Cognitive Science: An Introduction/Metaphor and Analogy

I. Part 1: Metaphor ~ An Introduction

The following is an introduction to George Lakoff’s highly influential Contemporary Metaphor Theory (hereafter: CMT). Lakoff, in true innovative fashion, redefined the way that we view human cognition by challenging the prevailing dogmas of the day. For Lakoff, this meant overcoming the central tenets and assumptions of our contemporary language philosophies and linguistic models[1]. Lakoff's dissent from the widely influential Chomskian paradigm sparked what has now come to be known as the ‘linguistic wars’ (a term coined by author and professor of English Language and Literature at the University of Waterloo, Randy Allan Harris). Lakoff’s subsequent works rest upon his fundamental separation with Chomskian generative syntax.

Because the following is intended as an introduction, it will prove beneficial to begin with the common, intuitive understanding of metaphor. Afterward, I will provide some basic terminology that will serve to facilitate a more technical discussion of the topic. Finally, I discuss the state of language research and Lakoff’s addendum to the field. We will begin our undertaking by way of an example: Consider the following utterances:

(1) Wikipedia is a goldmine!

(2) Lawyers are snakes.

Although you may not have heard (or read) an utterance such as (1) or (2) above, it is uncontroversial to claim that you had no problem interpreting their meaning. The above are examples of what, since Aristotle[2], we commonly refer to as metaphor: they are instances in which words, such as “goldmine” and “snakes,” are not used in their everyday sense, but in novel ways. The word “metaphor” comes to us from the Greek word metaphora, which means “transfer” or “to carry over.” This denotation implies an act of substitution, involving the transference of a word to a novel sense by comparing and/or juxtaposing two ostensibly unrelated entities (e.g., in this case, goldmine is used to predicate of Wikipedia something like ‘valuable’; in the same way, the word ‘snakes’ is used to predicate of lawyers something akin to deceit).

Classical theories of language regarded metaphor as being a matter of language and not thought. What this means is that metaphor was understood as a stylistic device; a deviant use of language that served to illuminate hidden similarities between objects. In this sense, metaphor contributes substantially to public oratory, social persuasion, and literature, but is disconnected from our thoughts and concepts—which are literal through and through. Importantly, metaphor was taken as markedly different from the domain of ordinary, conventional language: “everyday language had no metaphor, and metaphor used mechanisms outside the realm of everyday conventional language”[3]. It may come as no surprise to you that the word metaphor, today, still carries most of its original meaning. Although this definition is still with us today, Lakoff’s innovation has helped mold our understanding of metaphor into something far more technical and nuanced than the traditional view considers—and it has been the focus of controversy over the last few decades.

Before moving into the arena of scholarly dispute and controversy, we may point to a simple fact that no language theorist would deny: Metaphor is a ubiquitous linguistic phenomenon occruing in nearly every domain of linguistic communication. Indeed, some ways of speaking metaphorically are familiar to us, and have become so familiar that we fail to recognize their metaphoricity. A quick read through a newspaper is quite telling of our veritable agnosia, where for example, passages from the financial section will make mention of soaring stocks, or plummeting gas prices. At the same time, other metaphors are so novel and outlandish that they may preoccupy us for a long while. Take for example, T. S. Eliot’s Heart of light[4] in the IV Quartets—a perfect example of metaphor’s meditative and abstract conceptual qualities. Additionally, we use metaphors in a wide range of contexts: from every day conversation (the professor did not give the students any idea of what to expect), to literature (Juliet is the Sun), and even scientific discourse (genes are selfish, electricity flows).

Although this linguistic phenomenon has been a part of our linguistic communicative practices since the beginning, it has only recently begun to generate serious scholarly attention. Indeed, with only a few exceptions within literary criticism, hermeneutics, and classical rhetoric, it wasn’t until the twentieth century that work on metaphor came to be viewed as a credible field of study to a wider academic community. As Gerard Steen humorously puts it, “in the beginning there was Aristotle. Then there were the Dark Ages, which lasted until 1980. Then there was Lakoff…”[5]

II. Terminology

Before we delve into the contemporary study of metaphor, it will prove useful to first identify some basic features and terminology that will be used throughout the rest of this chapter.

Within metaphor research, it has become customary to think about a metaphor as composed of two halves. The reason, as was mentioned earlier, is that metaphor typically involves the comparison of two things[6].[7] The first half is customarily referred to as the topic or target domain[8]. You may think of it as the subject under comparison, or the things we are trying to understand. Consider an example from Shakespeare’s As You Like It (II.VII):

(3) All the world’s a stage.

Here, a melancholic Jaques compares the world to the stage of a theatrical performance. Using example (3) as our guide, we may designate ‘world’ as the topic. The second half of the metaphor is typically referred to as the vehicle, or source domain, from which we draw the meaning of the metaphorical expression. Using example (3), we can say that the source (of meaning) is how we understand the ‘world’ through our understanding of ‘stage’. Here we ask ourselves a very pertinent question: “how does the metaphor achieve this effect?” To this answer this, we add a third feature: the meaning arises through a system of associated commonplaces through which we filter understanding of the topic through related features from the vehicle. In plainer English, an aggregate set of attributes, traits, characteristics, relations and connotations from the source facilitate our understanding of the topic (the thing that is predicated of the vehicle).

Try and visualize all that you know in relation to ‘stages’. Now, not everything that we know about stages is transferred in the metaphor, otherwise it would quickly break down and become increasingly difficult to interpret the meaning; rather, we refine our understanding of the metaphor by mapping only those things that are relevant to the topic and appropriate within the larger context in which the metaphor occurs. The system of associated commonplaces will include features that transfer from ‘stage’ to ‘world’. If you are having trouble, allow me to suggest a few: “actors,” “artful,” “artifice,” “constructed,” “contrived,” “fictional/fictitious,” “ornate,” etc. These bits of knowledge come to imbue our understanding of ‘world’ as something like


or perhaps, more formidably


Finally, we may abstract away and generalize to the proposition


(and depending on the metaphor, the list of propositional import may go on and on, indefinitely).

Yet, an important question still lingers: how is it that an audience can be so accurate in determining the meaning of something which can be so enigmatic? Although there have been numerous answers offered, the remainder will provide a brief sketch of the standard model in order to highlight the major developments of the CMT.

III. Metaphor and the Standard Pragmatic Account

At one point, Paul Grice’s theory of implicatures had monopolized theorizing in all domains of indirect communication. His framework provided psychological plausibility and was alive to the state of semantic theorizing at the time. This became known as a theory of implicatures, as coined by Grice, and served to investigate those aspects of meaning that arise out of a conversational context where a speaker said one thing but meant another[10]. Following Grice, pragmatic theorists working within the Anglo-American tradition made some significant contributions to metaphor research by adopting his framework. Perhaps the most notable contribution was extending meaning to include speaker’s intentions and in turn, sketching a psychological explanation for deriving the meaning of figurative expressions by entertaining the speaker’s meaning. For example, Griceans typically view metaphor as a matter of what the addresser intends to communicate (which significantly departs from what they said) and that the retrieval of its meaning rests on a number of assumptions that the recipient reasons through about the addresser. In other words, successful communication results when the recipient recognizes that the addresser intends them to recognize what s/he is trying to communicate to them. In plainer terms, the meaning of a metaphor rests on what the recipient believes the communicator intends to communicate to them.

In this accounts most mature permutation, retrieval of metaphorical meaning involves three stages. First, the hearer utters a statement. Secondly, given the context of utterance, the recipient reasons that the communicator did not intend their words literally and rejects the literal interpretation. Finally, the recipient reasons about the communicators intentions in order to generate possible meanings that the speaker may have meant.

By focusing on a simple subject-predicate, such as S is P whereby the addressor means metaphorically S is R. Searle analyzes three sets of elements within this structure: the subject expression “S” and the object or objects it refers to; the predicate expression “P” that is uttered along with its literal meaning, truth conditions, and denotation(s) (if any); and there is the speaker’s utterance meaning “S is R” and its truth conditions[11].

Searle considers eight principles in total for metaphorical utterances. Briefly, the principles include “Rs being a salient feature of P-things, either by definition or contingency. Things which are P are often said or believed to be R; alternatively, it may be “a fact about our sensibility, whether culturally or naturally determined, that we…perceive connection, so that P is associated in our minds with R properties.”[12] Alternatively, if P things are not like R, and are not believed to be R like, the condition of P may be like the condition of being R[13]. Finally, after generating a set of possible meanings by these principles, the hearer must decide which she believes is most likely to be the speaker’s intended meaning. Camp and Reimer outline three virtues for motivating a Gricean pragmatic approach as follows:

First, it captures the fact that metaphors are meaningful, that they have a “cognitive content” other than literal content. Second, it does this without violating what Grice called ‘modified Occam’s Razor.’ This methodological principle is simply Occam’s Razor applied to linguistic meanings: Don’t multiply senses beyond necessity.[14] It does so because it incorporates literal sentence meaning with interpretive principles. Lastly, it accommodates a variety of linguistic tropes that are used in communication where there is a clear separation of speaker and sentence meaning. However, we can point to several shortcomings of the Gricean paradigm. Such theories typically assume that the metaphorical interpretation is prompted by the recipient identifying the utterance to be literally deviant. In that case, metaphors that are literally true on first reading seem to present a problem to this initial step. To highlight this, consider a metaphor such as

(4) No man is an island.

Where (4) could be taken as a true (and informative) given a literal context of utterance; yet, it also retains a metaphorical interpretation. Secondly, pragmatists generally assume that metaphorical meaning, like speaker meaning, can be fully cashed-out in propositional terms. However, Camp (2008) points out that metaphors, especially novel, poetic ones, ‘go beyond’ their propositional import. In this way pragmatists miss out on the most interesting features (i.e., those which seem irreducible to literal paraphrase).

Thirdly, notice that the Gricean-cum-Searlean paradigm makes an implicit empirical prediction: ‘the more stages required to retrieve the meaning of X, the longer X will take to process’ based on the fact that metaphorical interpretation happens indirectly because we must first retrieve and then reject the literal meaning of the linguistic input.[15] In a burst of creativity, this account has come to be known as the indirect hypothesis. In contrast, there is a prodigious amount of evidence which suggests that the literal interpretation need not be processed before we access a metaphorical meaning.[16] This is precisely the issue that Lakoff identifies.

The conventional wisdom of modern language philosophies posits that there is a divide between literal and figurative language. Given this distinction, it seems plausible to assume that one arrives at a metaphorical reading of a sentence by beginning with a literal meaning. Knowing that the literal meaning is unwarranted in the context in which it occurs, the recipient applies some algorithmic process in order to come up with a metaphorical interpretation. Lakoff’s theory begins by rejecting the dichotomy between the literal and figurative as infelicitous. Such a position, he claims, is tenable given we adopt the following fallacious assumptions:

1. All everyday language is literal, and none of it is metaphorical.

2. All subject matter can be comprehended literally, without recourse to metaphor.

3. Only literal language can be contingently true or false.

4. All definitions given in the lexicon of a language are literal, not metaphorical.

5. The concepts used in the grammar of a given language are all literal; none are metaphorical.

IV. The Lakoffian Shift

We shall look at these assumptions in a little more depth: As we have already mentioned, (1) is patently false. Metaphors abound in both our quotidian and technical discourses. (2) is also highly contestable. We may highlight our suspicion by pointing to instances such as the following: Astrophysicists often describe the distribution of mass in the universe as ‘foam-like’; chemists still ascribe orbitals to atoms as if electrons were planets spinning around a nuclear sun; biologists employ the term ‘genetic code’; environmentalists sometimes describe the Earth as if it were a living organism. In regards to (3) it may serve us better to understand issues of truth and falsity as instances of felicity (or whether the utterances follows the conventions of its context of utterance). Alternatively, rather than asking if a statement is true, it may be better to describe it as apt or appropriate.[17] (4) maybe disproven simply by looking in a dictionary. (5) is a little more tricky to tackle, and is one of CMTs central tenants which Lakoff has gone through great lengths to demonstrate. We shall spend some time devoted to Lakoff’s attempts to formulate his response to this.

Lakoff admits that the bulk of his ideas are indebted to Michael Reddy’s essay “The Conduit Metaphor,” claiming that Reddy has done “far more than he modestly suggested.”[18] In the article, Reddy analyzed the concept of communication and showed that the various ways in which we understand and talk about communication in everyday English is replete with metaphor. For a single, significant case, Reddy was also able to show that the locus of metaphor was thought, and not language; that metaphor is an indispensable part of our basic conventional way of conceptualizing the world, and that everyday habits reflect our metaphorical understanding of experience. Inspired by its potential, Lakoff and colleagues demonstrated through rigorous linguistic analysis (accompanied by voluminous examples) that Reddy’s findings could be generalized over numerous conceptual domains. All of the evidence, claims Lakoff, points to a system of conventional conceptual metaphors. If so, Lakoff claims that is it possible to show that large swaths of metaphorical expressions have a common root.

We will now consider an example by keeping the following two questions in mind: (1) is there a general principle governing the way in which a topic is characterized in everyday communication? and (2) is there a general principle governing how patterns of inference about the vehicle are used to reason about the topic of a metaphor?

Imagine a troubled, amorous relationship described as follows:

(5) We’ve hit a few bumps along the way, over the course of our relationship.

In the example, love is being conceptualized as a journey. The above implies that the relationship has had some set-backs, that the lovers find it difficult to keep going the way they’ve been going, and that they have experienced some obstacles. Lakoff suggests that this is not an isolated case. Rather, English utilizes many common expressions based on conceptualizing love as a journey, “and they are not only used for talking about love but also reasoning about it as well.”[19] For example, you may often hear people saying such things as look how far we’ve come. We can’t turn back now. We’re at a crossroads. It’s been a long and bumpy road. Our relationship is off the tracks.

Such expressions are not poetic, but common quotidian English expressions. Recall the two aforementioned questions. We may answer both in the affirmative: there is a single general principle that is part of the conceptual system underlying English. It is one that characterizes the domain of love in terms of the domain of journeys. Informally, Lakoff states the principle in the following way: The lovers are travelers on a journey together, with their common life goals seen as destinations to be reached. The relationship is their vehicle, and it allows them to pursue those common goals together. The relationship is seen as fulfilling its purpose as long as it allows them to make progress toward their common goals. The journey isn’t easy. There are impediments, and there are places (crossroads) where a decision has to be made about which direction to go in and where to keep traveling together.[20] Here, the metaphor involves a mapping from the vehicle domain (i.e., journeys), to the target domain (i.e., love). The mapping involves ontological correspondences between these two domains.

The LOVE-AS-JOURNEY mapping:

Lovers correspond to travelers.

Love relationships corresponds to a vehicle.

Lover’s common goals correspond to common destinations on the journey.

Difficulties in a relationship correspond to common destinations on the journey.

These correspondences are given a mnemonic, typically of a propositional form: TARGET-DOMAIN IS SOURCE-DOMAIN. In this case, the name of these variegated mappings receives the name LOVE IS A JOURNEY. In the literature on CMT, small capitals like LOVE IS A JOURNEY are used to represent these mappings. Thus, by using these small capitals, we are referring to the set of correspondences above. In addition, we are told that such ontological correspondences characterize epistemic ones by “mapping knowledge about journeys onto knowledge about love. Such correspondences permit us to reason about love using the knowledge we use to reason about journeys.”[21] It is imperative that we do not mistake the name of the mapping for the mapping itself. Such a mistake leads one to the misunderstanding that the mappings themselves are propositions, and that metaphors are propositional, by extension. Rather, metaphors are sets of conceptual correspondences.

Consequently, an understanding of metaphor as a conceptual cross-domain mapping acknowledges that it is not simply a matter of language, but of thought primarily.[22] If, for example, metaphor was a matter of language, then different metaphoric expressions would only be related to one another prima facie. Looking back to the above examples, it is hard to deny that each linguistic expression represents an isolated case of metaphor without any relation to the root metaphor LOVE IS A JOURNEY. In other words, with Lakoff, we observe that our above examples are not different metaphors, but simply different linguistic instantiations of the root or conceptual metaphor LIVE IS A JOURNEY. Furthermore, mappings between domains occur at a superordinate level. Consider LOVE IS A JOURNEY. Here the mapping corresponds to some means of transportation or vehicle. The vehicle is a superordinate category that includes more basic level categories such as car, train, boat, and plane. Typically, we find mappings at the superordinate level, and not general sub-mappings like LOVE IS A CAR.

The fact that these mappings are conventional (in that it is a fixed part of the conceptual apparatus giving way to our various linguistic instantiations) explains how novel uses of the mapping can be understood so quickly. Since the mapping is primary, in that it sanctions the use of source domain language and inference patterns for target domain concepts, any novel extension exploits the ontological and epistemic correspondences between the source and target. It is here that we are now given the technical definition of ‘metaphor’: an aggregate of conceptual cross-domain mappings at the superordinate level. Alternatively, we reserve the term “metaphoric expression” to designate the many individual linguistic expressions that arise from our conceptual apparatus.

V. Conceptual Metaphor and Experience

Undoubtedly, there are those who are wondering “why do we have the conventional metaphors that we do? For Lakoff and his collaborators postulate that conventional metaphors are grounded in embodied experience. Our everyday, real life experiences forms the basis for our metaphorical cognitions. For example, the experiential basis of the KNOWING IS SEEING metaphor[23] rests on the fact that we so rely on our vision to discover most of what we know. In the majority of cases, when we see something, we believe it to be true. Turning back to our original example, LOVE IS A JOUNREY exploits the same sort of process so that one familiar thing (seeing, sojourning) lends itself to imbue another, more elusive thing (knowing, amorous relationships). In other words, our interaction with the world provides the basis for spatial schemas that in turn provide the input for the creation of conceptual metaphors which organizes the way we view and understand it. Over time, single cognitive dispositions may then manifest in many different verbal instantiations.

VI. Research

If metaphors are basic conceptual patterns of thought, then simply pointing to a few examples on paper prove to be insufficient to those of us who are sceptically-minded. In order to rectify the claim that Lakoff has made, we will turn to some important experimentation in the field of cognitive linguistics. Recall our discussion about the Gricean-inspired model (pragmatic, indirect) and Lakoff’s competing theory (CMT, direct). On the one hand, the indirect hypothesis predicts that metaphors will require more processing time than their literal counterparts. On the other hand, the fact that many metaphors are understood unreflectively in everyday discourse leads the CMT to predict that metaphors are understood directly. Cognitive linguists reject the three-stage hypothesis posited by pragmatists which is backed up by further psycholinguistic studies showing the pragmatic account to be unnecessarily complicated.[24] [25] Each of the studies we will discuss assessed comprehension by asking subjects to make rapid responses to or decisions about the meanings of metaphorical and literal statements. Participants in these reaction time studies are presented with linguistic stimuli to which they must respond quickly. For example, participants may be asked to push a button as soon as they comprehend what they have read, or they may be asked to judge the similarity between two sentences. Reaction time studies generally show that people take between 1 and 4 seconds to read and understand simple figurative utterances (such as the examples above).

As mentioned, the pragmatic model does not explain the normal processing strategies when people comprehend metaphors in appropriate social and linguistic contexts as confirmed by Ortony, Schallert, Reynolds, & Anots (1978).[26] An example of these two contexts within the experiment is as follows:

Literal Context

Approaching the enemy infantry, the men worried about touching off landmines. They were anxious that their presence would be detected prematurely. Their fears were compounded by the knowledge that they might be isolated from their reinforcements. The outlook was grim. Regardless of the danger, the troops marched on.

Metaphorical Context

The children continued to annoy their babysitter. She told the little boys she would not tolerate any more bad Behaviour. Climbing all over the furniture was not allowed. She threatened to spank them if they continued to stomp, run, and scream around the room. The children knew that her spanking hurt. Regardless of the danger, the troops marched on.

Ortony et al (1978) concluded that with sufficient context, participants did not need to analyze the literal interpretation of the metaphorical utterance before coming to the intended meaning which was reflected in equivalent reaction times across both conditions. However, when participants read the target statements in short contexts, metaphorical targets took significantly longer to read than literal counterparts. A more recent study included a control in which the context preceding the target sentences were unrelated to either the figurative or literal interpretations of the targets.[27] Additionally, participants’ eye movements were tracked, and the amount of time people focused on the targets was recorded. These reaction time data, varying between 1500 and 2500 msec, confirmed the findings in that metaphors were comprehended as quickly as literal targets when preceded by longer contexts. Moreover, the study also concluded that there were no differences in reading times for literal and metaphorical targets preceded by shorter contexts.

Another important study conducted by Glucksberg, Gildea and Bookin (1982)[28] found that it actually took subjects longer to determine whether a sentence was literally true when they were also given a plausible metaphorical interpretation, even though they were asked to consider only literal truth-value. For example, although people correctly judged statements such as Some jobs are jails as literally false, the availability of a true metaphoric interpretation (i.e., “Some people are trapped in their occupations”) interfered with making literal judgments. When metaphoric interpretations of literally false sentences were available, participants took significantly longer to decide that such sentences were false. This suggests that metaphorical interpretation arises automatically, and does not require adjustment from the failure of literal meaning. The studies mentioned so far tested the psychological plausibility of the indirect hypothesis by examining the comprehension of nonliteral language.[29] The general picture we gather from them is that metaphor comprehension is very much akin to literal language comprehension. These studies are taken as canonical in the field and overwhelming evidence in favour of the CMT.

VII. Concluding Remarks

What is not metaphorical:

It may seem tempting to conclude that Lakoff has committed himself to the view that the entirety of our conceptual system is metaphorical. However, this would be to misrepresent his position. Although the old literal-metaphorical distinction was based on false assumptions that have proved to be false, one can make a different sort of literal-metaphorical distinction: those concepts that are not comprehended via conceptual metaphor might be called “literal.” Thus, although (…) a great many common concepts like causation and purpose are metaphorical, there is nonetheless an extensive range of metaphorical concepts. A sentence like “the balloon went up” is not metaphorical, nor is the old philosopher’s ‘”the cat is on the mat.” But as soon as one gets away from concrete physical experience and starts talking about abstractions or emotions, metaphorical understanding is the norm.[30]

A caveat:

So far, the studies we have looked at have not been able to deal well with the complexity and richness of fairly novel, poetic metaphors. Rather, they are concerned with establishing the basis of ordinary, conventional metaphorical constructs, such as

(5) my job is a jail


(6) I’m boiling with anger

which follow the standard A is a B structure. Because of this, there is a major inferential leap in theorizing. As Keysar, Boaz, Shen, Glucksberg, Horton (2000)[31] noted, as metaphors become lexicalized, they are no longer processed as metaphors, and it is clear that conventional metaphors are well on their way to becoming idioms in their own right. It may be the case that because of their familiarity, aptness, and conventionality, that they are understood just as quickly as their literal counterparts. Conversely, it may be the case that the structure of the metaphor used in these studies lends itself well to be interpreted as a simile, which in turn may allow for easier processing. In any case, there has been no systematic study done concerning the range of metaphors, and processing times for each. Without such systematicity, it is safer to limit the range and scope of metaphoricity to conventional verbal expressions.


Lakoff’s theory predicts that metaphors must be processed as quickly, and interpreted as directly as literal utterances. Coming out of this tradition, Gibbs (1990)[32] there have been numerous experimental studies supporting this claim. Suffice it to say that the most important feature of metaphor does not consist in its many verbal, linguistic instantiations, but its fundamental basis in our conceptual system. For Lakoff, linguistic metaphors are the direct manifestations of cross-domain mappings within our conceptual system, to which verbal metaphors are just “surface realizations.”[33] For example, Lakoff and his colleagues argue that we metaphorically transfer basic physical concepts such as up and over to other domains like our social interactions, emotions, science, and even mathematics. Lakoff and colleagues are interested in showing that many of our linguistic verbalizations exploit metaphorical conceptual structures. Even more ambitiously, Lakoff and his collaborators also attempt to show that many of our literal utterances are metaphoric at root. In order to buttress such a claim they point to common utterances such as: “your claims are indefensible”; or “he attacked every weak point of my argument.” Statements such as these purportedly reveal an underlying metaphorical concept that is utilized to express and make sense of a number of abstract ideas (i.e., knowledge, love). In the case just cited, the root metaphor is ARGUMENT IS WAR. The system of conceptual metaphor is largely unconscious, automatic, and used effortlessly, similar to our linguistic system.

Finally, many practitioners of language philosophy and linguistics maintain certain assumptions that are at odds with the views of the CMT. In brief, they are:

1. Correspondence theory of truth.

2. Meaning is equated to reference and truth.

3. Natural language semantics is characterized in terms of mathematical logic

Lakoff rejects 1 and instead develops an account of truth that is widely based on our embodiment and its relation to how we understand a given situation. Understanding, in this instance, is reliant on a series of experiential gestalts. These gestalts provide experience with the necessary pieces to structure our (metaphorical) concepts. Here, truth enters into the equation when our understanding of a sentence fits our understanding of a situation. This philosophical commitment necessarily leads to the rejection of premise 2 (in its traditional instantiation). If the truth of a situation is based on an experiential gestalt, it becomes increasingly difficult to explain how a sentence, for example, Bill is a bulldozer can actually say something meaningful (legendum: truthful) about the subject (Bill). This is due in part to the ephemeral nature of the utterance, the commitment of the speaker to the assertion, as well as the recipient’s charity to the description of Bill (seen as a bulldozer). Finally, Lakoff rejects 3 because Chomskian generative grammar does not permit metaphor into the logical form of semantics as it enters into grammar. Such a conception of grammar, claims Lakoff, makes an artifice out of natural language and is therefore untenable.

Interestingly, a number of continental philosophers have developed similar positions in terms of metaphoricity and truth-hood. Nietzsche regarded truth as a mobile army of worn-out metaphors, and we may read Lakoff making a similar claim (although somewhat less ambitiously). Paul Ricoeur held fast the idea that metaphor revealed the basic operations of the mind in its creative potential. The general idea tethering these theorists together is that metaphors are basic conceptual devices that have carved the paths toward understanding ourselves and the world around us. Our verbal instantiations range from the banal to the highly poetic and directly reflect our complex conceptual structures, whereby any account of language or thought is incomplete without directly addressing this phenomenon. At the risk of sounding too jocose, I want to conclude by reinforcing the point that metaphors are more than verbal wrapping paper; they are food for thought.


  1. Lakoff offers his model as an alternative to Chomskian generative grammar; this field has come to be known as cognitive linguistics and has numerous researchers working around the world producing a sizable amount of research. Briefly, Chomskian generative grammar holds that words intelligibly combine with one another based on hard and fast principles of Universal Grammar, found within a special-purpose language module. Cognitive linguistics denies this, and instead opts for a more general process whereby the way in which words meaningfully concatenate reflects “cognitive strategies for conceiving concrete situations.” On this account, conceptualization is defined as the unconscious manipulation of mental imagery of our embodied experiences. Concepts are first and foremost understood as coping mechanisms: they serve the purpose of coping with their subject matter. We use our most basic concepts to deal with our ever-changing environment and our relation to it. These mental images are called sensorimotor image schemas (i.e., experiential gestalts). These image schemas are redeployed in order to think about novel, ambiguous, or inherently unstructured subject matter in terms of something that is more concretely embedded in our cognitive hardware. It is safe to say that this notion of conceptualization takes us a bit off course that tradition has paved; it is more complex, more contextually malleable, and more intimately connected with behaviour and action than the traditional picture seems to grant.
  2. Aristotle. (2004) Rhetoric. (W.R. Roberts, Trans.) New York: Dover.
  3. Lakoff, G. (1993). The Contemporary Theory of Metaphor. In A. Ortony (Ed.), Metaphor and Thought (2nd ed., pp. 202-251). Cambridge: Cambridge University Press.
  4. “And the pool was filled with water out of sunlight, / And the lotos rose, quietly, quietly, / The surface glittered out of heart of light, /And they were behind us, reflected in the pool.
  5. Steen, G. (2000). Metaphor and language and literature: a cognitive perspective. Language and Literature, 9(3), 261-277.
  6. However, metaphors show variability across syntax and semantics. Syntactically, metaphors may not contain the subject being predicated (metaphorically), while others are adjectival (e.g., he drank the jovial wine). Semantically, the comparison between the items in the metaphor may be implicit (e.g., let slip the dogs of war, were ‘dogs’ are taken to mean ‘soldiers’). Finally, it is not always the case that metaphor achieves its effect through similarity. Rather, as I. A. Richards (1936) has pointed out, it may often be the dissimilarities between the target and source that produce a metaphor. Theoreticians’ inability to arrive at a standardized account aside, we may operationalize our understanding by providing ‘paradigmatic cases’: One means of standardization structures the paradigmatic cases as A is a B, where we may understand A and B as entities (concrete or abstract, noun or verb etc.), joined by a predicate.
  7. I use the word ‘things’ broadly in order to be sensitive to the fact that some metaphors do not follow the nominal (noun-noun) predicative structure (e.g., ‘he flew home’; ‘you gave me a great idea’).
  8. For consistency, I will henceforth use ‘topic’ to designate the subject of a metaphorical expression and ‘vehicle’ for the predicate.
  9. I use the standard notion (i.e., CAPITALS) to express propositions.
  10. Other forms of implicature include indirect speech.
  11. Searle, J. (1993). Metaphor. In A. Ortony (Ed.), Metaphor and Thought (pp. 83-111). Cambridge: Cambridge Univeristy Press; pg 89.
  12. Searle. Ibid, pp. 103.
  13. Searle. Ibid, pp. 103-107.
  14. Camp, E., & Reimer, M. (2006). Metaphor. In E. Lepore, & C. B. Smith (Eds.). The Oxford Handbook of the Philosophy of Language (pp. 845-863). New York: Oxford University Press, pp. 856.
  15. However, many claim (including Grice himself) that all he ever attempted to offer was a rational reconstruction, and was non-committal as the empirical ramifications of his theory.
  16. Gibbs, R. (1994). The Poetics of Mind: Figurative Thought, Language and Understanding. Cambridge: Cambridge University Press
  17. If metaphors can be true, in principle, there must also be false metaphors. What would be an example of a false metaphor? To think about the issue in terms of truth-conditionality seems a bit awkward.
  18. Lakoff (1980). Ibid.
  19. Lakoff (1980). Ibid, pp. 206.
  20. Lakoff, Ibid, pp. 206.
  21. Lakoff (1980). Ibid, pp. 207.
  22. In fact, this view is antithetical to the view that metaphors are only linguistic expressions.
  23. As in expressions such as ‘I see what you’re saying,’ his answer is clear,’ ‘his answer seemed murky.’
  24. Gibbs, R. W. (1984). Literal meaning and psychological theory. Cognitive Science, 8, 275-304.
  25. Gibbs, R. W. (1986). What does it mean to say that a metaphor has been understood? In R. Haskell (Ed.), Cognitive and symbolic structures: The psychology of metaphoric transformation (pp. 31-48). Norwood, NJ: Ablex.
  26. Ortony, A., Schallert, D., Reynolds, R., & Antos, S. (1978). Interpreting metaphors and idioms: Some effects of context on comprehension. Journal of Verbal Learning and Verbal Behavior, 17, 465-477.
  27. Gibbs (1984). Ibid, 100.
  28. Glucksberg, S., Gildea, P., & Bookin, H. (1982). On understanding nonliteral speech: Can people ignore metaphors? Journal of Verbal Learning and Verbal Behavior, 21, 85-98.
  29. For a more comprehensive survey of reaction-time studies, see Gibbs, R. W. (1994). The poetics of mind: figurative thought, language, and understanding. New York: Cambridge University Press.
  30. Lakoff. Ibid (1980) pp. 205.
  31. Keysar, B., Shen, Y., Glucksberg, S., & Horton, W. (2000). Conventional Language: How Metaphorical Is It? Journal of Memory and Language, 43, 576-593.
  32. Gibbs, R. W. (1990). Comprehending figurative referential description. Journal of Experimental Psychology: Learning, Memory & Cognition, 16, 56-66.
  33. Lakoff. Ibid, (1980), pp. 203.

Other Sources

Gibbs, R. W. (Ed.). (2008). The Cambridge Handbook of Metaphor and Thought. New York: Cambridge University Press.

Grice, H. (1975). Logic and Conversation. Syntax and Semantics, 3, 41-58.

Richards, I. (1966). Philosophy of Rhetoric (Reprint ed.). New York: Oxford University Press.

Ricoeur, P. (1977). The Rule of Metaphor: Multi-Disciplinary Studies in the Creation of Meaning in Language. (R. Czerny, K. McLaughlin, & J. Costello, Trans.) Toronto: University of Toronto Press.