10.Meaning, sense and usage
EPE is primarily concerned with the meanings of prepositions, not with how they pattern grammatically. For the sake of economy in wording, I use the word meaning broadly to cover all of the following:
a.the schematic visual and motoric (or kinaesthetic) mental images which speak- ers of English evidently associate with given prepositions. For example, from the standpoint of a viewer like the woman in Figure 1.1, the preposition behind will be associated with a schematic image (or images) of Thing B being on the other side of and at least partly hidden by Thing A:
Figure 1.1 From her point of view (but not from ours!) box B is behind box A
b.the notions and functional roles that speakers of English associate with a preposition. For example, behind is associated with the notion of ‘concealment,’ as in (11), while by is often used to signal that the landmark of the preposition fulfills the role of ‘agent’, as in (12):
(11)Suddenly, Metwo sensed a human approaching and leapt behind a bush.w
(12)The car was stopped by the police.w
In general I use the term meaning for big differences in meaning and sense for differences that are small. By this very rough and casual distinction, it can be said not only that to and from have different meanings from each other but also that by has different a meaning in stung by a bee than it does in a 2 meter by 2 meter rug. On the other hand, in step out of a room and roll out a carpet what we see are two different senses. Sometimes, instead of sense, I use the more specific, technical terms secondary meaning.4
4. One of the ways in which secondary meanings (a.k.a., extended or derived meanings) are thought to come into being is as follows. First, in order to express a new meaning or to express an old meaning differently, someone uses a word or phrase in a new, figurative way. Then, over time, this usage becomes ‘conventional’, which is to say that the expression’s original, figurative character
As it happens, there is controversy about how much of a preposition’s meaning comes from the preposition itself and how much comes from words that it occurs with. Either view can be taken to a problematic extreme. Thus, it is impossible that all a preposition’s meaning can come from its Subject or Landmark for in that case the book is on the table and the book is under the table would mean the same thing. In short, some meaning clearly does come from the preposition itself. On the other hand, many prepositions – let’s take under as an example – occur with thousands of other nouns in references to vast numbers of greatly or subtly different physical arrangements (e.g. under the carpet, table, cloud, ground…). Yet I would not want to say to students of English that under has thousands of meanings and senses. Some intermediate degree of description is both helpful and possible. This being said, there are cases where it is very difficult to exclude the possibility that an apparently independent sense comes from the preposition’s use in combination with a particular word or set of words. Thus, accomplishing something ‘in time’ means accomplishing it ‘before it’s too late’. It is far from certain how much of this (rather idiomatic) meaning comes from in and how much from the word time (or, indeed, from somewhere else). To be noncommittal in such cases, I sometimes avoid using the words meaning or sense, and instead say of a particular preposition that it has this or that usage (~ ‘kind of use’).
11.Literal, spatial meanings and abstract, figurative usages
The literal, spatial meanings of prepositions are the meanings we apply in order to describe physical scenes, as in this example:
(13)a. Put your drink on the table.W
Here, on ~ , where the blob represents the drink and the horizontal bar represents the table top.
Meanings like this are ‘grounded’ in our experience of the physical world, and it is clear that they are learned early in life. In this book we will repeatedly see that
is no longer noticed by the people who say and hear it. At the end of this development, a new, non-figurative meaning is stored in people’s long term memories. Even so, especially if the expression is a phrase, its original figurative conception may continue to be available in long-term memory via the memory traces for each individual word and it may continue to play a limited role in how some people understand and use the expression on particular occasions (cf., Cieslicka, 2008; Cutting & Bock, 1997; Sprenger, Levelt & Kempen, 2006). There are, though, a few cognitive linguists who believe that metaphor plays little or no role in how the different senses of a preposition (come to) relate to each other (Tyler & Evans, 2003: Chapter 2).
1 §12 Geometry, function and role 13
these physically grounded meanings underlie figurative usages like the one seen in (13b):
(13) b. We need to put the responsibility on students’ shoulders.W
Here too on ~ . That is, the blob represents ‘the responsibility’ and the horizontal bar represents ‘students’ shoulders’. Like this one, many figurative expressions involving a preposition are relatively easy to interpret in light of (one of) the preposition’s literal meanings. But there are other expressions which may require more deliberate analysis. Take, for example, the sentence, He shot himself in the foot. It could be used literally to refer to an event involving a physically real gun, bullet, foot, and injury. In this case, of course, in clearly has a literal meaning. But the sentence could also be used figura- tively, with the now conventional meaning, ‘He committed a foolish act very much to his own disadvantage’. Here, there are no physical guns, bullets, feet, or injuries. Never- theless, in itself still has the same literal meaning as before. That is, a preposition that is inside a figurative expression has not necessarily lost its literal meaning.
12.Geometry, function and role
The meaning of a preposition may include some or all of the following:
Geometrical (~ purely spatial, topological) meaning: This has to do with such purely spatial matters as whether the Subject and Landmark are near each other (e.g. close to, next to…); whether they are far from each other (beyond) or touching (against, on) or maybe near, maybe far, but not touching (above, below…); or whether the Subject is either near or in contact with the Landmark’s upper surface (over) or its lower surface (under…) or a side (alongside, beside…) or its front (in front of) or the back (behind…); or whether the Subject is near any surface of the Landmark except for the top and bottom (by) or near any surface at all (close to, near); and so on.
Functional meaning5: This goes beyond mere physical arrangement. For instance, the geometrical meaning of on in a mirror on the wall is simply that the mirror is in contact with the wall. The functional meaning is that the Subject (the mirror) is supported by the Landmark (the wall). That is, if the wall disappeared, the mirror
5. For a detailed, evidence-based account of functional meaning, and a survey of the literature, see Coventry and Garrod (2004).
would fall. As we will see, besides on certain other prepositions have a prominent functional meaning in many contexts while some others do not.
Role: Somewhat more idiomatic than function is the role that a preposition confers on a Landmark. For instance, in throw a ball to Person X, the preposition to tells us that Person X is a recipient whereas, in throw a ball at Person X, at tells us that Person X is a target.
13.The icons as aids to understanding
In this book you will see small, schematic pictures – or icons – such as the , introduced in §11 above as a representation of one meaning of on. Another example is , for toward (towardsBrE). As it happens, these two icons can be rotated to stand for a range of possible physical scenes. Thus, corresponds to a fly on the wall and corresponds to a fly on the ceiling. And Table 1.1 comments on three rotations of , as in, moving right toward the door.
Table 1.1 Alternate rotations of the icon for toward(s)
180o 90o counterclockwise 90o clockwise
e.g. Moving right
toward the door.
e.g. Moving left
toward the door.
e.g. Rising toward the exit from a mine.
e.g. Descending toward the exit from a lighthouse.
However, there are other icons which cannot be rotated in this carefree fashion. Thus, a 180° rotation of the icon for up, , creates the icon for a completely different preposition – namely, down, , which reflects the fact that the meanings of up and down are more specific about direction/orientation than toward and on are.
Many readers of the first edition have said they find these icons helpful. However,
people do differ on this point. There is evidence that some people tend to prefer words to images.6 Consequently, I have tried to create explanations which are not utterly reliant on the use of pictures and other icons. In any case, icons cannot always say a great deal about functional meanings and roles. I should add that a proportion of learners, especially very young ones, require pictures that are more
6.Katz (1982), Paivio and Harshman (1983), and Riding (1991). Experiments reported in Boers et al. (2008) relate directly to use of pictures in foreign language teaching.
detailed than most of mine. For these learners, a door may need to look a lot more like a door than the blob does in the icon . And with young learners especially, words and drawings should be complemented by physical demonstration whenever possible. Even so, icons can serve as convenient and revealing abbreviations both in class and in written feedback on learners’ written work – always provided that each icon has already been introduced and explained with reference to representa- tive physical scenes.
14. Classifying prepositions
I believe we may get a better idea of what prepositions are like by taking a quick look at some of the ways in which they may be classified. So as not to get bogged down in detail, most are presented in extremely abbreviated form.
The first of these classifications (Table 1.2) is quite traditional:
Table 1.2 Classification by domain of application
Space & time Time Space Neither space nor time
at, by, from, in, on, through, to… after, during, until… alongside, below, beside… concerning…
The classification sketched out in Table 1.3 addresses the issue of how prepositions that are somewhat synonymous may differ greatly in scope. For instance, in can be used with a far wider range of Landmarks than any of its three partial synonyms into, inside and within.
Table 1.3 Classification by specificity of meaning
General Specific
by alongside, beside, next to, in front of…
in into, inside, within
on onto, on top of
under below, beneath, underneath
As it happens, though, matters are not as neat as Table 1.3 suggests. For instance, in Come on!, on has a meaning which does not include the meanings of onto and on top of. Furthermore, a specific preposition is not always replaceable by the corresponding general preposition – e.g. crash into/in a wall.
The next classification asserts that by considering the shape of representative Landmarks we can discern four natural families of prepositions (Hawkins, 1993, with minor adaptations):
Table 1.4 Classification by shape of the Landmark
1. The Landmark is seen as an container/enclosure, space, or medium (e.g. water, air):
It’s in/out of the room. [location]
It went into/out of/through the room. [movement along a path]
It scattered papers throughout/all through the room. [distribution]
2. The Landmark is seen as a surface:
It’s lying on/off/across the carpet. [location]
It went onto/off/across the carpet. [movement along a path]
It scattered papers all across the carpet. [distribution]
3. The Landmark is seen as long and narrow:
There’s a ditch along the road. [location]
Go along the road. [movement along a path]
They scattered litter all along the road. [distribution]
4. The Landmark is seen as a point on a potential or actual path:
It’s toward/at/away from the school. [location]
We went to/from/via the school. [movement along a path]
Note that there are additional prepositions such as around that say almost nothing about a Landmark’s shape –
e.g. on a big enough map, you can draw a ring around a lake whether it is long and river-like in shape or whether it’s compact and round
Table 1.5 summarizes a classification that is similar in spirit to that in Table 1.4. It is one which I partly followed in deciding how to group prepositions for chapter by chapter discussion in this book.
Table 1.5 Classification by the relevant axis
Vertical above, under, below, beneath, underneath, on top of…
Horizontal, lateral to the viewer beside, alongside, on the right/left…
Horizontal, front-back in front of, behind, beyond…
Prepositions in more than one of these three groups across, from, in, near, out, through, to, toward(s),
with …
This factor is not applicable at, of
It is important to note, however, that for some of the prepositions listed in Table 1.5 point of view is sometimes an additional, crucial factor. In fact, it is also possible to classify prepositions according to the extent to which point of view influences their meaning. Thus, some prepositions are quite like the so-called deicticG words here, there, left and right, whose meanings depend hugely on point of view; on the left/right,
behind and in front of fall into this category (see, e.g. Figure 1.1 above and Chapter 8). At the other extreme are prepositions such as in and on top of whose meanings are substantially independent of point of view. For example, if I see that an apple is in a box from one viewpoint, it will probably seem to me to be in the box wherever around the box I stand, or if I lean over it, or even if I imagine the scene from the standpoint of the apple. In later chapters, we will see that prepositions such as up and down can be grouped in between the extremes of this continuum.
The very different classification is shown in Table 1.6 says relatively little about word meaning at all:
Table 1.6 Classification by (in)transitivity
Grammatical behavior
Example prepositions
Example of normal usage
Example of odd usage
�He crashed into.
Transitive into He crashed into a tree.
(I.e. There must be a grammatical object)
Transitive or
intransitive in
Let’s go in the house. Let’s go in.
�Go away me.
Intransitive away Go away.
(I.e. there can be no grammatical object.)
Table 1.7 returns us to the issue of meaning. Specifically, it can be helpful (both for stu- dents and teachers) to know that prepositions vary in ‘depictability’ and ‘imageability’. That is to say, the geometrical (~ purely spatial) meanings of prepositions vary in how easy they are to depict on paper or understand in terms of mental images. Hopefully, you will already be persuaded that some spatial prepositions, such as up, have spatial meanings that are substantially (but never completely) depictable – e.g. . near, on the other hand, is less depictable since the Subject can be on any side of the Landmark – including its upper and lower sides – which means that, strictly speaking, several individual depictions would be necessary to hint at the full range of scenes that near might describe. Depiction of the spatial meaning of at is problematic because it lacks imageable detail. Then, there are the prepositions of and for, which are undepictable simply because they now have no imageable spatial meanings at all. Note that the meaning of any robustly spatial prepositional is likely to consist partly of schematic (~ ‘undetailed’) motoric (~ ‘kinaesthetic’) images (see, e.g. the discussion of against in Ch. 15). And motoric images cannot be depicted on paper in any direct way.
Although of considerable practical use, the classification in Table 1.7 is complicated by the fact that some prepositions have both depictable and undepictable meanings. For instance, in the spatial expressions live by the sea and drive by the church, by has
Table 1.7 Classification by depictability
Generally easy to depict because they suggest relatively detailed images of spatial scenes. Some meanings are depictable and some are not. Several depictions are necessary to show the main options. Practically undepictable because the spatial meaning is not at all detailed. Undepictable because not spatial
around, from, in, to, toward, under… by, on… near, with… at for, of
meanings which are depictable (e.g. for pedagogical purposes). But this is not at all true of by in the passive construction (damaged by vandalsAGENT) and ‘manner’ expressions (open by turningMANNER), to give just two examples.
Table 1.8 represents another sort than any outlined so far. If fully developed, it would reveal not just which forms are acquired in what order but also which meanings of those forms. Knowing the order in which meanings are acquired might turn out to be particularly revealing about which sense(s) of any given preposition are conceptually the most basic. The research involved is especially tricky and time-consuming, however (e.g. Tomasello, 1987).
Table 1.8 Classification by order of acquisition between 1 and 8 years of age (based on Bowerman, 1996: 388, 405; Coventry et al., 2008)
Acquired earliest
Acquired latest In, on, up, down, under
Next to, beside, between
In front of, behind, in back of
above, left, right
The classification in Table 1.9 is rather subtle. If its meaning isn’t clear at this point, reading EPE should help.7
Table 1.9 Degrees of spatial/geometrical vs functional meaning (Adapted from Terzi, et al., 2007: 1)
Hang the picture on the wall. + spatial, + functional
Put down the cup.. + spatial, – functional
Bitten by a dog. – spatial, + functional
He ate it up. – spatial, – functional
7.It has been claimed that the classification shown in Table 1.9 helps explain the order in which children learn particular (meanings of) prepositions, with acquisition going from top to bottom. For references, see Terzi, et al. (2007). As far as I know, however, this claim has not been substantiated in detail. Evidence reported in Rice (1999a) suggests that things might not be so simple.
Aside from the classifications touched on above, a number of others are possible, as summarized by Tables 1.10–1.15, which I hope can speak for themselves.
Table 1.10 Classification by number of syllables
One syllable Two syllables Three syllables
at, by, for, from, of, to… across, before, over, under… in front of, alongside…
Table 1.11 Classification by lexical composition
A simple word Likely to have been regarded as a compound in the past Likely to be regarded as a compound now A phrase
at, by, for, from, in, out, over, through, under, with… before (~ by + fore), beside (~ by + side), toward (~ to + ward), … into, inward, onto, outward, throughout… in back of, in front of, on top of…
Table 1.12 Classification by number of quite distinct meanings
Few meanings Several meanings
underneath toward(s) away below against of on by
Table 1.13 Classification by frequency
High Medium Low
to, of, in, on, for, with, at, from, by, up, out* below alongside, underneath
*More or less in this order, these 11 prepositions are likely to appear in lists of the 50 highest frequency word forms of English (e.g. O’Keeffe, et al. 2007: 34–36).
Table 1.14 Classification by register (~ degree of formality)
Formal concerning, regarding…
Usable in all registers about…
Informal ‘bout
Table 1.15 Classification by source language
Germanic (Old English or Scandinavian) Latin Greek
at, by, for, from, in, off, on, out, over, through, till, to, under, with… Via French: across,
concerning, regarding…
meta, parallel to…
Direct from Latin: cum, per, qua, pro, versus, via…
15. Prototypical (~ ‘primary’ or ‘most representative’) meanings and secondary (~ ‘extended’) meanings
In §11 above, we briefly considered a common spatial meaning of on – as in, Your drink’s on the table. We can depict this meaning like this: . It may seem that this meaning is somehow more basic than other meanings which on might appear to have. To use more technical language, it may seem that this meaning is ‘prototypical’ (~ ‘conceptually basic’) compared to other meanings, which we might then consider to be secondary or extended meanings/senses. In any case, we may note that the meaning
involves contact with an upper surface and that this surface supports the Subject. In contrast, in the English town name St Leonard’s-on-Sea (and also in, I used to live on D StreetNAm), on has to do with non-supportive contact at the edge of a surface, like this: . We might therefore go ahead and decide that is more prototypical than , if it meets a number of criteria. These criteria have emerged from a stream of research in cognitive linguistics which has aimed to see if it is possible to identify a prototypical meaning for each polysemous (~ ‘multi-meaning’) preposition. Within this research stream, all of the following have been taken into consideration at one time or another:
1.The meaning recorded earliest in history, e.g. as given in the OED. (As it happens, both and are ancient.)
2.The first meaning acquired by native-speaking children. (It seems likely that would meet this criterion, but I know of no conclusive evidence that it does.)
3.The meaning which seems most grounded in physical experience. (On the face of things, meets this criterion, but again I know of no conclusive evidence.)
4.The meaning which appears to be the one that (most of) the other senses evolved from. ( meets this criterion quite well, although we will see in Chapter 3 that there may be at least one other prototypical meaning.)
5.The meaning that is most readily elicited from native-speakers when they are asked to give an example of a particular preposition. (It does seem that is relatively readily elicited, especially compared to . [Rice 1996: 156].)
6.The meaning that the preposition has in compound expressions. (Thus, one might suppose that it would be the primary meaning of on that would figure in a currently used compound like onto; and, indeed, onto does include not .)
7.The semantic relations between the target preposition and one or more other
prepositions. (For instance, a linguist might decide that the prototypical meaning for on should be consistent with the fact that on is sometimes the opposite of off; and it does seem that meets this criterion a bit better than .)
8.The degree to which a candidate prototypical meaning explains particular abstract usages of the preposition. (Thus, we might decide that is more prototypical
1 §16 Phrasal verbs (~ multi-word verbs) 21
than because it is not which is at work in a range of expressions such as depend on (money from your parents), where the Landmark [e.g. money from your parents] is a kind of abstract/metaphorical support.)8
In this book, I have not tried to identify prototypical meanings in this technical sense. However, the first meaning that is considered for each preposition is one that seems basic for pedagogical purposes – usually because it can be physically demonstrated and because it plays a role in common figurative usages. Most chapters then move on to look at what appear to be secondary literal usages/meanings and next, generally, comes discussion of noteworthy figurative and/or abstract usages.
One factor that seems especially important to me with respect to post-childhood learners of English is how easy it is to demonstrate or depict a particular prepositional meaning.9 This relates to criterion 3 above, but since I have paid so much more atten- tion to it than to some of the other criteria, I have thought it best, for the most part, to avoid the technical term prototypical meaning and use instead the less theory-bound term basic meaning – which readers should understand as meaning ‘apparently basic for pedagogical purposes and possibly also conceptually basic in the minds of many native-speakers’. (In order to strike out the words apparently and possibly, lots more research would need to be done.)
16.Phrasal verbs (~ multi-word verbs)
A phrasal verb is a combination of a verb and one or more prepositions plus possible other words in addition. A key feature of a phrasal verb is that the whole combination of words should function as a lexical unit that has its own meaning. This meaning may be relatively literal as in pick up (litter) ~ ‘gather and remove (litter)’, or not. But when students and teachers speak of phrasal verbs, it is generally the ones which are (semi) idiomatic that they have mind – e.g. the semi-idiomatic get over (a cold) and the very idiomatic put up with (bad behaviour).
8.For a fuller discussion of most of these and related matters, see Rice (1996, 1999a), Sandra and Rice (1995), and Tyler and Evans (2003, Chapter 3). On the Web, there is good material about prototype theory generally (i.e. not just relating to the different senses of one word), and biblio- graphical tips are easy to find. Eleanor Rosch is one good search term. See also Lakoff (1987, Part I) and Langacker (1987).
9.Another pedagogical criterion might be translatability. Thus, at beginner level, a teacher might privilege the meaning for which there is the most accurate L1 translation.
As it happens, grammarians tend to divide multi-word verbs into the following three classes:
–True phrasal verbs: e.g. look up a word/look a word up, in which up is considered to be a ‘particle’ rather than a full-blooded preposition.
–Prepositional verbs: e.g. look after a cat (�look a cat after), in which after is considered to be a preposition whose grammatical object is a cat.
–Phrasal-prepositional verbs: e.g. put up with bad behavior, each of which is considered to consist of a phrasal verb (put up) followed by a prepositional phrase (with bad behavior).
However, in an investigation of prepositional meaning this three-way classification is far less helpful than the classification we’ll look at next.
17.Ordinary, idiomatic phrasal verbs vs perfective phrasal verbs
One kind of phrasal verb consists of a very generic verb (get, give, put, take…) and one or more transitive or intransitive prepositions as in get over (a cold), put (a little money) by, take to (someone). These verb + preposition combinations are notoriously troublesome to post-childhood learners of English – and the more idiomatic, the more troublesome. For one thing, any given phrasal verb of this kind will be formally similar to a number of others. So get above (yourself) shares the word get with get around, get over, etc., while over occurs in get over, put one over on sb, take over and so on. Such repetition increases the chances that a learner will misremember which meaning goes with which phrasal verb – all the more so when a learner is unclear about the meanings of the prepositions. A great deal depends, therefore, on accurate, durable learning of the prepositions involved. (Hence this book.) Even so, the more idiomatic of these ‘ordinary’ phrasal verbs must still to some extent be learned one by one. Happily, recent years have seen the publication of a number of excellent phrasal verb dictionaries which can be very useful in this respect. But there are a great many phrasal verbs of another kind – so-called ‘perfective’ phrasal verbs such as cut up, close down, die out, think out, think through, and finish off. Because these expressions show a great deal of semantic regularity, learners should, potentially, be able to avoid having to look these expressions up in their dictionaries one by one. Let’s have a first look at how this is so.
The typical ‘perfective’ consists of a relatively non-generic verb (cut, close, die, think, finish…) and a preposition used intransitively (esp., up, down, out, off, & through). In these expressions, the preposition indicates that the action or transition named by the verb is definitive and/or thorough. Thus,
–‘cutting an onion up’ is more thorough than just cutting it.
–‘closing a restaurant down’ is more thorough and definitive than just closing it.
–‘dying out’ is more categorical (~ ‘broad in scope’) than mere dying.
–‘thinking something out or through’ is more thorough than just thinking about it.
–‘finishing something off’ is more definitive than just finishing it.
You will find out more about perfectives in the chapters where the relevant prepositions are discussed.
18.Prepositions, directional adverbs and particles
Some words, such as cow/s, are used both generally and specifically. In its general sense, the word cows can refer to a mixed group cows, bulls, heifers, calves and so on. In its specific sense, it refers to the adult female of the species. In this book, I use the word preposition in much the same way – often generally, sometimes specifically. When the issue is meaning rather than syntactical patterning, the term preposition covers all of the following: prepositions proper, directional adverbs, and particles. In fact, with respect to meaning, even adjectives and verbs may have prepositional character (e.g. an in joke and to down a plane). Occasionally, following others (e.g. Jackendoff and Landau, 1991), I use the blanket term intransitive preposition instead of directional adverb (Go away). This term is a good reminder that directional adverbs such as away may be very prepo- sitional in terms of meaning. One might object that the incorrectness of a sentence like (12a) indicates that away should be called something other than a preposition since the noun phrase that follows it (the car) is not in fact its Landmark.
(12) a. �Step away the car.
But if we consider matters in more detail, the prepositionality of away becomes evident. Take the following sentence, for example:
(12)b. •Step away from the car.W
The complete text (not given here) makes it clear that the speaker of (12b) is a policeman and the Subject (i.e. the person being addressed) is the driver of a car. ‘The car’ is not only the explicit Landmark of from but also the implicit Landmark of away. To see that this is so, we may need to consider the scenario step by step:
1.Just below, I have put ‘the car’ and ‘the driver’ together in a box in order to show that at the beginning of the scenario the driver is either in or very near the car:
2.If the driver obeys the order to step away, the physical scene changes, so that now the car and the driver are separated, like this:
If we assume that it is the driver, not the car, that moves, then we can pictorially sum- marize the scenario like this: . The blob stands for the car, the arrow depicts the path of the driver, and the dots say that the path began either exactly at the car or near it. All in all, there is a Subject, a Landmark, and a path. And the meaning of away is therefore prepositional in this case.
The shorter expression Go away! can be analyzed in much the same manner although in face-to-face communication the implicit Landmark, or point of reference, is likely to be the Speaker. If it is, the sequence of scenes is as follows:
At time 1:
At time 2:
Summary:
Again, the box in the depiction represents the Landmark and the arrow the path traveled by the Subject.
All other literal uses of away can be understood in a similar fashion. Here is another example:
(13)a. A pet cat or dog can drive away birds.W
Here, birds is the direct object of drive. In fact, one could rephrase (13a) as (13b):
(13)b. A pet cat or dog can drive birds away.
This makes it easier to see that, even in (13a), ‘birds’ is the Subject of away while the implicit Landmark of away is the birds’ initial location. Like this:
At time 1:
At time 2:
Summary:
In short, away is so much like a preposition (of path) that always calling it some- thing else – e.g. a ‘directional adverb’ or a ‘particle’ – could obscure the fact that in meaning away is quite prepositional. All this being said, sometimes it is fruitful to be aware of the preposition-adverb-particle distinction, even with respect to meaning – as Kreitzer (1997: 315) makes clear.10 In his view – with respect to their imageable,
10.My terminology is rather different than Kreitzer’s, however.
geometrical meanings – prepositions, particles and directional adverbs tend to differ as follows:
–A preposition (in the strict sense of the term) has a meaning which is relatively full and detailed, with Subject and Landmark explicitly stated – as in I jumped over the fence.
–A particle is not explicit about one component, the Landmark. A typical example is I turned the book over/I turned over the book where the Landmark is unstated. In fact, what the Subject (‘the book’) is turned over, is its own center of gravity, and the Landmark is therefore a location inside the Subject – at least in the case of over (see Ch. 9, §3.2.4).
–A directional adverb – e.g. I came over yesterday – says something about the shape
or direction of the path but other details (e.g. starting and end points) must be inferred.11
It seems, incidentally, that no one has conclusively solved the puzzle of why particles sometimes come right after their verb and sometimes they don’t (e.g. turn a book over vs turn over a book). So many factors appear to be involved that there is not nearly enough room in this book for an adequate survey (see Gries, 1999). That being said, Gries (1997:64; cited in Dirven, 2001: 48) has made the following proposal which seems to provide a large part of the answer:
–The verb+particle+np construction is preferred when the NP demands extra attention (on account of its high information value) – e.g. when the NP brings new information into the discourse (e.g. turn over the book instead of the…).
–The verb+np+particle construction is preferred when the NP demands rela- tively little attention – e.g. when the NP is a pronoun bringing little or no new information into the discourse (e.g. turn it over).
This tendency to place information-rich expressions in clause final position is actu- ally quite general in English – e.g. I gave it to him vs I gave him the book he had always wanted. This must be one of the reasons why particle placement does not seem to be a major problem for learners of English. For them a much bigger problem is learning phrasal verbs in the first place! (For a good short introduction to these matters see Dirven [2001], findable on the Web.)
11.All but one of the example sentences are Kreitzer’s: I added I turned over the book. Unlike Kreitzer, I do not describe the meaning of a particle precisely in terms of Subject and Landmark identity.
19. Prepositions and the guessability of idioms
One of the defining characteristics of an idiomatic phrase is that its overall meaning cannot be reliably guessed from the meanings of the individual words that make it up. Idiomaticity – and its opposite, transparency (~ ‘easy guessability’) – are in part personal matters. Person A might be able to guess the meaning of a phrase that Person B cannot, and vice versa.
Importantly, the class of ‘Idiom’ is graded along a continuum, as indicated in Table 1.16.
Table 1.16 Degrees of idiomaticity & transparency
idiomatic (the overall meaning is hard to guess from component the words)
4. make up after a quarrel
3. break up some ice
2. fill up a bag
1. climb up a tree
transparent (the overall meaning is easy to guess from the component words)
Let’s consider the four examples in Table 1.16 from the bottom up:
1.Climb up a tree seems highly guessable by anyone who knows the basic literal meaning of each of the four component words; in fact, just knowing up and tree might give learners all the information they need for an accurate guess.
2.Much the same can be said of fill up a bag, although a learner might well wonder why (perfective) up is included at all since, if you fill a container, the level of its contents seldom goes down.
3.Break up is similar to fill up except that it is much more difficult to see how up can have anything to do with the meaning that we see in (1) and (2).
4.Make up is the most idiomatic because knowing the common senses of make and also the spatial and the perfective senses of up is unlikely to help a learner guess that make up with sb means ‘re-establish good personal relations with sb’.
One of the central aims of this book is to help both learners and teachers of English to become more successful at using a knowledge of prepositions as a key for ‘unlocking’ the meanings of idioms in which prepositions occur. Sometimes, of course, just knowing the preposition isn’t enough. In phrasal verbs, for example, the meaning of the verb is very often more elusive than the meaning of the preposition. We see this in Table 1.16, Example 4; here, if only we could figure out what make means we would probably find that up has its normal perfective meaning. In a number of other expressions, it is the
1 §20 The roles of functionality and metonymy 27
pronouns it and one which hide meaning. Take, for instance, the expression put one over on sb (~‘fool sb’). What does one refer to? One what? If we knew that, this expres- sion would almost certainly become more guessable. Likewise, pull it off (~‘succeed despite a good chance of failure’) would doubtless be more guessable if we knew what ‘it’ was. An additional source of idiomaticity in pull it off is the absence of a landmark for off. The question now is off what? If we knew the answer to this question too, this expression might hardly be idiomatic at all.
Here and there throughout the book, we will consider similar cases where idioma- ticity comes mainly from the vagueness or absence of a (full) noun rather than from the preposition. Fortunately, or so we are informed by the authors of a comprehen- sive study (Grant & Bauer, 2004), only a few more than 100 of English idioms can be regarded as completely unguessable – and not all of them include prepositions. The rest, which Grant and Bauer call ‘figuratives’, are metaphorical and/or metonymicG – which means that learners of English should have some chance of eventually under- standing why any of these idioms has the meaning that it does even when they cannot guess the meaning from scratch without help. And this is good because learners who can see how a particular figurative meaning works, have a better chance of remember- ing the idiom in question (Boers & Lindstromberg, 2009: Chapter 5). While a good understanding of English prepositions is not the only key to learning and remember- ing idioms, it is certainly a very important one – particularly because prepositions are so common.
20.The roles of functionality and metonymy
With reference to the scene depicted in Figure 1.2, we might very naturally say that the chair is under the table.
Figure 1.2 A chair under a table (After Vandeloise, 1986: 19)
There are certain facts about this use of under12 which, in everyday life, we would generally overlook. Firstly, not all the chair is directly down from the table; some of it is off to one side. Secondly, some of the chair is actually higher than the table. Thirdly,
12. Vandeloise (1986) is concerned most directly with French prepositions, but what he says in this case is as applicable to English under as it is to French en dessous.
no part of the chair is lower than the lowest part of the table (i.e. both the table and the chair go right down to the floor). Why, then, do we use under? Almost certainly, we do so because the part of the chair that most directly supports a sitting person – the seat – is entirely lower than and almost completely directly down from the part of the table that we are normally aware of using – the top. In other words:
–We use under here to describe the arrangement of the elements that really matter to us (the chair-seat and the table-top) rather than all the elements of whatever functionally unrelated importance.
–The words table and chair are whole→part metonymsG (~abbreviations) for the top of the table and the seat of the chair (see Vandeloise 1986: 19–20, 48, and passim).
Metonymy and considerations of function are ubiquitous factors in language generally and in our use and understanding of English prepositions in particular. It is interesting though – and perhaps encouraging – that learners of English can find the functional meanings of English prepositions less difficult to grasp than the geometrical meanings (Coventry et al., 2008). And my experience is that they largely take metonymy in their stride as well, although a learner’s mother tongue (to name just one ‘learner variable’) may well influence the extent to which this is so.
21.Major non-spatial notions
In the field of TESOL the English preposition system was long under appreciated. Strangely so, not just because prepositions play a crucial role in specifying relationships of space and time, but also because they express many non-spatial notions which can be every bit as important as the notions expressed by verbs. Here are a few examples:
Abstract notion Example expression Agent written by Shakespeare
Cause die of/from TB
Function for sitting on
Instrument with a hammer
Manner with flair
Means by hand, through hard work
Purpose went to buy bread
These and other notions are discussed at different points in this book in connection with individual prepositions. Many of these scattered comments are brought together in summary form in Chapter 21.
Chapter 2
Toward(s), to, in/into, inward, outward, through, out (of), from (vs off),
away (from)
1.Overview
This chapter provides a first look at the meanings of several prepositions of direction and course of path; that is to say, the meanings covered in this chapter are fundamen- tally dynamic. Later chapters give additional details about most of the prepositions covered here (e.g. Chapter Four covers non-dynamic senses of in and out).
toward(s) to from away (from)
in/into
inward(s)
through
outward(s) out (of)
Figure 2.1 Preliminary overview of meanings
Opposites:
Toward(s) ≠ away (from); to ≠ from; in(to) ≠ out (of); inward ≠ outward
2.Toward(s)
2.1The meaning of Toward(s)
Toward or (especially in BrE) towards, means ‘nearer & nearer, in the direction of ’. The Landmark is not necessarily the endpoint of the path since the path may never reach the endpoint. For example, a sentence like the following is possible:
(1)She…started toward the house, but then turned toward the barn.W
2.2The suffix -ward(s)
-ward(s) can go on the end of several other prepositions, e.g. in-, out-, down-, up- and on- and dozens of different nouns, e.g. home-, sky-, land-, and west-. Home- ward, for example, means ‘in the direction of home’. One may even find humorous creations such as the following:
(2)He tottered off pigwards. [= ‘in the direction of the pig’] [P G Wodehouse. 1924. Summer Lightning, p. 24 (Barrie & Jenkins)]
All -ward(s) words, including toward(s), inward(s), and outward(s), can be used in expressions having to do with:
–orientation, e.g.
downward…
facing toward x, directed inward, turned outward, pointing
–direction of gaze, e.g. looking toward x, looking inward/outward/upward/skyward/ homeward…
–virtual movementG, e.g. a tree leaning toward the house, a route leading
northward…
We will return to toward(s), inward(s), and outward(s) later in this chapter.
3.To
3.1To: Basic, meaning
To specifies the endpoint of a path. If I say I went to their house, I mean that the end- point of my path was their house, that’s all. To says absolutely nothing about whether, when I arrived, I went in their house or whether, for instance, I just went to the front door, knocked, waited, and left.
3.2To ~ ’toward’
According to the OED, in early Old English to generally meant ‘toward’ while a now extinct preposition (oth) meant ‘to’. As it happens, to means ‘toward’ in a few contem- porary expressions – e.g. to the left/right/rear/east.
See Chapter 20 for more about to. The index lists additional locations, where to is contrasted with other prepositions such as at, of and for.
2 §4 In/inTo 31
4.in/into
4.1In/inTo vs To
4.1.1Basic differences
We use in or into when we think of the Landmark as something with a boundary and an interior – like this, – regardless of whether it is 2- or 3-dimensional. If we use to, however, or if we hear it, we do not think of the Landmark in this detailed way. Instead, we think of its being like a point or blob – perhaps something like this, . Thus, if we hear someone say (3) below, our knowledge of trees tells us that it is unlikely that the tree is entered in any sense at all.
(3)It took you 10 seconds to run to the tree and back again.W
But if we hear someone say (4), our knowledge of libraries tells us that she probably did enter it.
(4)She went to the library.
Using into in either (3) or (4) would force the understanding – surprising in the case of (3) – that the Landmark was entered. Regarding (3), we would then have to imagine that the tree was hollow and had an entrance of some kind.
4.1.2The difference in the scale of mental images for in/inTo and To
Apart from the question of whether a Landmark is normally enterable (like a library) or normally not enterable (like a tree), we can often use either in(to) or to with refe- rence to a given Landmark depending on the scale of our mental image of it. For example, if people are relatively close to a large city such as Chicago, they might say:
(5)a. I’m going into town/the city (to see the sights).
But if they live, say, 100 miles away, they are much more likely to say:
(5)b. I’m going to Chicago (to see the sights).
This is doubtless because, in everyday life, the boundaries and interiors of things become harder to see the farther away from these things we are. Here is another pair of examples;