwaan2005).

1. (26)

   1. b. You milk the goats and next you milk the cows.

   2. c. The well is next to the barn.

   3. d. She took off her shoes and started smoking.

Assuming that the temporal meaning of next comes from a meaning of proximity as in (26c), the reason it is used to talk about a later rather than earlier event would be that an earlier-to-later organization is assumed. This organization is also assumed in the iconic word order of (26b) and of (26d), assuming that (26d) means that she started smoking after she took off her shoes. The iconic word order is a material anchor (Hutchins, Reference Hutchins2005) for the conceived earlier-to-later organization of events.

Examples like (27) are consistent with the claim that people scan from earlier toward later, and that this direction is talked about as forward. In (27) a time period is being scanned rather than a process.

1. (27) Many physicians from that time forward were of the opinion that nearly all the late complications of syphilis were, in fact, the result of mercury poisoning [BNC ARH 142]. (Ground = that time.) [Moore, Reference Moore2014a, p. 125]

To summarize, this subsection has treated cases in which the direction of scanning or a metaphorical Mover other than Ego goes from earlier toward later. We see that even when Ego is not in the conceptualization, English speakers can talk about earlier-toward-later metaphorical motion as forward motion.

4.3. the grounding scenario for ‘sequence is relative position on a path’

This subsection discusses a grounding scenario that is strikingly different from the ones that motivate Moving Ego and Ego-centered Moving Time.

Minimally, this scenario involves two physical entities on a Path, and an event of arrival to motivate the mapping from an entity that is ahead to a time that is first. Fully elaborated, the scenario involves a sequence of two events of arrival. The sentence in (28) could be used in context to refer to an instance of the grounding scenario.

1. (28) Example sentence for the grounding scenario of sequence is relative position on a path

2. Pat [Figure] got to the office ahead of Kim [Ground].

In (28), when the first arrival occurs, the Figure (Pat) is ahead of the Ground (Kim), and this position ahead stands metonymically for the temporal relation ‘first’ which holds between Pat’s arrival and Kim’s presumed subsequent arrival. For further discussion, see Moore (2014a, Sections 6.2.2, 6.3, and 9.2). An example of sequence is relative position on a path is given in (29) for comparison with the non-metaphorical example in (28).

1. (29) There was fierce campaigning [Figure] ahead of the elections [Ground].

The Figure–Ground structure of the source frame of sequence is position requires some discussion because although sequence is position is a motion metaphor of time, the Figure does not necessarily move relative to the Ground. Movement is involved at a secondary level of Figure–Ground organization (Talmy, Reference Talmy2000, p. 336). This can be explained using an example sentence with follow.

1. (30) Fido [Figure] followed Gretchen [Ground] along the road.

Fido is analyzed as the primary Figure because it is his Location which is determined via the predicator follow relative to Gretchen, who accordingly serves as primary Ground. Next, there is a secondary level of meaning, since the ‘follow’ relation says that both entities in the relation are moving. Thus, Fido and Gretchen, considered together, constitute a secondary Figure which is moving relative to the road. Because they are moving in the same direction, the Figure and Ground maintain the same go-ahead/follow relation to each other regardless of the direction of motion relative to the secondary Ground or relative to the conceptualizer.

To summarize, in the grounding scenarios of the ego-perspective metaphors, there is one pertinent level of Figure–Ground organization, which has to do with relative motion between Figure and Ground. In the grounding scenario of sequence is position there are two levels of Figure–Ground organization: the primary Figure and primary Ground do not move relative to each other, even though as a unit (composite secondary Figure) they move relative to a secondary Ground.

4.4. past, future, and the psychological present

The ego-perspective grounding scenarios involve oppositions between past and present, and between present and future. This is explicit where the scenarios involve memory and expectation. The field-based grounding scenario (i.e., the one for sequence is position), by contrast, does not involve the concepts ‘past’ or ‘future’. In fact, the relation of succession cannot be analyzed in past–present–future terms. This is because successive events can happen during the present moment.

A possible alternative analysis would have the present as a ‘point’ that separates past and future. But the present as people experience it has duration, if we define the present as the time during which a person perceives, as opposed to times which are remembered or expected. Paul Fraisse demonstrates this point with the example of rhythm. When people experience rhythm, there are successive beats but they are perceived together. Fraisse’s (1963, p. 89) experimental subjects lost the sensation of rhythm when the interval between beats was increased to about 2 seconds. This is because the maximum duration of the psychological present for Fraisse’s subjects in that experiment was less than about 2 seconds.

In the sequence is position grounding scenario, one can imagine successive arrivals well within the limits of the psychological present. In such a scenario, one could perceive the go-ahead/follow relation and the relation of succession together simultaneously as different aspects of a gestalt. For this reason, there is not necessarily any expectation or memory involved in the grounding scenario of sequence is position.

To summarize, in the ego-perspective grounding scenarios, the movement of a Figure relative to a Ground correlates with Ego’s experiences of remembering the Past, perceiving the Present, and expecting the Future. In the field-based grounding scenario there are two successive events that are not analyzable in terms of Past–Present–Future because the events can happen together during the Present. This is true of any present moment. While Past, Present, and Future are constantly changing because the Present is constantly getting later, the temporal relationship between the successive events in the field-based grounding scenario does not change. Ego-perspective metaphors but not sequence is position depict experience of the Present.

5.1. the tendency of unmarked coding of ‘in-front/behind’

Given that a motion scenario is a plausible motivation for the earlier=in-front/later=behind pattern of temporal reference, it is interesting that there is evidence for a cross-linguistic tendency in which this pattern appears with deictically neutral expressions, among languages that use in-front or behind to talk about temporal concepts. This pattern can be seen in examples (31–32) above, which illustrate how Japanese and English conform to the tendency (Moore, Reference Moore2006, Reference Moore2014a, Section 11.2). The tendency is stated here:

Further illustrating the idea of the tendency with the words before and after, the hypothesis is that the words before and after have no tendency to be used deictically in their ‘earlier’ and ‘later’ meanings (respectively) because the motivating metaphor is neutral regarding perspective. Moreover, since the motivation is perspectivally neutral, the hypothesis does not offer any reason why before or after should not be used deictically. ^{Footnote 13} (This motivation may be only historical in the case of before/after, but is current in languages like Japanese that synchronically have the polysemy pattern.) Example (35a) illustrates a deictic use of before, and (35b) illustrates a non-deictic use.

1. (35)

   1. a. Before, I used to drink wine. (sequence is position, deictic)

   2. b. I wash my hands before I eat. (sequence is position, not deictic)

Example (35c) illustrates a use of before in which ‘later’ (future) is in front of Ego, presumably instantiating the Moving Ego metaphor. This illustrates the corollary of the tendency of unmarked coding, which is that later=in-front and earlier=behind expressions should tend to occur with deictic anchoring. (Note that the tendency does not make predictions about whether or not instances of the earlier=in-front/later=behind pattern will be deictic.)

1. (35)

   1. c. She has a wonderful future before her. (Moving Ego)

The ‘future’ meaning of before in (35c) is presumably due to the evocation of Ego’s bodily orientation via the personal pronoun her as complement of before. According to this account, Ego’s bodily orientation is not evoked in (35a) because the unstated complement of before in (35a) is understood to be the present time rather than ego. Consistent with the theory of Levinson (Reference Levinson2003), the frame of reference is not determined according to whether the Ground is associated with Ego. Rather than depending on the nature of the Ground itself, the relevant characteristics of the path-configured field-based frame of reference, stated for the source frame, are these:

1. i. The only criterion for determining the status of the Figure relative to the Ground is whether it is more or less advanced on the Path.

2. ii. The frame of reference determines an unchanging relationship between Figure and Ground: Their relative positions do not change, and the Figure does not get closer to or farther from the Ground. [Moore, 2104a, pp. 135–136, 139]

In the informal checking that I have done, languages that use deictically neutral front or behind terms for sequence exhibit the tendency. For example, the data in Haspelmath (Reference Haspelmath1997), a cross-linguistic survey of spatial expressions that are used to talk about time, mostly conform to the tendency. See Moore (2014a, chapter 11) for discussion. It is interesting that the earlier=in-front/later=behind pattern in deictically neutral or non-deictic expression types is found in many typologically different languages in different families in different parts of the world, from Yucatec Maya (Bohnemeyer, Reference Bohnemeyer1998, p. 257; Le Guen & Pool Balam, Reference Le Guen and Pool Balam2012, p. 7 ^{Footnote 14} ) to Aymara (Núñez & Sweetser, Reference Núñez and Sweetser2006) to Finnish (Huumo, Reference Huumo2015) to Mandarin Chinese (Yu, Reference Yu2012). The widespread occurrence of the pattern suggests that it is motivated by human experience.

6.1. location

The main way that the concept of location is relevant to temporal metaphor is that co-location (being at the same place) maps onto simultaneity; this is a primary metaphor: simultaneity is co-location. We have seen this mapping instantiated more specifically as Now is Here, which is a primary submapping of both Moving Ego and Ego-centered Moving Time. Now is here is exemplified in (36a) and (36b). Example (36c) shows the other primary variant of simultaneity is co-location: Times are Locations.

1. (36)

   1. a. Summer is here.

   2. b. Summer has arrived.

   3. c. We are in summer.

Summer is analyzed as the Figure in (36a) because it is the entity whose Location is at issue. Also, I interpret (36a) as analogous to (36b), in which Summer is figural because it is depicted as a Mover that is located at an unstated Ground, which is a Location that maps onto the moment of the speech event. Sentence (36c) is an example of Times are Locations because it takes Now as the Figure (expressed as we) and metaphorically locates it in summer, where summer is the Ground; i.e., (36c) tells us that the present time is simultaneous with part of summer. Sentence (37) is an example of Times are Locations in which Ego is not involved. In the example, the event of snowing is metaphorically located in May.

1. (37) It snowed in the middle of May.

Consistent with the logic of simultaneity is co-location (which is a statement of a mapping that generalizes over Now is Here and Times are Locations), entities that map onto Times and are not co-located with each other map onto different Times. For example, the distal Christmas in Christmas is coming is distinct from Now, which is the counterpart of the Location of the speech act. This inference pattern also applies to sequence is position, which does not depict simultaneity, since an essential feature of its structure is that Figure and Ground are not in the same position. This contrast between the ego-perspective and field-based metaphors is evident in the contrasting types of predicators involved. The ego-perspective metaphors may use predicators that involve co-location, such as arrive, but sequence is position does not use such predicators.

To summarize, there is a primary mapping from Locations to Times which is a component of both Moving Ego and Ego-centered Moving Time, and also exists independently. This primary metaphor yields predictable inferences.

6.2. motion

6.2.2. A static relation between primary Figure and primary Ground

In sequence is relative position on a path, the Figure is a Mover, and the primary Ground is also a Mover. The Figure and (primary) Ground are moving together, not relative to each other. The predicators in sequence is position expressions are well suited to expressing where an entity is located relative to another in an unchanging configuration. Such predicators include, for example, follow, be ahead, be in front. The mapping for sequence is position is in Table 3.

table 3. sequence is relative position on a path. Example: A period of calm followed the storm

Sequence is position expressions do more than tell us that two Times are not simultaneous; they also tell us which one was earlier or later. To the extent that they do this by presupposing direction on a Path, the information is not purely spatial, since the idea of a Path involves motion. However, in sequence is position the motion is not prominent. Rather, the motion is a feature of the background frame (or base; Langacker, Reference Langacker1987) relative to which the primary Figure and Ground are conceptualized. Although sequence is position does not profile change, this backgrounded motion corresponds to the understanding that Times are experienced as transient.

In addition to its temporal component, the notion of Path also has salient spatial structure. Specifically, the two ends of the Path differ from each other according to the direction of motion. Setting aside the notion of Path, direction in itself can be conceived of without motion or time. For example, my ‘line of sight’ gives me an instantaneous experience of direction. To summarize, we can isolate strictly spatial concepts in sequence is relative position on a path, but the full characterization of these concepts requires that we consider motion.

6.3. proximity

The concept of proximity functions in the metaphors in a way that is inferentially consistent with motion and location: if to be proximal is to be almost co-located, the metaphorical counterpart is to be almost simultaneous, as in (38b) Summer is approaching. Like location, proximity is a static idea. But while it makes sense to have a temporal mapping with static location as in (37) It snowed in the middle of May, proximity yields the right inferences only in a context of motion. For example, in The end of summer is near, we only get the inference that summer will end soon if metaphorical motion is involved.

6.4. the perceptive-interactive vs. derived ‘in-front/behind’ schemas

The later=in-front/earlier=behind pattern is motivated by Ego’s orientation and direction of motion, and also by her memory and expectations, as discussed in Section 4.1 on experiential motivation. The schema that gives rise to these characteristics can be called the perceptive-interactive in-front/behind schema (cf. Allan, Reference Allan1995). This schema divides the temporal universe into three regions: the Past behind Ego, the continuing Present at Ego’s changing Location, and the Future ahead of Ego. In (40a) we see an example in which the Time of an event is indicated by Ego’s bodily orientation, presumably motivated by the perceptive-interactive in-front/behind schema. To the extent that this schema is based on a moving Ego, it is dynamic. Example (40a) could also be structured by a static in-front/behind relation, as is also the case with (41). Whether examples like (40a) and (41) are understood as dynamic is an empirical question. (Cf. Lakoff & Johnson, Reference Lakoff and Johnson1999, p. 140 on the Time Orientation Metaphor.)

1. (40)

   1. a. I had an experience back in 2002 that alerted me that there was such a thing as vaccine injury. [KPFA radio 19 June 2015. Italics added.] (Moving Ego)

   2. b. The farther ahead we go, the more I understand.

Example (40b) shows that the ‘ahead’ direction is compatible with the expression of the Present as a continuing Time. In other words, the Present is given special treatment and depicted as having structure. Like (40b), example (41) expresses the Present as a locus of consciousness rather than just another Time.

1. (41) I’m looking forward to the match.

In contrast to the perceptive-interactive in-front/behind schema, we have also seen a mapping that uses a metaphorical in-front and behind derived from the relative positions of two entities that arrive in sequence (Section 5 on sequence is relative position on a path). We can call this the derived in-front/behind schema. This schema divides the temporal universe into two regions, ‘before’ and ‘after’, with no special treatment for Ego’s Now. The spatial relation that is mapped is essentially static, but with a background of motion, as discussed in Section 6.2.2 on sequence is position. The concepts of in-front and behind in this schema have to do with motion and location, and not with the front or back of any perceiving or interacting entity.

6.5. the significance of relative motion between primary Figure and primary Ground

Looking at the discussion so far, it will be noticed that relative motion between Figure and primary Ground occurs only in ego-perspective metaphors. This generalization is simple, but there is a type of apparent counter-example, represented by (23c) above (#As Christmas moves forward into the 21st century …), and by (42) below.

1. (42) Ramadan is getting closer to spring break every year. (This is because Ramadan is defined in a lunar calendar and spring break is defined in a solar calendar.)

Taken at face value, (42) has a Time moving relative to another Time. But (42) does not involve a mapping from ordinary motion to time. Rather, the metaphoric motion in (42) is analogous to pattern-path fictive motion (Talmy, Reference Talmy2000, p. 128) as in (43). ^{Footnote 17}

1. (43)

   1. a. As I painted the ceiling, paint spots slowly progressed across the floor.

   2. b. From noon to eight, the paint spots got closer to the radio every hour.

In (42), the Figure (Ramadan) changes Location relative to the Ground (spring break), but there is no Ego’s Now or decentered locus relative to which Ramadan is becoming more imminent. Thus (42) is not an instance of Ego-centered Moving Time. Instead, (42) is about successive occurrences of Ramadan as a temporal role instantiated on different days of the year in successive solar years (earlier each year); for example, Ramadan of 2014, Ramadan of 2015, etc. The most parsimonious analysis of (42) is that occurrences of Ramadan are construed as places via times are Locations (Section 6.1), and each of these places is metaphorically closer to spring break than the one in the preceding year (immediacy is proximity) (cf. Sweetser, Reference Sweetser1996, Reference Sweetser, Nuyts and Pederson1997). Thus the generalization holds that relative motion between primary Figure and primary Ground only occurs in ego perspective metaphors. This motion depicts Ego’s experience of time, as we will see in the following sections.

6.6. summary and conclusions to ‘spatial and temporal concepts in the metaphors’

We have identified one strictly spatial mapping: simultaneity is co-location. This mapping has immediacy is proximity as a kind of corollary, with the caveat that I do not know if people conceive of immediacy as proximity independently of motion. Additionally, we have identified sequence is relative position on a path as a metaphor that is primarily spatial but is structured by a background of motion or at least motivated by a motion scenario. The ego-perspective metaphors prominently involve motion, and the motion in turn highlights the spatial concepts of co-location, separation, and proximity.

We analyzed two different in-front/behind relations: a dynamic one in Moving Ego and a static one in sequence is position. These two in-front/behind relations are motivated in different scenarios that share virtually nothing except motion. All of the spatial concepts we have seen are involved to one degree or another in scenarios of motion. These scenarios range from the unbounded motion of Christmas is coming to the mere inference that motion has occurred, as in Summer is here.

7.1. the generic structure of the ego-perspective metaphors

The ego-perspective metaphors share two kinds of generic structure: perspectival structure and aspectual structure. This structure is present in the source and target frames of each metaphor. The perspectival structure is realized as the presence of Ego in all four frames. This is represented in Tables 1 and 2 and in Figures 2 and 3 as the word here or now. Additionally the words deictic center are included as a kind of shorthand to indicate that each frame is organized around Ego’s situation. This organization obtains at all stages of each frame.

Fig. 2. Ego-centered Moving Time as a conceptual integration network.

Fig. 3. Moving Ego as a conceptual integration network.

The generic aspectual structure consists of the four stages identified in Section 6.2.1 for the schematic motion event. The stages are indicated in the Figures by the numbers 1, 2, 3, and 4, corresponding to the order of the stages. This aspectual structure can be thought of as X-schema (Execution schema) structure in the sense of Chang, Gildea, and Narayanan (Reference Chang, Gildea, Narayanan, Gernsbacher and Derry1998). Each element of the generic aspectual structure maps onto the element with the corresponding number in source and target frames of the metaphor. The generic aspectual stages are given here, and the stages of the schematic motion event are repeated below so the reader can see how each aspectual stage corresponds to a stage of the motion event.

In Figures 2 and 3, inputs 1 and 2 are the respective source and target frames of the metaphor. The relationship between these two frames is the same as it is typically analyzed in the standard two-frame representations of conceptual metaphors, with an arrow leading from each source-frame element to its target-frame counterpart. The addition of the generic space says that each element of the source-to-target pair is an instance of a generic stage. For example, the ‘Distal Mover headed toward Ego’s Location’ and ‘A Future Time’ instantiate the initial state (Stage 1) of the generic aspectual structure. In Stage 2, the approaching Mover in the source frame and the Time that is becoming more imminent in the target frame share a generic temporal structure of continuation with expected culmination. Solid undirected lines in the diagrams indicate mappings from the generic structure to the inputs. (Whether a line is straight or curved does not matter.) In all cases, mappings from motion to time are structured by a shared temporal component. Each generic stage maps onto the stage in both inputs that has the same number.

The blended space at the bottom of Figures 2 and 3 makes the metaphorical construal explicit. Dashed lines indicate mappings from the inputs to the blend. In each case, elements in the inputs map onto elements in the blend that have the same number (and instantiate the same stage of the event). The dashed lines are omitted from Figure 3 to simplify the diagram.

Moving Ego is represented in Figure 3. Most of the details are the same as those of Ego-centered Moving Time except that Ego is in the role of Mover, and Locations map onto Times. Other than that, the main difference in English is that the ‘passing’ phase can be talked about more easily in Ego-centered Moving Time than in Moving Ego, as shown in (44).

1. (44)

   1. a. Sunday is passing slowly.

   2. b. ?We are slowly passing Sunday.

   3. c. The deadline is passing.

   4. e. We are passing the deadline.

Table 4 shows how Moving Ego and Ego-centered Moving Time have the same generic aspectual structure. Although the table is reminiscent of a blending diagram, note that it is not possible to create a blend that combines both metaphors (Lakoff, Reference Lakoff and Ortony1993, p. 218). The current analysis highlights the observation that motion and ‘time passing’ share temporal structure; for example, the target-frame phenomenon of the Future Time becoming more imminent and the continued motion of the Mover toward the Location in the source-frame both instantiate the generic stage 2 continuing process. Additionally, by identifying the target-frame stages, I try to state in target-frame terms some of what is understood as ‘time passing’.

table 4. Moving Ego and Ego-centered Moving Time with their generic structure

Moving Ego and Ego-centered Moving Time contrast very neatly with sequence is relative position on a path. One contrast is that motion can be factored out of the source frame of sequence is position. By this I mean that if only the relation between primary Figure and primary Ground is considered, there is no motion (cf. Section 4.3). Motion cannot be factored out in the case of the ego-perspective metaphors.

7.2. the generic structure of ‘sequence is relative position on a path’

Before we discuss the generic structure there is something that should be made explicit about the analysis regarding the sequence is position grounding scenario and the use of in-front/behind terms to designate the anterior or posterior semantic relations. We need to consider two possibilities: (i) the use of in-front or behind evokes a metaphorical motion scenario; (ii) the expressions evoke a metaphoric in-front/behind relation that does not involve actual motion, but may involve fictive motion or scanning. This latter version has the mapping in Table 5. (Compare Table 3 in Section 6.)

table 5. sequence is relative position on a path (schematic version; Moore, Reference Moore 2014a, p. 149)

As one would expect, the generic structure (in Figure 4) generalizes over both versions of sequence is position (Tables 3 and 5). The generic structure of sequence is position consists of an ordered relation between two discrete entities. The structure can be iterated so that there are more than two entities, but the ahead/behind and earlier/later relations are essentially binary. This generic structure maps onto each input as suggested by Figure 4. The individual entities of the ordered relation are not specified in the generic structure because the source and target do not have the same kind of order. (The nature of this generic notion of order is a matter for future research.) A salient indication that motion is not a necessary ingredient of the formulation of sequence is position in Figure 4 is that the generic structure does not mention anything temporal.

Fig. 4. Sequence is relative position on a path as a conceptual integration network.

To summarize: this analysis highlights the fundamental contrast between the ego-perspective metaphors with their single Mover moving relative to a Location and sequence is relative position on a path with its two Movers in a static relation to each other. The ego-perspective metaphors share considerable structure at the level of the schematic motion event. The generic structure of the ego-perspective metaphors is temporal, but that of sequence is position is not temporal.

10.1. motivation for the two-mover constraint

At this point the question arises of why English and other languages have a primary-metaphor mapping with two Movers in which the Mover that is ahead maps onto an earlier Time, but no counterpart mapping with two Movers in which the Mover that is ahead maps onto a later Time. The proposed motivation for this situation has to do with the observation that a metaphorical Mover that goes from a region that maps onto an earlier Time toward a region that maps onto a later Time is understood as an evolving Present Time. In the experiential correlations that motivate primary mappings, there is only one Present moment, so there can be only one earlier-toward-later Mover in a primary metaphor under the conditions stated above for the two-Mover constraint. Metaphorical motion from later toward earlier is not restricted in this way.

Support for the claim that a metaphorical Mover that goes from earlier toward later tends to map onto Now was given in Section 4.1.1 above on ‘The contrasting metaphorical directions of motion in Moving Ego and Ego-centered Moving Time’. The essential idea is that English speakers understand processes to occur in the present and develop as Now gets later. In Section 4.2, we noted some cases in which the earlier-toward-later metaphorical direction does not strictly speaking involve Now as a frame element, as in (57) (repeated from above).

1. (57) The beans boiled from dawn to dusk.

Even though examples like (57) do not have the present moment as a frame element, it is still reasonable to assume that a conceptualizer’s understanding of a continuing Time evokes an experience of Now. The reason for assuming this has to do with the observation that examples like (57) involve a certain Time which progresses through the period in question, getting later as natural processes happen. The only way that I have of understanding this concept of the special Time during which processes happen is by analogy to my own experience of Now. Indeed, conceiving of the evolution of a process requires a Now because the conceptualizer has to conceive of successive overlapping states that evolve from earlier toward later, as in the example of the cooking eggplant in Section 4.1.1 (cf. James, Reference James1891, chapter 15; Langacker, Reference Langacker, Pecher and Zwaan2005). Instead of profiling that evolution, the earlier=in-front relation profiles a comparison between earlier and later Times. That is, such constructions (e.g., Spring follows winter) basically just tell us which one is earlier and which one is later. They deal with two or more different Times, not necessarily with the development or evolution of any process.