Compositionality and cognitive semantics

Clarifications. Modes of combination. Cognitive semantics. Extensions of meaning # Gramática. Semántica

  • Enviado por: HaRPy
  • Idioma: inglés
  • País: España España
  • 24 páginas




Compositionality or syntagmatic delimitation of lexical units (Cruise, 1986. 2000) is a basic principle in grammatical semantics which claims that single meanings combine together to form more complex meanings. That is to said, we must interpret utterances from our knowledge of the meanings of simple expressions and the constructions used in combining them.

Its strong version (Cruse, 2000) says that “the meaning of a grammatically complex form is a compositional function of the meanings of its grammatical constituents”.

The traditional view is that the relationship is fairly tight: the meaning of a complex expression is fully determined by its structure and the meanings of its constituents—once we fix what the parts mean and how they are put together we have no more leeway regarding the meaning of the whole. This is the principle of compositionality, a fundamental presupposition of most contemporary work in semantics.

  • “a red hat” = a + red + hat

  • “the girl is playing football” = the + girl + is + playing + football

Proponents of compositionality typically emphasize the productivity and systematicity of our linguistic understanding. We can understand a large—perhaps infinitely large—collection of complex expressions the first time we encounter them, and if we understand some complex expressions we tend to understand others that can be obtained by recombining their constituents. Compositionality is supposed to feature in the best explanation of these phenomena. Opponents of compositionality typically point to cases when meanings of larger expressions seem to depend on the intentions of the speaker, on the linguistic environment, or on the setting in which the utterance takes place without their parts displaying a similar dependence. They try to respond to the arguments from productivity and systematicity by insisting that the phenomena are limited, and by suggesting alternative explanations.

If we know what the words mean, using our knowlegde of syntactic and semantic rules we can work out the meanings of sentences, even ones we have not previuosly encountered. The examples are based on Marchand (1969):

watchmaker writing table speech-writer

bookseller waiting room ironing board

1.1 Clarifications

There are many theses called `the principle of compositionality'. The following can serve as a common reference point:


The meaning of a complex expression is determined by its structure and the meanings of its constituents.

Important variants of the compositionality principle will be presented below in a form most similar to (C) to facilitate comparisons. When formulating more precise versions it is crucial to keep the pre-theoretical intuitions that led many to accept compositionality firmly in mind.

1.2 Languages

The principle of compositionality is normally taken to quantify over expressions of some particular language L:

(C )

For every complex expression e in L, the meaning of e in L is determined by the structure of e in L and the meanings of the constituents of e in L.

Questions of structure and constituency are settled by the syntax of L, while the meanings of simple expressions are given by the lexical semantics of L. Compositionality entails (although on many elaborations is not entailed by) the claim that syntax plus lexical semantics determines the entire semantics for L.

It makes a big difference whether L is a natural or an artificial language. Syntactic and semantic questions about a natural language are settled by and large through empirical investigation; syntactic and semantic questions about an artificial language are settled usually by checking what the appropriate stipulations are. Prima facie, natural languages might turn out not to be compositional, whereas many artificial languages were designed to meet such a requirement. (Compositionality is a bonus when it comes to proof-checking in computer languages, or inductive proofs in logical calculi.) Unless explicitly noted, talk of compositionality is to be taken as talk of compositionality of some particular natural language, or of natural languages in general.

If thought is a kind of language, we can raise the question whether it is compositional. Thought would not have to be much like Swahili or the language of set theory for the question to make sense, but we do need the assumptions that thoughts have meanings (and so, presumably, are not themselves meanings) and that they have meaningful constituents. These assumptions follow from the language of thought hypothesis. Those who reject this hypothesis may still speak of the compositionality of thought—but only in an extended sense.

What would such an extended sense be? The key to generalizing compositionality for non-linguistic representational systems is to relax the syntactic ideas of `constituent' and `structure'. Consider, for example, the No-Left-Turn sign:

'Compositionality and cognitive semantics'

This could be viewed as a complex sign decomposable into meaningful features—the shape, the color pattern, the arrow, etc. These features are the analogues of simple expressions: they appear in many other complex signs and they appear to contribute more or less uniformly to their meanings.

'Compositionality and cognitive semantics'
'Compositionality and cognitive semantics'
'Compositionality and cognitive semantics'
'Compositionality and cognitive semantics'
'Compositionality and cognitive semantics'

Once we have an initial grip on what counts as a constituent and how those constituents compose, and once we settle on a sensible way to articulate meanings for traffic signs, we can legitimately raise the question whether this system of representations is compositional. We may even be able to answer it.

1.3 Meaning

The principle of compositionality is not committed to a specific conception of meaning. In fact, it is frequently announced as a principle that is applicable to whatever a semantic theory might assign to expressions of a language. Furthermore, although the reference of an arbitrary expression is definitely not something one would normally call its `meaning', versions of the following principle are frequently called `the principle of compositionality':


For every complex expression e in L, the reference of e in L is determined by the structure of e in L and the references of the constituents of e in L.

(I use the word `reference' here roughly the way Frege used his `Bedeutung' after 1892. But it could be taken the way Barwise and Perry use `reference', the way Lewis uses `extension', or the way Kaplan uses `content'. The differences are significant, but they do not matter for present purposes.) To avoid confusion, we should call this the principle of compositionality of reference, and (C) the principle of compositionality of meaning; when I speak of compositionality unqualified, what is meant is always the latter. Since the arguments in favor of compositionality tend to be based on general considerations about linguistic understanding—which, I shall suppose amounts to nothing more or less than understanding what linguistic expressions mean—proponents of (Cref) have a choice to make. They can advocate (Cref) on different grounds or they can claim that an appropriate theory of the sort that assigns (relative to a variety of factors) references to expressions can serve as a theory of meaning.

1.4 Structure

Much of what was said above about the need to constrain what counts as meaning applies to structure as well. Janssen (1986) has a proof that we can turn any meaning assignment on a recursively enumerable set of expressions into a compositional one, as long as we can replace the syntactic operations with different ones. If we insist—as we should—that any acceptable semantic theory must respect what syntax tells us about the structure of complex expressions, this result says nothing about the possibility of providing an adequate compositional semantics; cf. Westerstahl (1998).The moral of Janssen's result is that although commitment to compositionality requires allegiance to no particular sect of syntacticians, one cannot be oblivious to syntactic evidence in semantic theorizing.

(C) does not require the kind of tight correspondence between syntax and semantics we intuitively associate with compositionality. To illustrate this, consider a view, according to which the meaning of a declarative sentence s is the set of possible worlds where s is true. According to such a view, tautologies are synonymous, even though (since Rudolf presumably has some tautological beliefs and lacks others) sentences resulting from embedding tautologies under `Rudolf believes that…' are not. Intuitively, this is a violation of compositionality. Still, the semantics is not in conflict with (C): tautologies might differ structurally or in the meaning of their constituents, which could explain how their embedding can yield non-synonymous sentences; cf. Carnap (1947) and Lewis (1970).

1.5 Arguments for compositionality

The simplest argument for compositionality is that it is supported by intuitions many claim to have about meaning and structure. Although there are interesting putative counterexamples, they probably can be explained away through modest revisions of our syntactic and/or semantic theories. This defense is reasonable but much too modest. For even if it succeeds in convincing some who aren't already convinced, it leaves us all in the dark why compositionality is true. Defenders of compositionality should do better than this.


Two importants concepts are related to compositionality: determination and predictability: the meaning of a complex expression is determined by the meanings of it's constituents and predictable by general rules from the meanigs of it's constituents.

There is a rationale which explains this principle:

The first is that a language has an infinite number of grammatical sentences; the second is that language has unlimited expressive power, that is, anything which can be conceived of can be expressed in language. Besides these, it is a fact that our mind can not store this infinite inventory of sentences and expressions that language can produce.

2. Modes of combination

Cruse (2000) views two basic modes of combining two meanings to make a third: additive and interactive.

Common examples of additive modes are simple syntactic co-ordinations:

  • [A manageress and two clerks] entered the meeting room and sat down.

  • My son is [tall and fair].

In interactive combinations, at least one of the constituents is modified. Two types of interactive combinations can be identified:

  • Endocentric, where the resulting meaning is of the same type as one of the constituents.

  • Exocentric, where the resulting meaning is of different type to either of the constituents.

2.1 Some endocentric combinations

It is the most basic one and can be illustrated by the example “black shoes”. Which are things that are simultaneously shoes and black.

Now look at some combinations of a noun modified by an adjective, constructions that are grammatically exactly the same:
bad apple, cold turkey, dark horse, dead meat, dirty linen, heavy metal, hot potato, white elephant ...
Most of them (though not all) can have a literal endocentric meaning, that is we can really mean "the apple is bad, the elephant is white" and so on. But all these are much more commonly used in a metaphorical, exocentric, and therefore unpredictable, meaning.

2.1.2 Relative descriptors

This is a more complex interaction between meanings. It can be illustrated by “blue pine-tree” and it can not be glossed “something which is “pine-tree” and “blue”. “Blue” must be interpreted relative to the colour of the class “pine-tree” and means something like “significantly more bluish green than the average green pine-tree”. “pine-tree” determines how “blue” must be interpreted and “blue” limits the application of “pine-tree”.

2.1.3 Negational descriptors

Here, the effect of the modifier is to negate the head, while at the same time it indicates which the real referent is.

  • A former Governor

  • An ex-singer

  • An imitation Picasso painting

A “former Governor” is not at the same time “Governor” and “former”. If it is “former” it is not anymore “Governor”. However, “Governor” and “former Governor” belong to the same basic ontological class, thus being an interactive combination.

2.1.4 Indirect types

Indirect combinations require a more complex compositional process. If we take the example of “a beautiful dancer” we find that this phrase is ambiguous. One reading denoting someone who is simultaneously beautiful and a dancer. But the other reading requires some reconstruction of the meaning so that beautiful becomes an adverbial modifier of the verbal root dance, being “someone who dances beautifully”.

2.2 Exocentric combinations

Are of a radical different ontological type from that of any of the constituent meanings. An example would be the preposition on which denotes a relation and a noun phrase like the table which denotes a thing. When combined, they produce the prepositional phrase on the table which denotes a place, thus being of a different ontological class.

The exocentric compounds show the metaphorical richness of the language, so let's dig into them a little further. Only this time, we'll look at a slightly different type of compound, the ones made up of Adjective + Noun.

We can paraphrase Adjective + Noun compounds by saying "a bird that is blue", or using A = Adjective and N = Noun, "an N that is A". Try this now with high chair, lowlands, bad mood, wildfire.

Now let's turn to the other type of combination, Adjective + Noun phrases. We can paraphrase a phrase of this kind as "the N is A": try
good idea, green grass, heavy stone, hot weather.

3. Restrictions to compositionality: conventionality and creativity

This is a process rooted in individual psychology, called routinization. If two or more actions are often performer together, we make them into a routine, that is, we create a package of automatized parts for which a single decision is sufficient.

Conventionalization produces cases like the idioms “kick the bucket” or “paint the town red” which are not predictable from the meanings of their constituents, thus restricting compositionality. The effects of conventionalization go even further, since languages are full of expressions which have come to mean something that is in principle compatible with the compositional meaning, but is in fact something much more specific.

Creativity can be considered another restriction to compositionality. Linguistic communication always relies on extra-linguistic knowledge. Using it is in fact a precondition for linguistic creativity: without it, a speaker could not rely on a listener's capacity to read `between lines' .

The following examples show this necessary `creativity':

There was no fish pudding left on the kitchen's table. The cat climbed down from it licking its mouth.”

“Some money disappeared from the Bank. The director was seen the following day flying to Brazil.”

The result of conventionalization is often that an expression comes to be used routinely in certain kind of situations, thus leading to a loss of expressive power.

Expressions like:

“You are welcome”

“I would be so grateful if you…”

“As far as I know...”

are so routinized and grammaticalized that they have lost their original meaning. It is here that creativity comes in, when speakers try to find new ways of saying things.

4. Semantics constituents and the “recurrent contrast test”

Is based on the assumption that meaning implies choice. An expression cannot have meaning unless it was chosen from a set of possibilities. The “recurrent contrast test” states that semantic constituents can be substituted by some other constituent belonging to the same grammatical class, therefore giving a different meaning.

Example: in “the girls play tennis”, “tennis” can be substituted by “marbles” changing the meanings of the phrase, so, the recurrent contrast test would have been passed. However, in “I want to read a book” “to” cannot be substituted since it is an obligatory element.

These constituents need to be minimal, i.e. that they cannot be divided into smaller parts.

5. Limits to compositionality

Is not a universally valid concept. When analysed compositionally, its resulting meaning does not coincide with the meaning of its individual constituents. Phrases like spill the beans or kick the bucket cannot be decoded compositionally, since they do not have semantic constituents, although it might seem so.

Transparency and opacity are the two extreme points of a continuum of degrees. Following Cruse (1986) there seem to be two factors to this notion of degree of opacity and/or compositionality. The first deals with the question: are constituents of opaque expressions full semantic indicators? Blackbird with two full indicators is les opaque than ladybird with one partial indicator only (-bird) which in turn is less opaque than red herring. The second factor is the discrepancy between the combined contribution of the indicators, whether full of partial, and the overall meaning of the idiom. This is difficult to measure, since some “irreversible binomials”, also known as “restricted collocations like fish and chips are less opaque then, blackbird, even though both contain only full semantic indicators.

5.1 Indicators, tallies and categorisers

There are certain types of element which fail the recurrent contrast test but are not words. They have semantic relevance and cannot be dismissed. It is the case of cran-, rasp-, goose-, of cranberry, raspberry and gooseberry. They do not carry any meaning at all.

One of the challenges that a morphologist faces is providing an analysis that distinguishes what is of purely arcane historical interest from what is relevant in an analysis of the contemporary language. What is in principle should be compositional for some native speakers may not be for all speakers. The classic example of this is provided by the cranberry words. These words highlight the problem of treating morph-like units that do not belong to any recognisable morpheme in the language as it is used today. A cranberry is a kind of berry. So, we can identify berry as a morpheme in that word. The only loophole available is to treat cranberry as a compound word made up of the free root morpheme and the bound root morpheme cran- which has the peculiarity of only occurring as part of the word cranberry. They are termed semantic tallies and their partner elements (-berries), semantic categorisers.

Another category which are not constituents but have a semantic function, are: semantic indicators. It is the case of black- and bird- in blackbird or green- and house- in greenhouse. Other cases could be: im- of impertinent, impudent; dis in disappoint, disgust, etc

5.2 Idioms

An idiom is an expression whose meaning is not compositional-that is, whose meaning does not follow the meaning of the individual words of which it is composed. For example, the English phrase to “kick the bucket” means to die. A listener knowing the meaning of “kick the bucket” will not thereby be able to predict that the expression can mean to die. Idioms are often, though perhaps not universally, classified as figures of speech.

Idioms typically admit two different interpretations: a literal one and a nonliteral (or figurative) one.

Continuing with the previous example, the phrase to kick the bucket can, in fact, refer to the act of giving a kick to a bucket, but this interpretation is usually not the intended one when a native speaker uses the phrase. This aspect of idioms can be frustrating for students of a new language.

Idioms are often colloquial metaphors. The most common ones can have deep roots, traceable across many languages. Many have translations in other languages, some of which are direct. For example, get lost! — which means go away or stop bothering me—is said to be a direct translation or calque from an older Yiddishs idiom.

While many idioms are clearly based in conceptual metaphors such as "time as a substance", "time as a path", "love as war" or "up is more", the idioms themselves are often not particularly essential, even when the metaphors themselves are. For example "spend time", "battle of the sexes", and "back in the day" are idiomatic and based in essential metaphors, but one can communicate perfectly well without them. In forms like "profits are up", the metaphor is carried by "up" itself. The phrase "profits are up" is not itself an idiom. Practically anything measurable can be used in place of "profits": "crime is up", "satisfaction is up", "complaints are up" etc. Truly essential idioms generally involve prepositions, for example "out of" or "turn into".

To let the cat out of the bag

This idiom was probably originated in English marketplaces many years ago. Traders would put a cat in a bag and would try to deceive possible customers by saying it was a pig. People would buy the bag without close inspection of the contents, but then they would let the cat out of the bag and realize that they had been deceived, which is the meaning it has nowadays.

To be born with a silver spoon in one's mouth

This expression is used to refer to a member of a wealthy family. It was probably originated from an old custom whereby godparents used to present the child at the christening with a silver spoon.

Never look a gift-horse in the mouth

This is an old saying and it's related to the way that a horse's age is calculated: by looking at its teeth. The expression now means that you should accept a received gift without trying to find a fault in it.

Dressed to the nines

This may come from an alteration of "dressed to the eyes", which was written in Old English as "to then eyne". The letter n of then was removed and moreover, eyne was changed to nine. Another explanation is that the number ten is considered the ultimate point of perfection. Therefore, if someone is dressed to the nines, he must look really smart.

It's raining cats and dogs

This is a very old expression linked to the ancient beliefs of sailors and also to Norse mythology. Cats were associated with heavy rain and dogs with storms and wind. Therefore, this expression was used to refer to severe rainstorms. Nowadays, it has become old-fashioned.

To catch someone red-handed

This meaning of this idiom is "to find someone when he's doing something wrong". It alludes to the discovery of the murderer so soon after committing the crime that blood is still on his hands.

To bury the hatchet

This comes from the American custom of burying hawks and other weapons as a sign that hostilities between the American Indians and the White had ended. Nowadays, this idiom is used to refer to coming to peaceful terms with an opponent.

5.3 Frozen metaphors

Frozen metaphors are a kind of idiomatic expressions which have a certain degree of compositionality, in the sense that synonym substitution is possible, without destroying the non-literal reading. They keep, however, some syntactic frozenness typical of idioms.

  • The ball's in your court now

On your side of the net

  • I can read her like an open book


  • He has one foot in the grave

Both feet tomb

Are units spelled in the form of non-contiguous sequences of tokens (e.g. Take…into account), and their recognition requires computation traditionally performed by syntactic parsers. Over 30,000 French frozen expressions have been described in the tables. We show how to use these tables to automatically construct FSTs that are capable of recognizing and tagging frozen expressions in text. Representing the result as tags poses some formal problems that we discuss.

5.4 Collocations

The collocation of a word (i.e. the combination that a word enters into) plays a major role in determining its meaning in context. For instance, red is interpreted differently in red rose, where it is the flower that is red, as opposed to red grapefruit, where it is the flesh of the fruit that is red, and red apple, where it is the skin of the fruit that is red.

Collocations, like idioms, need to be individually learned. This type of non-compositional item must be seen both from the side of the speaker/writer/sender and from the side of the hearer/reader/recipient. They are important when focused semantically, since their meaning must be decoded non-compositionally.

Examples: heavy rain, high speed, high wind, severe frost, but not

*great rain, tall speed, heavy wind, deep frost.

5.5 Clichés

A cliché (from French name cliché) originally was a printing term for a semi-permanently assembled piece of type which could easily be inserted into the document being printed. It has come to mean a phrase, expression, or idea that has been overused to the point of losing its intended force or novelty, especially when at some time it was considered distinctively forceful or novel. By extension, "cliché" applies also to almost any situations, subjects, characterisations, or objects that have similarly become overly familiar or commonplace. Their meanings may also be misunderstood leading to them being often misused. As a result, many feel that they should not be used and are seen as an indicator of lack of creativity, innovation, or sincerity. Because the novelty or frequency of an expression's use varies between different times and places, whether a given expression is a cliché depends largely on who uses it and who makes the judgment.

On the other hand, there can also be advantages to using clichés. The use of a cliché that is well known to the audience can help keep the storytelling on a fast pace without as much explanation and elaboration. They can also help in connecting with the audience by showing them something with which they are familiar or can relate to.

Some characteristics are:

  • They don't need to be decoded globally.

  • These phrases are stored as complete units in the brains, being easy to retrieve and easy to decode.

  • Their propositional features of truth or falseness are never seriously examined, funtioning as discourse markers.

Many examples are: `I'm sure about this', `Ladies and Gentlemen', `I won't take no for answer', `as far as I'm concerned', `Behind the clouds, the sun is shining', `I will survive', `Every rose has its thorn', `Truth is nothing but a feeling that something is true', `The truth is just as meaningful as the lie', `Making a bad decision is better than making no decision at', `When the work is done, I will have time for myself', `You get nothing for free', `No pain, no gain', `It'll feel better when it stops hurting', `Goodbye is not an easy word to say', `I will always remember you', `Parting is such sweet sorrow', ` The sun in your eyes made some of the lies worth believing', `There will be another one', `I am better off alone', `I love being free, it's the best way to be', `There are plenty more fish in the sea', `Love hurts', `One lost, ten found'.

6. Phonetic elicitors of semantic traits

There are some phonetic sequences, though, which have semantic value without being grammatically valid. They seem to go straight from sound to meaning, bypassing grammar.

One term is termed onomatopoeic phonetic sequences,(n'mt'p') is a figure of speech, that employs a word, or occasionally, a grouping of words, that imitates, echoes, or suggests the object it is describing, such as "bang", "click", "fizz", "hush" or "buzz", or animal noises such as "moo", "quack" or "meow". They are also a very common feature of comic strip writing, where words such as "Pow", or "Ka-pwing" help the reader to better imagine what is being described, and make up for the lack of literary description.

Onomatopoeic words exist in every language, although they are different in each. For example:-

In Latin, tuxtax was the equivalent of "bam" or "whack" and was meant to imitate the sound of blows landing.

In Ancient Greek, koax was used as the sound of a frog croaking.

In Japanese, dokidoki is used to indicate the beating of a heart.

Sometimes onomatopoeic words have a very tenuous relationship with the object they describe, such as bow-wow in English and wang-wang in Chinese for the sound a dog makes.
Some animals are named after the sounds they make, especially birds such as the cuckoo and chickadee. This practice is especially common in certain languages such as Maori and therefore in names for birds borrowed from these languages.

Here are cool mosses deep,
And through the moss the ivies creep,
And in the stream the
long-leaved flowers weep,
And from the craggy ledge the poppy hangs in sleep.

Onomatopoeia can also represent harsh and unpleasant sounds, as in Browning's “Meeting at Night”:

A tap at the pane, the quick sharp scratch
And blue spurt of a lighted match.

Some other very common English-language examples include:







*ping pong




Aside from the above, machines are usually described with:

automobile - "honk" for the horn, "vroom" for the engine, "screech" for the tires

train - "clickety-clack" crossing a junction, "choo-choo" for the whistle.

cash register - "kaching"

For animal sounds, these words are typically used in English:

bee - "buzz"

cat - "mereow" (U.S.) / "miaow" (UK), "meow" (U.S.) / "miow" (UK), "purr"

bird - "chirp", "tweet"

chickadee - "chickadee"

chicken - "cluck", "cackle", "bawk,"

crow - "caw"

dove - "coo"

duck - "quack"

owl - "hoo" or "hoot"

rooster - "cockadoodledoo"

turkey - "gobble"

cow - "moo"

dog - "woof", "arf", "grrr"

dolphin - "click"

insects - "buzz"

frog - "ribbit", "croak". Note: many species of frog make different calls.

lion - "roar"

horse - "neigh", "whinny", "snort"

pig - "oink", "wee-wee-wee"

sheep - "baa"

Some of these words are used as nouns and verbs when describing the noise.

The second type are cases of sound symbolism. Sound symbolism or phonosemantics is a branch of linguistics and refers to the idea that vocal sounds have meaning. An important concept for understanding this idea is phoneme: phonemes are written between slashes like this /b/. Therefore a flashing beacon called a Gleeper (analogy with bleeper) would have some chance of succeeding, while a new breakfast called Slub would have little chance of success. However, since these elements are not semantic constituents (plate, plane, pane and plain have nothing in common), we shall call them phonetic elicitors of semantic traits.


Cruse, D.A. 1986. Lexical Semantics. Cambridge University Press. Cruse, A.2000. Meaning in language. Oxford University Press. Francis Katamba, 2005. English Words. Second edition. By Routledge. Columbia University press The book of clichés Stanford Encyclopedia of Philosophy




1. Literal and figurative meaning.

We call a sentence or a word meaning literal, when it means what it affirms on the face of it, and nothing else.

Literal meaning is a property of linguistic expressions. Roughly speaking, the literal meaning of a complex sequence of words is determined by its grammatical properties and the meanings that are conventionally assigned to those words. The literal meaning of a statement should be distinguished from its conversational implicature - the information that is implicitly conveyed in a particular conversational context, distinct from the literal meaning of the statement.

For example, suppose we ask Lily whether she wants to go to the cinema and she replies, "I am very tired." Naturally we would infer that Lily does not want to go to the cinema. But this is not part of the literal meaning of what is said. Rather, the information that she does not want to go is conveyed in an implicit manner. Similarly, suppose we hear Lala says, "Po likes books". We might perhaps take Lala to be saying that Po likes to read. But this is only the conversational implicature, and not part of the literal meaning of what is being said. It might turn out that Po hates reading and she likes books only because she regards them as good investment. But even if this is the case, Lala's assertion is still true.

Many traditional academic analyses of language divided linguistic expressions into two classes: literal and figurative. Expressions said to be in figurative language are called figures of speech.

1.1 Traditional analyses

In the traditional analyses, words in literal expressions denote what they mean according to common or dictionary usage, while words in figurative expressions denote something other than what they mean according to common or dictionary usage. Often, in this framework, a particular instance of figurative language can/must be reduced to literal language in order to find out what the expression might be intended to mean.

Sometimes the literal meaning of a particular figure of speech is clear. We can confidently interpret the figure, "The ground is thirsty," to mean "the ground is dry" because we know that the ground cannot literally feel thirst (or anything else, for that matter). Other times, it is harder to pinpoint the literal meaning of a figure of speech. If someone says, "When I first saw her, my soul began to quiver," he might mean, "When I first saw her, I began to fall in love," or "When I first saw her, I began to panic," or something else entirely. Whereas the ground's thirst can only sensibly apply to its dryness, the soul's quivering could refer to a whole range of feelings, including mutually exclusive ones. Only someone familiar with the speaker's feelings could accurately interpret this statement.

How many kinds of figurative language are there? Classical and traditional linguistics by some counts identified more than two hundred and fifty different figures. More recently, some have boiled the number into a much smaller number; some, for example, claim to be able to classify all figurative language as either metaphor or metonymy.

1.2 Figurative as used by Searle and friends

...basically anything that fails the above "letter test". This is supposed to contrast not only with metaphor and poetic language, but also with, e.g., indirect requests.

1.3 Contesting the literal/figurative distinction

Note that most modern academic analyses of language no longer maintain a strict distinction between literal and figurative language. Cognitive linguistics, in particular, may ultimately declare all distinction between literal language and figurative language outdated. Consider what cognitive linguists Gilles Fauconnier and Mark Turner have to say:

...what gets called literal meaning is only a plausible default in minimally specified contexts. It is not clear that the notion "literal meaning" plays any privileged role in the on-line construction of meaning. (Fauconnier and Turner, p. 69)

In other words, "literal meaning" is not a special sort of meaning; it is only the meaning we are most likely to assign to a word or phrase if we know nothing about the context in which it is to be used.

1.4 Establish and non-establish meaning extensions

Cruse (2000) claims three kinds of extension: naturalized, establish and nonce readings for words. Naturalized extensions are those which speakers do not feel anymore as figurative meaning:

The kettle is boiling

She's in love Establish extensions are figurative, but speakers know they are using figures of speech. So, these meanings presumably have an entry in the mental lexicon.

You, lazy bones, come on to work!

She had too many mouth to feed Finally, nonce readings have no entry in the mental lexicon, since these meanings are not yet collectively used, thus not being establish. This is the province of individual creation of figurative meaning and it is usually generated and interpreted using strategies of meaning extension like metaphor and metonymy. They can be found in the field of literature (poetry, drama, narration) as well as in our everyday use of language:

I think I could eat an entire cow

This 5 Euros note is my lunch today

1.5 Metaphors and Metonymy

Within the theory of conceptual metaphor, the claim has been put forward that metaphors and metonymies are grounded in bodily experience; hence the term embodiment (Lakoff and Johnson, 1980, Johnson, 1987). Yet, relatively few studies deal with the analysis of specific, individual body parts in relation to conceptual metaphors, and even more rare are contrastive studies that deal with different languages.

This paper intends to shed more light on these aspects by means of a cross-linguistic analysis of the semantic field of eye compared to their Spanish and Danish counterparts, ojo and øje, respectively.

The analysis shows that non-literal uses of eye are, indeed, linked with mappings of mind and body and these mappings appear to be related directly to the constitution, shape and function of this body part. The analysis is inspired by Pustejovsky's (1995) qualia roles, the data being divided into three main groups from which the semantic field of eye is envisaged: 1) the partitional, 2) the functional and, 3) the telic. Although the overall findings indicate similarities between the languages analysed, there are also clear inter-linguistic divergences - a fact that might indicate differences in conceptualizations of this semantic field.

The analysis supports Deignan and Potter's (2004) view that metonymy is a powerful tool for generating figurative expressions, but suggests a need to distinguish between metonyms that are (more) directly linked to the body (which I will call intrinsic body metonymies), e.g. Then he lifted his eyes from the glass and those that are less directly linked to the body (which I will call extrinsic body metonymies), e.g. The Prime minister has to lift his eyes from this matter in order to look into the future. It would seem that the languages examined differ as to their ability to permit the step from intrinsic to extrinsic metonymy.


In language, a metaphor (from the Greek: metapherin) is a rhetorical trope defined as a direct comparison between two seemingly unrelated subjects. In a metaphor, a first object is described as being a second object. Through this description it is implied that the first object has some of the qualities of the second. In this way, the first object can be economically described because implicit and explicit attributes from the second object can be used to fill in the description of the first. This is exploited in literature and especially in poetry, where with few words, emotions and associations from one context can powerfully be associated with another, different subject.

Metaphor is a subset of analogy and is closely related to other similar concepts such as comparison, simile, allegory and parable. The exact classification of the type of analogy can be unclear when the subjects, the points of comparison or the manner of comparison are omitted.

Types of metaphor

An extended metaphor is one that sets up a principal subject with several subsidiary subjects or comparisons. The above quote from As you like it is a good example. The world is described as a stage and then men and women are subsidiary subjects that are further described in the same context.

A mixed metaphor is one that leaps, in the course of a figure, to a second identification inconsistent with the first one. Example: "Clinton stepped up to the plate and grabbed the bull by the horn". Here, the baseball and the activities of a cowboy are implied. Other examples include: "That wet blanket is a loose cannon"; "Strike while the iron is in the fire"; or (said by an administrator whose government-department's budget was slashed) "Now we can just kiss that program right down the drain".

A dead metaphor is one in which the sense of a transferred image is not present. Example: "to grasp a concept" or "to gather you've understood." Both of these phrases use a physical action as a metaphor for understanding (itself a metaphor), but in none of these cases do most speakers of English actually visualize the physical action. Dead metaphors, by definition, normally go unnoticed. Some people make a distinction between a "dead metaphor" whose origin most speakers are entirely unaware of (such as "to understand" meaning to get underneath a concept), and a dormant metaphor, whose metaphorical character people are aware of but rarely think about (such as "to break the ice"). Others, however, use dead metaphor for both of these concepts, and use it more generally as a way of describing metaphorical cliché.

Other types of metaphor have been identified as well, though the nomenclatures are not as universally accepted:

An active metaphor is one which by contrast to a dead metaphor, is not part of daily language and is noticeable as a metaphor. Example: "You are my sun."

An absolute or paralogical metaphor (sometimes called an antimetaphor) is one in which there is no discernible point of resemblance between the idea and the image. Example: "The couch is the autobahn of the living room."

A complex metaphor is one which mounts one identification on another. Example: "That throws some light on the question." Throwing light is a metaphor and there is no actual light.

A compound or loose metaphor is one that catches the mind with several points of similarity. Example: "He has the wild stag's foot." This phrase suggests grace and speed as well as daring.

An implicit metaphor is one in which the tenor is not specified but implied. Example: "Shut your trap!" Here, the mouth of the listener is the unspecified tenor.

A submerged metaphor is one in which the vehicle is implied, or indicated by one aspect. Example: "my winged thought". Here, the audience must supply the image of the bird.

A simple or tight metaphor is one in which there is but one point of resemblance between the tenor and the vehicle. Example: "Cool it". In this example, the vehicle, "cool", is a temperature and nothing else, so the tenor, "it", can only be grounded to the tenor by one attribute.

A root metaphor is the underlying personal attachments that shape an individual's understanding of a situation. It is different to the previous types of metaphor in that it is not an explicit device in language but merely a part of comprehension. Religion is considered the most common root metaphor since birth, marriage, death and other life experiences can convey a very different meaning to different people based on their level or type of religious adherence. An individual's political affiliations are another root metaphor that may affect the message conveyed by such terms as conservatism and liberal. In the example: "He is a very conservative politician", "conservatism" is the vehicle, "he" is the tenor and the attributes conveyed are dependent on the root metaphor: is it a good or a bad thing to be considered conservative?

A dying metaphor is a derogatory term coined by George Orwell in his essay Politics and the English Language. Orwell defines a dying metaphor as a metaphor that isn't dead (dead metaphors are ok since they are treated like ordinary words), but has been worn out and are used because they save people the trouble of inventing phrases for themselves. Example: Achilles' heel.

The category of metaphor can be further considered to contain the following specialized subsets:

allegory: An extended metaphor in which a story is told to illustrate an important attribute of the subject

catachresis: A mixed metaphor (sometimes used by design and sometimes a rhetorical fault)

parable: An extended metaphor told as an anecdote to illustrate or teach a moral lesson


Originally, metaphor was a Greek word meaning "transfer". The Greek etymology is from meta, implying "a change" and pherein meaning "to bear, or carry".

In modern Greek the word metaphor also means transport or transfer.

Metaphor and Simile

Metaphor and simile are two of the best known tropes and are often mentioned together as examples of rhetorical figures. Metaphor and simile are both terms that describe a comparison: the only difference between a metaphor and a simile is that a simile makes the comparison explicit by using "like" or "as." Despite the similarity of the two figures, and the fact that they have historically been used as synonyms, it is the distinction between them which is normally focused upon in teaching. Ironically, "not knowing the difference between a simile and a metaphor" is sometimes used as a euphemism for knowing little about rhetoric or literature. Of course, someone truly versed in rhetoric understands that there is very little difference between metaphor and simile, and that the distinction is trivial compared to, for example, the difference between metonymy and metaphor.

The Colombia Encyclopedia, 6th edition, explains the difference as:

a simile states that A is like B, a metaphor states that A is B or substitutes B for A.

Thus, "You are my sunshine" is a metaphor, whereas "Your eyes are like the sun" is a simile. However, some describe similes as simply a specific type of metaphor (see Joseph Kelly's The Seagull Reader (2005), pages 377-379). Most dictionary definitions of both metaphor and simile support Kelly's assertion, and historically it appears the terms were used essentially as synonyms. Nonetheless, many lists of literary terms define metaphor as "a comparison not using like or as", showing the emphasis often put on teaching this distinction.

Usually, similes and metaphors could easily be interchanged. For example remove the word 'like' from William Shakespeare's simile, "Death lies on her, like an untimely frost," and it becomes "Death lies on her, an untimely frost," which retains almost exactly the same meaning. However, at other times using a simile as opposed to a metaphor clarifies the analogy by calling out exactly what is being compared. "He had a posture like a question mark" (Corbett, Classical rhetoric for the modern student (1971), page 479) has one possible interpretation, that the shape of the posture is that of a question mark, whereas "His posture was a question mark" has a second interpretation, that the reason for the posture is in question. At other times use of a simile rather than a metaphor adds meaning by calling to attention the process of comparison, as in "A woman without a man is like a fish without a bicycle" (Irina Dunn). The point is not to compare a woman to a fish, but to ask the reader to consider how the woman is like the fish. Finally, similes are often more convenient than metaphors when analogizing actions as opposed to things: "Wide sleeves fluttering like wings" (Marcel Proust) does not translate easily from simile to metaphor.

Metaphors in literature and language

Metaphor is present in written language back to the earliest surviving writings. From the Epic of Gilgamesh (one of the oldest Sumerian texts):

My friend, the swift mule, fleet wild ass of the mountain, panther of the wilderness, after we joined together and went up into the mountain, fought the Bull of Heaven and killed it, and overwhelmed Humbaba, who lived in the Cedar Forest, now what is this sleep that has seized you? - (Trans. Kovacs, 1989)

In this example, the friend is compared to a mule, a wild ass, and a panther to indicate that the speaker sees traits from these animals in his friend.

The Greek plays of Sophocles, Aeschylus, and Euripides, among others, were almost invariably allegorical, showing the tragedy of the protagonists, either to caution the audience metaphorically about temptation, or to lambast famous individuals of the day by inferring similarities with the characatures in the play.

Even when they are not intentional, parallels can be drawn between most writing or language and other topics. In this way it can be seen that any theme in literature is a metaphor, using the story to convey information about human perception of the theme in question.


In rhetoric and cognitive linguistics, metonymy (in Greek  (meta) = after/later and  (onoma) = name) (IPA: m-tn'-m) is the use of a single characteristic to identify a more complex entity. It is also known as denominatio or pars pro toto (part for the whole).

In cognitive linguistics, metonymy is one of the basic characteristics of cognition. It is extremely common for people to take one well-understood or easy-to-perceive aspect of something and use that aspect to stand either for the thing as a whole or for some other aspect or part of it.

In rhetoric, metonymy is the substitution of one word for another with which it is associated. There are two senses for metonymy:

  • Metonymy is, broadly defined, a trope in which one entity is used to stand for another associated entity.

  • Metonymy is, more specifically, a replacive relationship that is the basis for a number of conventional metonymic expressions occurring in ordinary language.

  • Examples of Metonym

    Some common examples of metonymy are:

    The pen is mightier than the sword (pen is a metonym for writers and the exchange of ideas; sword is a metonym for war and violence).

    "The White House", to refer to the US President and his advisors.

    "The crown", to refer to the king

    "The press", literally referring to the printing press

    "The media", literally referring to the various media of news communication (print, radio, television, etc.)

    • The pen is mightier than the sword.

    Pen and sword represent publishing and military force, respectively.

    Nixon stands for the armed forces that Nixon controlled.

      • A Mercedes rear-ended me.

    The word me stands for the car that the speaker was driving.

    Metonymy and Metaphor

    Metonymy works by contiguity rather than similarity. The White House is associated with, or near, the president; a writer always has a pen close at hand. However, when people use metonymy, they don't typically wish to transfer qualities as they do with metaphor: there is nothing house-like about the president, crown-like about the king, or pen-like about a writer. Rather, metonymy transfers a whole set of associations which may or may not be integral to the meaning.

    In linguistics, as in rhetoric, the distinction between metaphor and metonymy is important. Two examples using the term "fishing" help make the distinction clear (example drawn from Dirven, 1996).

    The phrase "to fish pearls" uses metonymy, drawing from "fishing" the idea of taking things from the ocean. What is carried across from "fishing fish" to "fishing pearls" is the domain of usage and the associations with the ocean and boats, but we understand the phrase in spite of rather than because of the literal meaning of fishing: we know you do not use a fishing rod or net to get pearls.

    In contrast, the metaphorical phrase "fishing for information", transfers the concept of fishing (waiting, hoping to catch something that you can't see) into a new domain.

    Thus, metonymy works by calling up a domain of usage and an array of associations (in the example above, boats, the ocean, gathering life from the sea) whereas metaphor picks a target set of meanings and transfers them to a new domain of usage.

    "Metaphors We Live By" by George Lakoff and Mark Johnson

    Our selection comprises chapters 1, 2, 3, of Metaphors We Live By (1980).


    Metaphor is for most people device of the poetic imagination and the rhetorical flourish--a matter of extraordinary rather than ordinary language. Moreover, metaphor is typically viewed as characteristic of language alone, a matter of words rather than thought or action. For this reason, most people think they can get along perfectly well without metaphor. We have found, on the contrary, that metaphor is pervasive in everyday life, not just in language but in thought and action. Our ordinary conceptual system, in terms of which we both think and act, is fundamentally metaphorical in nature.

    The concepts that govern our thought are not just matters of the intellect. They also govern our everyday functioning, down to the most mundane details. Our concepts structure what we perceive, how we get around in the world, and how we relate to other people. Our conceptual system thus plays a central role in defining our everyday realities. If we are right in suggesting that our conceptual system is largely metaphorical, then the way we thinks what we experience, and what we do every day is very much a matter of metaphor.

    But our conceptual system is not something we are normally aware of. in most of the little things we do every day, we simply think and act more or less automatically along certain lines. Just what these lines are is by no means obvious. One way to find out is by looking at language. Since communication is based on the same conceptual system that we use in thinking and acting, language is an important source of evidence for what that system is like.

    Primarily on the basis of linguistic evidence, we have found that most of our ordinary conceptual system is metaphorical in nature. And we have found a way to begin to identify in detail just what the metaphors are halt structure how we perceive, how we think, and what we do.

    To give some idea of what it could mean for a concept to be metaphorical and for such a concept to structure an everyday activity, let us start with the concept ARGUMENT and the conceptual metaphor ARGUMENT IS WAR. This metaphor is reflected in our everyday language by a wide variety of expressions:


    Your claims are indefensible.

    He attacked every weak point in my argument.

    His criticisms were right on target.

    I demolished his argument.

    I've never won an argument with him.

    you disagree? Okay, shoot!

    If you use that strategy, he'll wipe you out.

    He shot down all of my arguments.

    It is important to see that we don't just talk about arguments in terms of

    It is important to see that we don't just talk about arguments in terms of war. We can actually win or lose arguments. We see the person we are arguing with as an opponent. We attack his positions and we defend our own. We gain and lose ground. We plan and use strategies. If we find a position indefensible, we can abandon it and take a new line of attack. Many of the things we do in arguing are partially structured by the concept of war. Though there is no physical battle, there is a verbal battle, and the structure of an argument--attack, defense, counter-attack, etc.---reflects this. It is in this sense that the ARGUMENT IS WAR metaphor is one that we live by in this culture; its structures the actions we perform in arguing. Try to imagine a culture where arguments are not viewed in terms of war, where no one wins or loses, where there is no sense of attacking or defending, gaining or losing ground. Imagine a culture where an argument is viewed as a dance, the participants are seen as performers, and the goal is to perform in a balanced and aesthetically pleasing way. In such a culture, people would view arguments differently, experience them differently, carry them out differently, and talk about them differently. But we would probably not view them as arguing at all: they would simply be doing something different. It would seem strange even to call what they were doing "arguing." In perhaps the most neutral way of describing this difference between their culture and ours would be to say that we have a discourse form structured in terms of battle and they have one structured in terms of dance. This is an example of what it means for a metaphorical concept, namely, ARGUMENT IS WAR, to structure (at least in part) what we do and how we understand what we are doing when we argue. The essence of metaphor is understanding and experiencing one kind of thing in terms of another.. It is not that arguments are a subspecies of war. Arguments and wars are different kinds of things--verbal discourse and armed conflict--and the actions performed are different kinds of actions. But ARGUMENT is partially structured, understood, performed, and talked about in terms of WAR. The concept is metaphorically structured, the activity is metaphorically structured, and, consequently, the language is metaphorically structured.

    Moreover, this is the ordinary way of having an argument and talking about one. The normal way for us to talk about attacking a position is to use the words "attack a position." Our conventional ways of talking about arguments presuppose a metaphor we are hardly ever conscious of. The metaphors not merely in the words we use--it is in our very concept of an argument. The language of argument is not poetic, fanciful, or rhetorical; it is literal. We talk about arguments that way because we conceive of them that way--and we act according to the way we conceive of things.

    The most important claim we have made so far is that metaphor is not just a matter of language, that is, of mere words. We shall argue that, on the contrary, human thought processes are largely metaphorical. This is what we mean when we say that the human conceptual system is metaphorically structured and defined. Metaphors as linguistic expressions are possible precisely because there are metaphors in a person's conceptual system. Therefore, whenever in this book we speak of metaphors, such as ARGUMENT IS WAR, it should be understood that metaphor means metaphorical concept.


    Arguments usually follow patterns; that is, there are certain things we typically do and do not do in arguing. The fact that we in part conceptualize arguments in terms of battle systematically influences the shape argument stake and the way we talk about what we do in arguing. Because the metaphorical concept is systematic, the language we use to talk about that aspect of the concept is systematic.

    We saw in the ARGUMENT IS WAR metaphor that expressions from the vocabulary of war, e.g., attack a position, indefensible, strategy, new line of attack, win, gain ground, etc., form a systematic way of talking about the battling aspects of arguing. It is no accident that these expressions mean what they mean when we use them to talk about arguments. A portion of the conceptual network of battle partially characterizes file concept of an argument, and the language follows suit. Since metaphorical expressions in our language are tied to metaphorical concepts in a systematic way, we can use metaphorical linguistic expressions to study the nature of metaphorical concepts and to gain an understanding of the metaphorical nature of our activities.

    To get an idea of how metaphorical expressions in everyday language icon give us insight into the metaphorical nature of the concepts that structure our everyday activities, let us consider the metaphorical concept TIME IS Money as it is reflected in contemporary English.


    You're wasting my time.

    This gadget will save you hours. I don't have the time to give you.

    How do you spend your time these days? That flat tire cost me an hour.

    I've invested a lot of time in her.

    1 don't have enough time to spare for that.You're running out of time.

    He's living on I borrowed time.

    You don't use your time, profitably.

    I lost a lot of time when I got sick.

    Thank you for your time.

    Time in our culture is a valuable commodity. It is a limited resource that we use to accomplish our goals. Because of the way that the concept of work has developed in modern Western culture, where work is typically associated with the time it takes and time is precisely quantified, it has become customary to pay people by the hour, week, or year. In our culture TIME IS MONEY in many ways: telephone message units, hourly wages, hotel room rates, yearly budgets, interest on loans, and paying your debt to society by "serving time." These practices are relatively new in the history of the human race, and by no means do they exist in all cultures. They have arisen in modern industrialized societies and structure our basic everyday activities in a very profound way. Corresponding to the fact that we act as if time is a valuable commodity--a limited resource, even money--we conceive of time that way. Thus we understand and experience time as the kind of thing that can be spent, wasted, budgeted, invested wisely or poorly, saved, or squandered.

    TIME IS MONEY, TIME IS A LIMITED RESOURCE, and TIME IS A VALUABLE COMMODITY are all metaphorical concepts. They are metaphorical since we are using our everyday experiences with money, limited resources, and valuable commodities to conceptualize time. This isn't a necessary way for human beings to conceptualize time; it is tied to our culture. There are cultures where time is none of these things.

    The metaphorical concepts TIME IS MONEY, TIME 1S A RESOURCE, and TIME IS A VALUABLE COMMODITY form a single system based on sub-categorization, since in our society money is a limited resource and limited resources are valuable commodities. These sub categorization relationships characterize entailment relationships between the metaphors: TIME IS MONEY entails that TIME IS A LIMITED RESOURCE, which entails that TIME 1S A VALUABLE COMMODITY.

    We are adopting the practice of using the most specific metaphorical concept, in this case TIME IS MONEY to characterize the entire system. Of the expressions listed under the TIME IS MONEY metaphor, some refer specifically to money (spend, invest, budget, probably cost), others to limited resources (use, use up, have enough of, run out of), and still others to valuable commodities (have, give, lose, thank you for). This is an example of the way in which metaphorical entailments can characterize a coherent system of metaphorical concepts and a corresponding coherent system of metaphorical expressions for those concepts.

    The very systematicity that allows us to comprehend one aspect of a concept in terms of another (e.g., comprehending an aspect of arguing in terms of battle) will necessarily hide other aspects of the concept. In allowing us to focus on one aspect of a concept (e.g., the battling aspects of arguing), metaphorical concept can keep us from focusing on other aspects of the concept that are inconsistent with that metaphor. For example, in the midst of a heated argument, when we are intent on attacking our opponent's position and defending our own, we may lose sight of the cooperative aspects of arguing. Someone who is arguing with you can be viewed as giving you his time, a valuable commodity, in an effort at mutual understanding. But when we are preoccupied with the battle aspects, we often lose sight of the cooperative aspects.

    A far more subtle case of how a metaphorical concept can hide an aspect of our experience can be seen in what Michael Reddy has called the "conduit metaphor."' Reddy observes that our language about language is structured roughly by the following complex metaphor:




    The speaker puts ideas (objects) into words (containers) and sends them (along a conduit) to a bearer who takes the idea/objects out of the word/containers. Reddy documents this with more than a hundred types of expressions in English, which he estimates account for at least 70 percent of the expressions we use for talking about language. Here are some examples:


    It's hard to get that idea across to him.

    I gave you that idea.

    Your reasons came through to us.

    It's difficult to put my ideas into words.

    When you have a good idea, try to capture it immediately in words.

    Try to pack more thought into fewer words.

    The meaning is right there in the words.

    Don't force your meanings into the wrong words.

    His words carry little meaning..

    Your words seem hollow.

    In examples like these it is far more difficult to see that there is anything hidden by the metaphor or even to see that there is a metaphor here at all. This is so much the conventional way of thinking about language that it is sometimes hard to imagine that it might not fit reality. But if we look at what the conduit metaphor entails, we can see some of the ways in which it masks aspects of the communicative process.

    REFERENCES Institute of Language and Communication, University of South. Lakoff, G. and Johnson 1980 Francis Katamba, 2005. English Words. Second edition. By Routledge