iv Table of contents
Radden Günter, Klaus-Michael Köpcke, omas Berg and Peter Siemund
(eds.), Aspects of Meaning Construction. Reviewed by Annalisa
Annual Review of Cognitive Linguistics (), 1–26.
issn ‒ / e-issn – © John Benjamins Publishing Company
Before and aer
Relations of anteriority and posteriority along ‘paths’
of conceptual structure*
Aristotle University of essaloniki, Greece
e paper deals with the way events are ordered in a sequence with respect to
each other and with respect to the speaker. When ordered with respect to each
other, one event serves as the temporal reference point for establishing mental
contact with another event, the target. Taking the speaker as the reference point,
events are organized on the time line around the speaker and the present moment
of speaking (speech time) from which the speaker looks back at past time and
forwards at future time. e event-ordering relations are studied by before and af-
ter in physical, temporal and abstract space. Moreover, the variety of meanings of
before and aer is not only motivated by the spatial and nonspatial domains they
extend to, but also by the dierent patterns and modications of image schemas.
Keywords: anteriority/posteriority relations, event-ordering relations, event
time/reference time, physical/abstract space, image-schema transformations,
e paper deals with the systematic usage of the relations of anteriority and pos-
teriority, lexically expressed by the relators before and aer and grammatically by
various means, e.g. the perfect tenses for anteriority or the prospective construc-
tions for posteriority. It basically focuses on the way events are ordered with re-
spect to each other in a sequence and with respect to the speaker. is may also be
expressed by a sequence of clauses — a fact which reects the iconicity between
clause order and event order. If events are moments or intervals in physical or ab-
stract space, event-order relations link these moments or intervals.
In addition to before and aer, event-ordering relations can be expressed
by ahead, behind, in front of, in back of (but also by until, past). Event-ordering
2 Angeliki Athanasiadou
relations locate the target event they aim at in physical, temporal or abstract space
with respect to some reference point. Reference points, according to Langacker,
pertain to our ability to invoke one conceptual entity for mentally accessing an-
other conceptual entity. To serve as a reference point, a conceptual entity must
rst be established. One may then be led along a mental path towards a target with
respect to some reference point (Langacker, 1993, pp. 1, 6).
Before and aer are orientational relations on the horizontal axis. I will argue
that there is a systematic polysemy of the above relations which is not only due to
the dierent domains, i.e. spatial or nonspatial, they may belong to, but also to the
possibility of image schemas to be modied (by undergoing various operations).
e paper is organized in two parts: the rst part deals with the way the rela-
tions of anteriority and posteriority are expressed and extend to a variety of do-
mains. e second part interprets the polysemy of before and aer by means of
specic cognitive operations, which allow spatial and temporal orientation, focus-
ing on various purposes, thus creating new, slightly dierent or totally dierent
image-schema structures. ese cognitive operations are the image-schema trans-
formations by means of which “we make use of our ability to manipulate abstract
structure in mental space” (Johnson, 1987, p. 26).
2. Physical or abstract space
2.1 Physical space
One way of characterizing the relations of before and aer is by specifying the
horizontal alignment of the two entities related. Horizontality is the dimension
of orientation, which is made use of along the front/back axis. Another way of
characterizing the relations of before and aer is in terms of motion. A motion
event evokes a scene, which is composed of a set of sub-events. ese include the
starting point of an entity’s motion, its moving along a path and its arrival at an
endpoint. e mention of any of these sub-events is sucient to bring to mind the
whole scene. A scene of motion is understood on the basis of the image schema
of SOURCE-PATH-GOAL, which also holds for time as Point of Origin-Duration-
e temporal uses of before and aer are derived from the spatial senses, which
are in present-day English expressed by the prepositions in front of and behind.
PLACE TIME ORDER
in front of/behind (BrE) before/aer John before Mary
in front of/in back of (AmE) Mary aer John
Before and aer 3
e original spatial senses of English before and aer have given way for purely
temporal and order relations. In fact, in English, time is seen to be so important
that the double use of before/aer has not been kept for both space and time; be-
fore and aer hold for time and order whereas new prepositions have emerged for
place, namely in front of/behind. Before has no productive space sense any longer.
But, of course, some spatial ‘le-overs’ have still survived. Before in a spatial sense
is still found in rare idiomatic expressions such as to appear before (the) court and
has largely been replaced by the complex preposition in front of, especially when
no motion is involved. In fact, in a number of instances one can use in front of or
behind instead of before/aer. But, as will be argued, they dier in the sense that
they construe the spatial scene in two dierent ways, namely in terms of purely
physical space or in terms of abstract space.
2.2 Abstract space
In the spatial (or temporal) world, it is natural to expect physical things that are
subsequent to be arranged behind a thing, and things that are antecedent or earlier
to be arranged ahead or in front of others. In this way, both aer and before may
also involve an ordering of things: the ordering can be of customary conventions,
as for instance, the house numbers in a street or the case of the letters of the alpha-
bet, but it can also be an ordering with respect to degrees of priority and impor-
tance. At any rate, things are “located”. Aer all, there is the metaphor IMPORTANT
IS FIRST, as e.g. prime in Smoking is the prime cause of heart disease. In fact, as can
be seen in the examples in (1), taken from dictionaries,1 in addition to IMPORTANT
IS FIRST, there is also the metaphor IMPORTANT IS FRONT (Lako and Johnson,
1980, 1999, p. 50):
(1) a. He ranked Johnson aer Shakespeare.
b. He’s the tallest, aer Richard.
c. Your name comes aer mine in the list.
d. Aer you. (= please go rst)
e. Your name is before mine on the list.
f. He puts his work before everything.
g. e task before us is a daunting one.
In fact, “conceiving of an entity in terms of degrees of importance necessarily
involves mental scanning through the spatial domain along a number of items.
We are usually not aware of the mental scanning because it is immanent in (or
manifested through) the richer, more contentful conception of spatial motion by a
physical object” (Langacker, 1993, p. 4).
4 Angeliki Athanasiadou
When people or things follow, they may be ranked as next in excellence or
importance (1a–c) or priority (1d). When people or things are ahead, they are in
a position superior in terms of order or importance (1e–g). In all instances of (1),
behind or in front of are impossible or unlikely. When situations are ahead of us
they may not only be in an important or superior position but they may also be
unknown to us as they belong to the future (2a); in our Western cultures, the fu-
ture is conceptualized as being in front of us. Situations ahead of us may also be in
store for, awaiting for us to be taken into account (2b). In both instances of (2) we
have the metaphor THE FUTURE IS BEFORE US (Lako and Johnson, 1999, p. 140):
(2) a. e whole summer lay before me.
b. A young person’s whole life lies before him.
In both examples of (2) there is an objective: ‘the summer is there for me to engage
in a number of activities’ (2a), ‘a young person’s life is there in order to make the
best of it’ (2b). ese two examples are closely related to the implicatures in (1f )
and (1g). e importance of the work is given priority (1f) and the task is a chal-
lenge for us (1g). Moreover, as was briey mentioned, before seems to have a few
‘traces’ le of its original spatial sense, as in:
(3) e case is now before the court.
‘e case is under the consideration or jurisdiction of the court’. is expression is
no longer a creative pattern; it is an idiom of the earlier use of before in its spatial
sense. is spatial sense has been preserved possibly because the court metonymi-
cally stands for the judge and the jury.
Examples (1e–g), (2), and (3) reect the historical evolution of before. Its ini-
tial spatial meaning of ‘in front of ’ is now almost extinct.2 Before has been split
from its spatial sense, which has found new forms (in front of). In its metaphorical
usages as in the examples under (1e–g), its spatial sense of “order” has been pre-
served in the source domain; it can still be very “locational”. In addition to “order”,
the meaning of “target”, that is ‘in order to go at it’, implies challenging important
situations when X before Y as in example (2).
Moreover, before invites the implicature of ‘in co-presence of ’ because for a
trajector to be in front of a landmark implies their being co-present. It also implies
interaction or being brought face to face with each other:
(4) a. She asked that the visitor be brought before her.
b. ey had the advantage of playing before their home crowd.
c. He was brought before the judge.
In (4) the sense of appearance (in the presence of) is involved. If the situations in
(4) were purely spatial, being brought in front of her, playing in front of the home
Before and aer 5
crowd, being brought in front of the judge would mean that ‘she’, ‘the home crowd’
and ‘the judge’ were sitting behind ‘the visitor’, ‘the team’ and ‘him’ respectively,
which would be a case of oversimplication. But, before, in the course of its evolu-
tion, has developed an inherent meaning of “priority” in taxonomic (order) and
appearance senses. It is worth mentioning that in (3), (4a), (4c), and to a lesser
degree in (4b), before is employed for entities denoting authority, superior posi-
tion or high rank and one needs to ask for their decision, approval or verdict. So,
the visitor in (4a) is brought in the presence of or face to face with her, the crowd
is lling the whole stadium in (4b), and he is brought in the presence of or face to
face with the judge in (4c).
Aer invites the implicatures ‘aiming for somebody’ or at least ‘following
somebody’ whereas before is used to say that ‘somebody or something is a chal-
lenge or in higher power than somebody or something’:
(5) a. I’m always having to clean up aer the children.
b. He ran aer her with the book.
c. He’s aer a job at our place.
d. She asked aer you.
e. Before you is a list of the points we have to discuss.
f. ey knelt before the throne.
e implicature invited by aer in (5a) is that this person is the one doing the
cleaning aer the children had messed up the room and possibly repeatedly or
on a regular basis. In (5b) he ran behind her in order to catch up with her. In (5a)
we have the particular purpose of being charged with consecutive responsibili-
ties and in (5b) of hurrying to catch her. Aer in (5b) and (5c) has extended from
the meaning of ‘following in time or space’ to the meaning of ‘pursuit or search
of’: when in quest or pursuit of someone or something, one tries to nd or catch
somebody or something. e aer-phrase has been conventionalized with go aer
as a “goal”. e goal lying in front of us may be an animate one ‘going aer those
high in rank’ which is a possible extension from the “order” meaning of aer; it
may be an inanimate goal seeking aer fame, going aer big money even hunting
aer money; it may also be “order+goal” as in going aer/chasing aer the thieves,
or police being aer people. Purpose or goal events can thus be metaphorized as a
race in which the goal to be attained is itself in motion with the human running
aer it. e metaphor is purposes are destinations (Lako and Johnson, 1999,
pp. 52, 61), and going “aer” is only a special kind of motion.
e purpose of following someone may also extend to the conventional im-
plicature of ‘being concerned about someone’, such as asking to nd out about
someone or inquiring aer one’s health (5d). e goal now is asking information
about how somebody is doing.
6 Angeliki Athanasiadou
e goal in (5e) is the task we have to work through. Were it purely physical
space it would be something like: In front of you there is a huge le; please do not
touch it. Abstract space would be reected in: Before you is a huge le; start as soon
as possible. e metonymy in (5f), that is the throne standing for the monarch,
communicates the goal very eectively. e particular set of circumstances inher-
ent in this situation is probably worship or submission.
2.3. More abstract space
Our understanding of spatial and temporal relations as expressed by before and
aer involves more than situating entities in physical and temporal space: it si-
multaneously means relating the entities to each other along even more abstract
dimensions. eir choice is largely motivated by our metaphorical understanding
of the various abstract concepts in terms of spatial relations. erefore, it is only
natural for before and aer to transcend their spatial and temporal senses and to be
used for more abstract relations such as contrast, cause or purpose among others.
Due to the abstract nature of these domains only a few of the properties of before
and aer can be related to the domain of physical or temporal space. Of course, we
cannot expect concrete spatial or temporal dimensionality to be preserved in ab-
stract concepts. From the two orientational prepositions, aer exhibits a remark-
able extension to a variety of domains. It extends to the domains expressed by the
(6) a. It was pleasantly cool in the house aer the sticky heat outside.
b. I’ll never forgive him aer what he said.
c. ey are still friends aer all their dierences.
d. I can’t believe she’d do that, not aer all I’ve done for her.
e. He won the race aer all!
e domain of contrast, including both mild contrast or strong contrast, with con-
cession, and adversativeness, is present in the examples in (6). In (6a) the events
in contrast are the cool atmosphere in the house and the sticky heat outside. So
‘it was cool inside against the heat outside’. However, the contrast, which mainly
results from the lexical items in the house/outside, is further strengthened by af-
ter, which here designates a relation of sequence, i.e. it implies the succession of
events (they were outside rst and have now gone inside). e implicatures in (6a)
and (6b) are based on our common experience that earlier events may cause later
events. In (6b) aer invites the implicature of ‘as a result of or because of the things
he said’. e sense of contrast in (6c) implies concession ‘regardless of’ or ‘in spite
of’. Concession, in its turn, merges with adversativeness in (6d) ‘in spite of all I’ve
done’ or ‘although I’ve done things’. Similarly, in (6e), the sense of contrast is met
Before and aer 7
and the meaning is ‘in spite of expectations and/or eorts he won the race’. us,
events that follow other events in time or space may be in contrast, concession and
adversativeness or may be the result of other events. e cause of the contrast, of
the concession or the adversativeness and of the result may also be given.
(7) a. Aer what you have done, you shouldn’t complain.
b. Aer all, why worry about it?
e aer-phrase in (7a, b) may be in both initial and nal position (You shouldn’t
complain, aer what you have done and why worry about it, aer all?). is exten-
sion to the meaning of ‘considering’ illustrated in (7a) is found with expressions
of communication and mental state and implies a mental goal one has to take into
account. is has given rise to the use of the expression aer all as a sentence ad-
verb in (7b), which means ‘when everything is considered’. (is meaning of aer
all is not met in (6e)). Finally, following someone or something may metaphori-
cally extend to following a style, or a model, imitating or being in the manner of
someone or something:
(8) a. A painting aer Goya.
b. A man aer my own heart.
c. We named the baby ‘Ena’ aer her grandmother.
is is illustrated in (8a). When following a style or when imitating, results in be-
ing in accordance with the nature or the desires of someone or something, it may
be in conformity with someone or something, that is, meeting the aspirations of,
as in (8b), or even it may be a convention or a customary act to be followed, as
3. Conceptual and linguistic temporality
According to Langacker “an image schema is immanent in the corresponding
conceptual archetype, i.e. activation of the former is inherent in the activation of
the latter (though not conversely). For example, conceiving of an object moving
through space involves mental scanning through the spatial domain” (Langacker,
1993, pp. 2–4). If spatial path is the prototype, one has the ability to make exten-
sions and chains of extensions that reach higher and higher levels of abstraction.
Extensions from a prototype are based on some perceived similarity between the
Time, like space, is a fundamental domain of our experience and cognition.
Research on temporality is basically oriented towards grammatical marking,
namely tense and aspect of the verb. ere is also research on temporal adverbials
8 Angeliki Athanasiadou
or particles though these cases have not been a focal area of linguistic investigation,
possibly because tense and aspect are deeply rooted in the structural organization
of languages: tense refers to the grammatical expression of speech time relative to
some other time, and aspect designates the internal temporal organization of the
situation described by the verb.
e speaker or the hearer ‘scans’ the time line in order to locate events in
time. We understand motion through time metaphorically in terms of the spatial
SOURCE-PATH-GOAL image schema: time has a starting point, a path or duration
and an endpoint. Most of the prepositions describing temporal motion are also
used in a spatial sense in English and even those we think of as being purely tem-
poral ones, like till, until, and ago, historically derive from spatial concepts.
Indeed, time is conceived of as a ‘time line’ upon which events are situated. e
time line corresponds to a path in physical space, events correspond to entities in
physical space. In this way, our metaphorical understanding of time as space is to
a large extent reected in the temporal use of spatial prepositions. Just like spatial
relations, temporal relations may be characterized according to state and motion
and according to dimensions and orientations (Radden and Dirven, 2007).
When it comes to the specic analysis of event-order relations, temporal in-
formation only constitutes a source (very basic though) that creates implicatures
of various other orders, the temporal order referring to one, though very frequent,
domain, namely the domain of time. Concerning the relations of before and aer,
the picture exhibited by dierent dictionaries (though mainly by Collins/Cobuild
and Oxford) is the following:
(9) a. e year before last he won a gold medal, and the year before that he won
b. Leave your keys at reception before departure.
c. Before I made a decision, I thought carefully about it.
(10) a. ey arrived shortly aer 5.
b. Aer winning the prize she became famous overnight.
c. We’ll leave aer lunch.
Before refers to events at a previous point in time (9), while aer refers to events
subsequent in time (10). Bohnemeyer (1998, p. 239), who investigates the event-
ordering relations of before and aer in Yucatec Maya, emphasizes their role in
time reference and temporal coherence. Indeed, as is shown, the function of before
and aer is to establish and maintain time reference and temporal coherence in
discourse. ey express two-place relations between a target event and either a
reference event, which serves as reference time, in adjacent discourse (9b, c), (10b)
Before and aer 9
or the conventionalized temporal point of reference (9a), (10a): the temporal point
of reference is when he won medals or when the time of the arrival was set.
But to begin with, and following Radden and Dirven’s Cognitive English Gram-
mar (CEG) (2007), it is important to show the dierence between iconic order,
that is word or clause order reecting event order, and construed order. In iconic
order, each event constitutes the reference point for following events:
(11) She married; she got a baby.
(12) She got a baby; she married.
In construed order, the speakers build in gestalt order; the events are not only
marked with respect to their succession but also with respect to foregroundedness
and backgroundedness, which is linguistically expressed in terms of subordina-
tion. e foregrounded information is in the main clause and the backgrounded
information in the subordinate clause. All this leads to very great ‘freedom’ of
(11) a. She married before she got a baby.
b. She got a baby aer she married.
(12) a. She got a baby before she married.
b. She married aer she got a baby.
In our attempt to capture the way events are ordered in a sequence, we make refer-
ence to the domain of time, which is conceived of as one-dimensional and direc-
tional. e temporal ordering of time units metaphorically corresponds to their
spatial sequencing. e sequential order of events can be conceived of as con-
sisting of points on the time axis with horizontal, that is, front and back (ahead/
behind) or vertical (up/down) directionality. We start with the horizontal axis due
to our spatial experience of motion, which is mainly directed to the front. Given
the front-back orientation, goals are located in front of the experiencer, who self-
orients with respect to these goals. us, entities and events count as goals located
in front of the experiencer. e future, then, derives from an anticipation the expe-
riencer makes in the present about an objective or goal. e future is elaborated in
terms of prepositions as ahead, in front of, before: e future lies before us, She has
a bright future ahead of her. In contrast, the past is derived from moments held in
memory. Tasks and goals are nished, and as we turn to the next tasks, we turn our
back on tasks that are completed. In the Western languages, past events are seen
as being located behind the experiencer. e past is elaborated in terms of behind:
My childhood is behind me. Temporal events can be represented in their own right
when they can be checked from our experience of the world or they can be repre-
sented metaphorically when they are le unspecied from our world experience.
10 Angeliki Athanasiadou
is is most oen done through spatial metaphors. So when we put forward our
ideas or we leave the past behind us we rely on space to talk about time.
Indeed, we tend to think of time metaphorically in terms of the most tangible
domain of experience, which is space. Past, present and future are units of time
lined up along a horizontal axis. We, the speakers, are observers positioned in the
present, seeing the future in front of us and the past behind us. Using the relations
of ‘before’ or ‘anterior’ and ‘aer’ or ‘posterior’ we make use of the spatial model
ere have, however, been dierent arrangements of the two sets of concepts,
namely of past and future on the one hand, and of earlier and later on the other.
e latter set is expressed by before, beforehand, formerly, previously for ‘earlier’
and aer, aerwards, hereaer for ‘later’. Both sets are elaborated in terms of their
location on the front/back axis, that is, the location of earlier or later events is
similar to the location of events in the past or in the future. e dierent arrange-
ments have to do with whether the reference point is the experiencer or another
temporal event. In the examples below the reference point is a temporal event,
lunch in (13a) and breakfast in (13b):
(13) a. Breakfast comes before lunch.
b. Lunch comes aer breakfast.
Breakfast is located with respect to lunch (13a), while lunch is located with respect
to breakfast (13b). Traugott (1978) has observed that sequencing of events is in-
dependent of the concepts of past and future, as the only information we need is
relative sequencing, rather than notions of past and future.
e concepts of past and future are tied to the human experiencer as they are
related to the perceptual moment. us, they do not aect our understanding of
events being sequenced with respect to one another.
An example oen cited for a temporal event being the reference point is the
case of a race (Evans, 2003). e winner, that is, the earliest arrival is in front of the
later arrival, who is behind the winner, in such a way that the sequence of arrival
correlates with the relative location of the athletes with respect to one another. is
correlation, which establishes an event in a sequence as the reference point, moti-
vates our conceptualization of events as a sequence in terms of before or aer. e
events are elaborated in terms of the concepts earlier and later located on a path,
having nothing to do with the use of the concepts of past and future.
Taking the human experiencer as the reference point, there are, in fact, three
notions of time (Radden and Dirven, 2007, Chapter 9). Time can be organized
on the time line around the speaker as the observer and the present moment of
speaking. is is referred to as speech time, which is an idealized present moment
from which the speaker/observer can look back at past time and forward at future
Before and aer 11
time and locate events on the time line. e time at which a situation, typically an
event, holds, held or will hold is known as event time. e times of events are de-
scribed by means of dierent tense forms. In English, there are two types of tenses:
absolute and relative tenses. Present, past and future are absolute tenses which
allow speakers to locate any situation in three spheres of conceived time. ey are
determined with respect to speech time. In their dependence on speech time, the
three absolute tenses are, therefore, deictic categories in the sense that they ground
situations with respect to speech time: the present tense grounds a situation at,
around, or including, speech time, the past tense grounds a situation before speech
time, and the future tense grounds a situation aer speech time. For example, the
event described by I’ll go sailing next Sunday is located in future time as seen from
present speech time — some days later it would turn into a present event and aer
that into a past event. But in addition to the three basic tenses, we also have tenses
which relate event times to another time, which serves as reference point, and thus
only indirectly to speech time. ese are the relative tenses, which denote relative
times. A relative time may be specied anterior or posterior to each of the three
absolute times; it is located relative to a reference time serving as temporal refer-
ence point. For instance, yesterday serves as the reference time in the past relative
to which we locate the day before yesterday as an anterior day, tomorrow serves as
the reference time relative to which we locate the day aer tomorrow as a posterior
day. Relative tenses, therefore, allow us to express times anterior or posterior to
present, past or future time. And we have perfect forms for the expression of ante-
riority and prospective forms for the expression of posteriority.
In the case of absolute tenses, event time and reference time coincide: in the
simple past e train le as well as in the simple future e train will leave in ten
minutes both event time and reference time fall before or aer speech time respec-
tively. In the simple present (I see our train over there) speech time falls within the
same time sphere together with reference time and event time. But in the case of
relative times we witness the separation of event time and reference time. Relative
times only relate events to speech time (S) indirectly via reference time. is cre-
ates a special link between event time (E) and reference time (R).
Anterior times are expressed by the perfect tenses. Perfect tenses denote a rela-
tive time anterior to a reference time. e reference time may be each of the three
absolute times: present, past or future time as in the examples below where we
mentally locate events anterior to a present, past or future time:
12 Angeliki Athanasiadou
a. e train has just le. .
event time reference time
anterior to present present time E R, S
b. e train had le when I arrived.
event time reference time
anterior to past past time E R S
c. e train will have le when we get there.
event time reference time
anterior to future future time S E R
All sentences display the same basic conceptual pattern: a moment in present (a),
past (b) or future time (c) serves as the reference time from which we mentally ac-
cess the anterior time of an event.
Posterior times are expressed by prospective constructions: they are located
at a later time than reference time. ey involve intentional future (d), a past mo-
ment relative to later past, present or future moment (e), or they predict a future
event at a later point in time (f):
d. I am going to leave this week.
present posterior time S, R E
e. First, I was going to leave last week. R (e) S, (e) (E)
past posterior time
f. Now, I’ll be going to leave next week.
future posterior time S R E
In posterior constructions with going to, reference time refers to the moment when
an intention about an event in later time is conceived. An intention may be formed
in the present, past or future, and the event conceived is posterior relative to one
of these reference times, but it is not obvious when it is to occur (this explains the
parentheses put around the three Es in (e)).
All the above situations from (a) to (f)4 involve a temporal relation of anteri-
ority or posteriority represented by event time relative to reference time. So, the
relations of anteriority and posteriority are used only in connection with relative
time, not with absolute time. Anteriority and posteriority is not the same as before
and aer but only the same as before and aer in relation to a reference point that
is dierent from speech time. In other words, the relations of anteriority and pos-
teriority do not apply to the absolute tense system. In (14), the events are ordered
on the time axis.
(14) a. ey nished cooking.
Before and aer 13
b. ey will nish cooking.
In (14) both event time and reference time fall before (14a) or aer (14b) speech
With the relations of ‘before’ or ‘anterior to’ and ‘aer’ or ‘posterior to’ the per-
fect tenses need not be expressed; the events they refer to are located with respect
to some reference point. We have a spatial, temporal, or abstract path, along which
events occupy a certain position (the spatial, the temporal axis, or otherwise) and
certain boundaries, intervals or spans within which they occur. One of the events
is a prominent or focused entity. is is a fundamental capacity of ours to “…in-
voke the conception of one entity as a cognitive reference point for purposes of es-
tablishing mental contact with another, i.e. to single it out for individual conscious
awareness” (Langacker, 1993, p. 5). In (15) a bounded event is specied in relation
to another, contextually given event.
(15) a. ey nished cooking before going to school.
b. ey nished cooking aer coming back from school.
By means of the conjunctions before and aer, the event time, i.e. the time of the
activity of cooking, is located as anterior to the reference point, i.e. school (15a),
whereas in (15b) it is posterior to it. e reference point, however, may be an event
described in the preceding sentence:
(16) a. I arrived at the station at 12. e train had le. R (E+S)
b. I will arrive at the station at 12. e train will have le (by then). R (S+E)
e perfect tenses in (16) (past perfect, future perfect) order the event of the leav-
ing of the train as anterior to the reference event of the arrival. Finally, adverbs
such as aerwards or beforehand can be considered expressions of relative times:
(17) ey nished cooking. Aerwards they le for school.
R (E+S) R E
Event-order relations allow us to account for a wide range of phenomena in linguis-
tic temporality: the concept of time, the system of tenses, the connectors employed.
Such phenomena formed the basis of previous studies on linguistic temporality. Let
us briey compare the analysis above by Radden and Dirven against some other
approaches. Quirk and Greenbaum, in particular, regard before and aer:
i. as prepositions of time followed by a temporal noun phrase (before next week)
or a subjectless V-ing clause (before leaving school) (Quirk and Greenbaum,
ii. as subordinators in nite adverbial clauses of time (buy your tickets before you
reach the station). -ing clauses without a subject are also used to express time
14 Angeliki Athanasiadou
relationships and they are common in initial position (e stranger, having
discarded his jacket, moved threateningly towards me (aer he had discard-
ed…)) (Quirk and Greenbaum, 1976, 11.21).
Comrie (1985, p. 122) proposes a simple analysis of the temporal relations of an-
teriority and posteriority, namely E before S and E aer S. S is the time of speech
and E is not necessarily a point, it is “the time point or interval which is occupied
by the situation to be located in time”. e relations before and aer mean ‘properly
before’ and ‘properly aer’.
e above two approaches are limited in scope: Quirk and Greenbaum fail to
distinguish the three time concepts of speech time, event time and reference time
whereas Comrie does not take the notion of reference time into account.
Klein (1994, pp. 59, 61–62) introduces the ‘Basic Time Concept’ and charac-
teristics such as segmentability, linear order, proximity, and duration, which are
indispensable for notions of time. Klein conceives of time as consisting of units
(moments or intervals) and of event-order relations linking these units. He sees
communication as the mere ‘coding’ of content packages without interactional pa-
rameters. But events are not just intervals that follow or precede other intervals
in time. According to Bohnemeyer (1998, p. 243) pre- and post-states5 are not
just time units in temporal ordering only, so that they follow or precede the target
event in time; they rather have their own status. A post-state reects some kind of
trace the target event leaves behind, whereas a pre-state precedes the target event
not only in time, but it is related to the target event in other ways, not necessarily
temporal ones (Bohnemeyer 1998, p. 243). Klein’s view would be adequate strictly
for the relations of anteriority and posteriority having no causal or other implica-
tion at all. What is lacking in his view is that the relations of antecedent and sub-
sequent are also causally linked, in addition to being temporally linked, or due to
the fact that we have complex event structures which may be linked in a variety of
ways even more complex than the purely temporal one.
It is true that the order the events will have basically depends on the amount
of temporal information, which interacts with world knowledge and reects an
iconic relationship between clause order and event order. So one thing is the solid
amount of temporal information for the sequence of events: they are time intervals
that follow or precede S. However, the occurrence of an event preceding S or fol-
lowing S may additionally be related to S in some non-temporal way. It may be the
consequence or the purpose with respect to other events. In this way, events are
not just spans or intervals but idiosyncratic entities.
e picture of the uses of before and aer discussed so far is a very solid one.
Both relators derive from the domain of physical space. At a certain point in time,
they had both the meaning of space and time and in present-day English their
Before and aer 15
meanings are primarily restricted to time.6 So their evolution could be schemati-
cally represented as:
Spatial Spatiotemporal Temporal
At each one of these phases there have been steps of extension into other more
abstract senses. In particular, the spatiotemporal meaning of aer, which is ‘fol-
lowing in space or time’, gave rise to the meaning of ‘following a given order’, which
in turn gave rise to the meanings of priority and importance. Similarly, the mean-
ing of ‘following a goal’ derives from the central spatiotemporal meaning of aer.
us, ‘being in pursuit or in search of ’ or even ‘following a mental goal’ leads to
the meanings of ‘imitating’ or ‘being in accordance to’, which imply time.
4. Before and aer: Variations on a schema
One of the important claims of cognitive semantics is that much of our knowl-
edge is not static, propositional and sentential but is grounded in and structured
by various patterns of our perceptual interactions, bodily actions, and manipula-
tions of objects (Johnson, 1987; Lako, 1987, 1990; Talmy, 1988). ese patterns,
called image schemas, emerge throughout sensorimotor activity as we manipulate
objects, orient ourselves spatially and temporally, and direct our perceptual fo-
cus for various purposes (Johnson, 1991). Dierent image schemas and several
image-schema transformations appear regularly in people’s everyday thinking,
reasoning, and imagination (Johnson, 1987; Lako, 1987). Among these image
schemas are CONTAINER, SOURCE-PATH-GOAL, PATH, CENTER-PERIPHERY, etc. Im-
age schemas can be dened as analogue representations of spatial relations and of
movements and orientations in space. However, image schemas are not just single
entities, but they are linked together to form very natural relations (causal, tempo-
ral, part–whole, agent–patient, …) through image-schema transformations. ese
latter operations play a special role in people’s cognition. Path-focus to endpoint
focus, multiplex to mass, following a trajectory are some of the most important im-
age-schema transformations. Schema transformations are anything but arbitrary.
ey are direct reections of our experiences, which may be visual or kinesthetic
( Lako, 1987, pp. 442–3; Johnson 1987, p. 26).
Lexical meaning, then, is based on the abstract, conceptual structures of image
schemas (Lako 1987; Langacker 1987, 1991, pp. 398–403); in fact, image schemas
to a large extent motivate people’s understanding of lexical meaning. e dier-
ent physical and abstract senses of polysemous words are related in systematic and
recurrent ways throughout the lexicon. ese meanings “are not arbitrary for na-
tive speakers but are motivated by people’s recurring bodily experiences in the real
16 Angeliki Athanasiadou
world” (Gibbs et al., 1994; Gibbs and Colston, 1995, p. 352). I will, then, attempt to
account for how dierent patterns of image schemas motivate dierent uses of be-
fore and aer, seeing them from a dierent perspective. Along these lines some ex-
amples from Section 2 are here discussed again but from a dierent point of view.
4.1 Metaphorical mapping from physical space to temporal space
ere is a very strong relationship between time and space. “e spatialization of
time is so obvious and so pervasive a phenomenon in the grammatical and lexical
structure of so many of the world’s languages that it has been frequently noted,
even by scholars who would not think of themselves as subscribing to the hypoth-
esis of localism” (Miller and Johnson-Laird, 1976, p. 375; Lyons 1977, p. 718). As is
noted, the temporal and spatial terms in many languages overlap considerably At
the corner of the street — at 7:00 o’clock, from Athens to London — from 7:00 to 9:00,
etc. One reason for considering time and space having an intimate link is due to
Lako and Johnson (1980) who drew attention to the power of the metaphor TIME
AS SPACE. In conceptualizing time as space, it is thought of as a unidimensional,
directional and dynamic entity (Clark, 1973; McGlone and Harding 1998; Radden
2003, among others). In this paper, the focus is on terms used to show orientation
of the time-line. As time is thought to be directional, we think of time as moving
front or back (before/aer), up or down and not le or right by means of lateral
symmetric terms (Clark, 1973; McGlone and Harding, 1998; Traugott 1978). Time
is a phenomenon, which is characterized by the appearance and disappearance of
entities and events situated on or moving along an axis. We, the conceptualizers,
experience this on-going change of the sequential order of events and we import
spatial terms such as before/aer, ahead/behind or up/down (Boroditsky, 2000, pp.
3–4). However, there are many aspects of time that are le unspecied in our expe-
rience of the world but are specied in our language — most oen through spatial
metaphors (Lako and Johnson, 1980, 1999). For instance, conceptually, time is
unspecied with respect to its motion, i.e. whether it moves horizontally or verti-
cally or whether it moves forward or back. But metaphors such as look forward to
tomorrow, being behind schedule or proposing theories ahead of our time express
horizontal movement to the front; above all, these expressions show the perva-
sive correspondence between space and time in language. In this paper, following
Boroditsky (2000, p. 5), the focus is on the way events are ordered in a sequence
with respect to each other and to the speaker. And this rst takes place in physical
space and then it is metaphorically extended to temporal (and/or abstract) space.
Two distinct models of time as motion are implicit in English temporal ex-
pressions. Clark (1973) referred to these two conceptualizations of time as the
moving-ego and moving-time. In the moving-ego model, time is conceptualized
Before and aer 17
as a stationary entity and the observer moves from the past via the present into
the future; alternatively, in the moving-time model, time is conceptualized as a
dynamic entity that ows past a stationary observer from the future toward the
past. is can be schematically represented in the Figure below:
Figure 1. Spatial depictions of the moving-ego and moving-time perspectives (Clark,
In the moving-ego perspective, the front-back orientation of time is due to an
observer’s frontward movement, which shows in expressions like the weeks ahead
of us, Christmas lies ahead and concerns events in the future. It is also due to an ob-
server’s backward movement, which shows in expressions like Christmas is behind
us, the worst is behind us and concerns events in the past.
In the moving-time perspective, in which the observer is stationary, still the
front region determines future events and the back region past events Troubles
lie ahead, at was way back in 1900 (examples from Radden, 2003, p. 229). As-
suming that events are moving from the future to the past (see the moving-time
perspective in Figure 1 above), they may be assigned a front and a back. If events
occur earlier than a reference point, they are in front of it (Summer is ahead of the
fall semester) and if they occur later, they are behind it (e Christmas season fol-
lows the fall semester) (examples from McGlone and Harding, 1998, p. 1212).
ese two dominant spatial metaphors used to sequence events in time are
also discussed in Boroditsky (2000, pp. 5–6). In her example e revolution is be-
fore us, the revolution “is a later or future event which is said to be before because it
is further along in the observer’s direction of motion”. So, just as we move along a
path and the objects are ordered we can say they are in front of us because they are
further along in our direction of motion. is is the moving-ego perspective. But
in the moving-time perspective the front is assigned to a past or earlier event. So
in Boroditsky’s example e revolution was over before breakfast, the revolution “is
the earlier event and is said to be before because it is further along in the direction
18 Angeliki Athanasiadou
of motion of time”. e results of Boroditsky’s experiments also showed that the
domains of space and time share relational structure. So in:
(18) a. Did she leave a message before she went?
b. He arrived before me.
e meaning of before in (18) is derived by a metaphorical mapping of the hori-
zontal axis from the spatial to the temporal domain. Before is moving or lying in a
position along a temporal path.
e mapping from physical to temporal space is one of the principles that
contributes to the picture provided by before and aer. e horizontality schema,
to which they belong, includes various horizontal and vertical axes and paths dif-
fering in orientation and direction. In accounting for the polysemy of before and
aer, I will only refer to the cognitive operations allowing us to use the frontward/
backward expressions with respect to the location of a reference point in relation
to the poles of a frontward directed axis. is will be discussed on a par with the
degrees of dynamicity characterizing the uses of before and aer. So before and
aer establish and maintain spatial, temporal or more abstract coherence in dis-
course. However, they do not only locate events on the path axis, as relations of
event order can be either anterior or posterior to some reference point, but they
can also express abstract or mental motion.
e reference point (R) may be singled out to invoke the conception of one
entity for purposes of establishing mental contact with another, the target (T)
through a mental path followed by the conceptualizer (Langacker, 1993, pp. 5–6).
e reference point may be located either in a subsequent position (Schema I) or
in an antecedent position (Schema II) in relation to the frontward directed axis.
Schema (I) R E
aer you shut the door
Schema (II) E R
turn lebefore bank
Figure 2. Orientation along the frontward horizontal axis
(19) a. Shut the door aer you.
b. Turn le just before (= before you reach) the bank.
e use of aer in (19a) shows that you is located in a subsequent position to the
pole of a frontward axis (Schema (I), Figure 2), and that mental motion is implied
(you have rst passed through it). If we have Shut the door in front of/behind you,
there is no implication of mental motion, (that you have passed through it). *Shut
the door before you is impossible. e use of before in (19b) shows that you is lo-
Before and aer 19
cated in an antecedent position (Schema (II), Figure 2) and mental motion takes
place before the reference point. With respect to the horizontal axis, before and
aer may be used metaphorically in abstract domains.
(20) a. He is going aer big money.
b. ey are still friends aer all their dierences.
c. See me before you leave.
d. He puts his work before everything.
e use of aer in (20a) shows that the human observer tries to arrive at an event
as the desired target and accomplish it, whereas in (20b) dierences are le behind
and the reference point is ahead of them. In (20c) we have a purely temporal hier-
archy where seeing is an earlier event with respect to the reference point of depar-
ture. In (20d) the reference point work is high up in the hierarchy and everything
4.2 Horizontal Axis → Vertical Axis
Ekberg (1997, p. 70) explicitly deals with the manipulation of the vertical axis to
refer to nonvertical relations (Vertical Axis → Horizontal Axis). She shows “how
the image schema of a Vertical Axis may be transformed so that it will t contexts
where there is no (apparent) vertical relation between trajector and landmark”
(1997, p. 84). Furthermore, she supports the view that the possibility of using
expressions of verticality for nonvertical relations is due to the importance of the
vertical axis on the basis of biological and psycholinguistic factors. However, the
vertical and the horizontal axis have a common experiential basis: when we walk
or swim, our heads become fronts exactly like the fronts of moving objects (Yu,
1998 p. 111; Svorou, 1994, pp. 73, 150 also shows that “terms for 'head' may give
rise to either front-region or top-region grams”). So there may also be prefer-
ences for the horizontal axis due to our spatial experience of motion, which is
directed to the front. Hence, the horizontal axis may be transformed into the
e relation between the frontward pole of the horizontal axis and a specied
reference point is preserved when the horizontal axis is transformed into a vertical
one. As was said earlier, image schemas generate spatial concepts, but also new,
extended, though interrelated, meanings or instances of polysemy occur because
various cognitive operations may apply to image schemas.
Since image schemas are kinesthetic in nature they concern awareness of
many aspects of functioning in space: orientation, motion, balance, shape judg-
ments, etc. (Lako, 1987, pp. 445–6). e transformation of the horizontal axis
into the vertical one is based on the fact that we can perform mental operations on
20 Angeliki Athanasiadou
schemas that are abstract analogs of physical processes or spatial operations (John-
son, 1987, pp. 23, 25). ese operations are embodied; our orientation towards the
front or back may be transformed into the vertical up or down due to our upright
position to denote things we experience in the vertical axis such as list of names
(21a, b), height (21c) or priority (21d). is operation is lexically manifested in the
extended uses of before and aer as can be seen in the examples below, which are
some of the examples cited in (1):
(21) a. Your name comes aer mine in the list.
b. Your name is before mine on the list.
c. He’s the tallest, aer Richard.
d. He puts his work before everything.
e mental imagery of a list (21a, b) or of the height of people (21c) or of tasks that
take place in an order (21d) recall classication (see also the oen-cited case age
before beauty) and is of a horizontal or vertical orientation (names can be listed
one aer another according to alphabetical order, persons can be ordered one af-
ter another according to height, tasks are ordered according to the priority given
to them). As image schemas motivate abstract metaphorical concepts — in fact,
people project image schemas from one domain onto another — we conceptual-
ize quantity or importance in terms of verticality: MORE IS UP and LESS IS DOWN
(Lako and Johnson, 1980).
According to Ekberg (1997, p. 77), when transforming the vertical axis into a
horizontal one, an entity A which is located “above” an entity B will, by implica-
tion, be located “behind” B (e gure below given by Ekberg shows the corre-
spondence between “above” and “behind”; x stands for the reference point).
Along the same lines, my claim is that when the horizontal axis is transformed into
the vertical one, an entity located before or behind will now be located ahead or
above another entity (22a), and reversedly an entity located aerwards will now be
behind and follow (22b):
Before and aer 21
(22) a. Your name is before mine on the list.
b. ‘Against’ comes aer ‘again’ in the dictionary.
As was said in 4.1., we think of time as moving on a horizontal axis (before/aer)
and on a vertical axis (up/down). On the horizontal axis, an earlier point in time
is back as in at’s all behind us now, Let’s put that in back of us. A later point is
in front of us as He has a great future in front of him, We’re looking ahead to the
Figure 3. Temporality on the frontward and backward pole of the horizontal axis
Ekberg showed the correspondence between “above” and “behind”. She also dealt
with the temporal meanings of “up” (and “down”) noticing that time may be con-
ceptualized as moving upward (1997, p. 83). In the present study, the horizontal
ahead8 may be transformed to vertical up so that future may also be seen as “up”.
is is in accordance with Lako and Johnson’s view, when they relate the un-
known with “up” and the known with “down”, as at’s up in the air (UNKNOWN
IS UP) versus e matter is settled (KNOWN IS DOWN) (Lako and Johnson, 1980,
Figure 4. Correspondence between “ahead” and “up”
is is also consistent with what Yu (1998, p. 111) claimed, namely that “up” and
“front” have a common experiential basis. But earlier time can be “up”, that is, past
can be up as Yu (1998, p. 112) says about older generations being ascendants and
younger generations being descendants. To this respect see also Radden’s example
(2003, p. 227) ese stories have been passed down from generation to generation.
22 Angeliki Athanasiadou
4.3 Path focus → Endpoint focus
e SOURCE-PATH-GOAL schema is one of the basic image schemas that arise from
our bodily experience and perceptual interactions with the world. When we expe-
rience an object moving along a path, we can later mentally trace the path it has
followed. “It is a common experience to follow the path of a moving object until it
comes to rest, and then to focus on where it is. is corresponds to the path focus
and end-point focus transformation” (Lako, 1987, p. 442). is operation which
is based on the human ability to mentally scan a construed path (Langacker, 1990,
pp. 326–330) has a metonymic basis: one names the whole but means the part,
that is, the end. Before and aer may involve summary scanning of the scene with
the focus on the endpoint; they denote a location before or aer the endpoint of
an imaginary spatial or temporal path which is mentally traced by the speaker or
In cases of implicit and abstractly presented paths, one does not focus on the
steps followed but on the nal point. In these cases we do not focus on the proces-
sual character of the path but on the endpoint of this condensed path. By mentally
scanning paths, one ‘rests’ before or aer the path’s endpoint. By means of mental
scanning, the extended horizontal orientation of a stretch of time (23a, b) or of the
mile (23c) is construed as having a starting point and an endpoint:
(23) a. It may be many years before the situation improves.
b. It was some time before I realized the truth.
c. e house is about a mile aer the rst stoplight.
In (23a), the focus is on an improved situation reaching its endpoint; this is a static
situation and one does not need to trace a sequence of points but, by means of vi-
sual or mental scanning, refers to the situation as a whole. (23b) is an achievement,
so it only has an endpoint focusing on the realization of the truth. (23c) focuses
on locating things; one traces the route from the stoplight to the house which is
construed as having a starting point along some path toward a goal.
e above observations should not lead us to the oversimplied conclusion
that we link the senses of a polysemous entity on the basis of an abstract construct
such as an image schema or that we, as Gibbs and Colston (1995, p. 354) say, “auto-
matically access some pattern of image schemas each time we encounter a particu-
lar use of a word”. We are, however, able to recognize, and this claim is supported
by experimental work carried by Gibbs and Colston, the bond linking schematic
bodily experiences with the dierent aspects of meaning, not only physical but
also abstract and/or metaphorical.
Before and aer 23
5. Concluding remarks
In this paper I have attempted to show the various senses of before and aer trying
to nd out small steps of extension from one sense to the other. If the analysis had
been restricted to the identication of the polysemy of these lexical items, I would
only be led to the systematic organization and linking of their senses in terms of
the dierent domains they refer to. But this is part of the picture. I have shown
that before and aer each have a basic schematic meaning covering all their senses.
Moreover, specic operations such as the transformation of the horizontal axis
to a vertical one or the focus on the endpoint of a path help me process linguistic
and nonlinguistic information. e variations of the horizontality schema may
even lead to senses that only seem contradictory as a consequence of the dierent
cognitive operations that apply.
e question, however, is why would speakers of English express the relations
of anteriority and posteriority in such a variety of ways. Although one could take
the cognitive operations into account, there still remains the question of the va-
riety of preferred ways for expressing anterior and posterior relations (Time Ori-
entation, Moving Time, Moving Observer). Moreover, the evidence presented in
this paper provides only a small part of the data that might be related to how
image schemas and their transformations mediate and constitute dierent aspects
of cognitive functioning. e only plausible answer seems to be that the poly-
semy and variability of before and aer arise from our most common everyday
embodied experience of functioning in the world. We normally move in relation
to the others and others move in relation to us. We tend to link motion with time-
dening events. In such motion situations, we are typically looking ahead, either
in the direction of our motion or at things or people moving toward us. In motion
situations, those things and people that we will be coming close to, in the near fu-
ture, are ahead of us. us, motion situations involve the experiential bases for the
expressions of Time Orientation, Moving Time, and Moving Observer.
* Many thanks are due to the reviewers of ARCL for constructive comments and suggestions on
an earlier version of the paper. I am solely responsible for the views that follow.
1. e majority of the examples of the paper are taken from the dictionaries cited in the
2. is might possibly be due to the importance of the time concept. But this is a very specula-
tive observation which needs to be further elaborated.
24 Angeliki Athanasiadou
3. All these extensions of meanings have been studied from the point of view of grammatical-
ization, especially by Traugott and König (1991, pp. 189–218) and by Heine, Claudi and Hün-
nemeyer (1991, pp. 123–140).
4. (a) to (f ) are taken from Radden and Dirven, 2007, Table 9.1. Patterns of time and tense.
5. “Pre-state” and “post-state” are terms taken from Klein (1994).
6. According to SOED both before and aer were originally spatial. In particular, aer as an
adverb in OE meant ‘behind in place or order’ while as a preposition its original meaning was
‘moving in the rear of, behind’ with vbs., adj., and sbs. Of action: ‘in pursuit of, in search of’. e
temporal meaning as well as the meanings of priority or imitation and importance follow. Before
of motion meant ‘ahead, in front’ while of position or direction it meant ‘in front, in or on the
fore side’. is spatial meaning gave rise to the meaning of ‘in front of so as to be in the sight of,
the (mental) view of, the knowledge of’. e temporal meaning ‘in time previous, earlier, sooner’
7. e examples are from Lako and Johnson (1999, p. 140).
8. e meanings of ahead in SOED, which are (i) ‘at the head’, ‘in advance’, (ii) ‘in the direct
line of one’s motion’, (iii) ‘pointing forward’ and (iv) ‘forward or onward rapidly; headlong’, all
contribute to its horizontality and to the front position or movement.
Bohnemeyer, J. (1998). Temporal reference from a radical pragmatics perspective. Why Yucatec
does not need to express ‘aer’ and ‘before’. Cognitive Linguistics, 9(3), 239–282.
Boroditsky, L. (2000). Metaphoric structuring: understanding time through spatial metaphors.
Cognition, 75, 1–28.
Clark, H.H. (1973). Space, time, semantics, and the child. In T.E. Moore (Ed.), Cognitive Devel-
opment and the Acquisition of Language (pp. 27–63). New York: Academic Press.
Collins Cobuild English Language Dictionary (J. Sinclair (Ed.)). (1988). London and Glasgow,
Comrie, B. (1985). Ten s e . Cambridge: Cambridge University Press.
Ekberg, L. (1997). e mental manipulation of the vertical axis: How to go from “up” to “out”,
or from “above” to “behind”. In M. Verspoor, K.D. Lee and E. Sweetser (Eds.), Lexical and
Syntactical Construction of Meaning (pp. 69–88). Amsterdam: John Benjamins.
Evans, V. (2003). e Structure of Time: Language, Meaning and Temporal Cognition. Amster-
dam: John Benjamins.
Gibbs, R., D. Beitel, M. Harrington and P. Sanders. (1994). Taking a stand on the meanings of
stand: Bodily experience as motivation for polysemy. Journal of Semantics, 11, 231–151.
Gibbs, R. and H. Colston. (1995). e cognitive psychological reality of image schemas and their
transformations. Cognitive Linguistics, 6(4), 347–378.
Heine, Bernd, Ulrike Claudi and Friederike Hünnemeyer. (1991). Grammaticalization. A Con-
ceptual Framework. e University of Chicago Press. Chicago and London.
Johnson, M. (1987). e Body in the Mind: e Bodily Basis of Meaning, Imagination, and Rea-
son. Chicago and London: e University of Chicago Press.
Before and aer 25
Johnson, M. (1991). Knowing through the body. Philosophical Psychology, 4, 3–18.
Klein, W. (1994). Time in Language. Routledge. London and New York.
Lako, G. (1987). Women, Fire, and Dangerous ings. e University of Chicago Press. Chicago
Lako, G. (1990). e invariance hypothesis: Is abstract reason based on image-schemas? Cog-
nitive Linguistics, 1(1): 39–74.
Lako, G. and M. Johnson. (1980). Metaphors We Live By. Chicago and London: e University
of Chicago Press.
Lako, G. and M. Johnson. (1999). Philosophy in the Flesh. New York, Basic Books, 137–169.
Langacker, R. (1987). Foundations of Cognitive Grammar. eoretical Prerequisites. Volume I.
Stanford: Stanford University Press.
Langacker, R. (1990). Concept, Image, and Symbol. e Cognitive Basis of Grammar. CLR 1. Ber-
lin: Mouton de Gruyter.
Langacker, R. (1991). Foundations of Cognitive Grammar. Descriptive Application. Volume II.
Stanford: Stanford University Press.
Langacker, R. (1993). Reference-point constructions. Cognitive Linguistics, 4(1), 1–38.
Longman Dictionary of Contemporary English. (1995). ird Edition. Longman Dictionaries.
Lyons, J. (1977). Semantics. Volume 2. Cambridge: Cambridge University Press.
McGlone, M. and J. Harding. (1998). Back (or forward?) to the future: e role of perspective in
temporal language comprehension. Journal of Experimental Psychology: Learning, Memory,
and Cognition, 24(5), 1211–1223.
Miller, G. and Johnson-Laird, P. (1976). Language and Perception. Cambridge, MA: Harvard
Oxford Advanced Learner’s Dictionary. (2000). Oxford University Press.
Quirk, R. and S. Greenbaum. (1976). A University Grammar of English. London: Longman.
Radden, G. (2003). e metaphor TIME AS SPACE across languages. In Nicole Baumarten, Clau-
dia Böttger, Markus Motz, Julia Probst (Eds.), Übersetzen, Interkulturelle Kommunikation,
Spracherwerb und Sprachvermittlung — das Leben mit mehreren Sprachen: Festschri für
Juliane House zum 60 Geburtstag (pp. 225–238). Bochum: AKS-Verlag.
Radden, G. and R. Dirven. (2007). Cognitive English Grammar. Amsterdam: John Benjamins.
Svorou, S. (1994). e Grammar of Space. Amsterdam: John Benjamins.
Talmy, L. (1988). Force dynamics in language and cognition. Cognitive Science, 12, 49–100.
e American Heritage Dictionary of the English Language. (1996). Boston, New York: Hough-
ton. ird Edition.
e Shorter Oxford English Dictionary on Historical Principles (C. T. Onions (Ed.)). (1956).
Oxford at the Clarendon Press. ird Edition.
Traugott, E.C. (1978). On the expression of spatio-temporal relations in language. In J. H.
Greenberg (Ed.), Universals of Human Language (pp. 369–400). Stanford: Stanford Uni-
Traugott, E.C. and E. König. (1991). e semantics-pragmatics of grammaticalization revisited.
In Traugott, E.C. and B. Heine (Eds.), Approaches to Grammaticalization. Amsterdam: John
Yu, N. (1998). e Contemporary eory of Metaphor: A Perspective from Chinese. Amsterdam:
26 Angeliki Athanasiadou
Aristotle University of essaloniki
School of English
54 124 essaloniki
About the author
Angeliki Athanasiadou is Professor of Linguistics in the Department of eoretical and Applied
Linguistics, School of English, Aristotle University of essaloniki, Greece. She has researched
such areas of Cognitive English Grammar as questions and answers, conditionals (jointly with
R. Dirven), the setting of events in physical and abstract space, adjectives and intensiers. She
has also worked on areas of cognition such as subjectivity and on the language of emotions
(jointly with E. Tabakowska).
Annual Review of Cognitive Linguistics (), 27–54.
issn ‒ / e-issn – © John Benjamins Publishing Company
A bi-polar theory of nominal and clause
structure and function
Jerry T. Ball
Air Force Research Laboratory, Mesa, Arizona
A bi-polar theory of the structure and function of nominals and clauses is
presented in which a specier, functioning as a referential pole, and a head,
functioning as a relational pole, combine to form a referring expression. e
theory applies to both object referring expressions, in the case of nominals, and
situation referring expressions, in the case of clauses. e bi-polar theory is
contrasted with X-Bar eory — a uni-polar theory in which the head uniquely
determines the type of the larger expression in which it occurs. Uni-polar
theories adopt a strong notion of endocentricity, which is rejected in the bi-polar
theory, where both the specier and the head make signicant and meaningful
contributions to the larger expressions in which they occur. e bi-polar theory
is also contrasted with Langacker’s conception of the basic structure and func-
tion of nominals and clauses.
Keywords: nominal, clause, specier, head, structure, function, meaning,
referential pole, relational pole, referring expassion
Grammar encodes meaning (Wierzbicka, 1988). “Grammar is simply the structur-
ing and symbolization of semantic content” (Langacker, 1987, p. 12). Grammatical
variation is largely the result of a compromise between the diering requirements
for the encoding of both semantic and discourse pragmatic aspects of meaning
(Givón, 1984). “One should prefer a semantic theory that explains otherwise arbi-
trary generalizations about the syntax and the lexicon…a theory’s deviations from
ecient encoding must be vigorously justied, for what appears to be an irregular
relationship between syntax and semantics may turn out merely to be a bad theory
of one or the other” (Jackendo, 1983, pp. 13–14).
e above statements support the position that there is a close relationship be-
tween form and function, between syntax and linguistic semantics. e statement