LINGUIST List 28.1813

Thu Apr 13 2017

Review: Cog Sci; Ling Theories: Gibbs (2016)

Editor for this issue: Dhaval Niphade <>

Date: 07-Dec-2016
From: Eric Heaps <>
Subject: Mixing Metaphor
E-mail this message to a friend

Discuss this message

Book announced at

EDITOR: Jr. Gibbs
TITLE: Mixing Metaphor
SERIES TITLE: Metaphor in Language, Cognition, and Communication 6
PUBLISHER: John Benjamins
YEAR: 2016

REVIEWER: Eric Heaps, Indiana University Bloomington

Reviews Editor: Helen Aristar-Dry


Raymond Gibbs was inspired to put together “Mixing Metaphor,” a collection of essays, after an experience with psychology students several years ago. After asking the students to write down anything they wanted on the topic of metaphor, a rather vague, open-ended question that could have gone in many directions, he found that 40% of all respondents referred to the illegitimacy of mixing metaphors. When seeking further explanation from the students, he discovered that this statement stemmed from explicit teachings in high school related to books and essays on writing style and composition. Upon further consideration, Gibbs found it ironic that this most notable of beliefs about metaphor is so rarely studied in the world of metaphor scholarship, a vast field that covers the manifold ways in which metaphor is manifest in human experience as a fundamental scheme of thought.

Starting with the hypothesis that the sheer commonality of mixed metaphors indicates that they are something more than failed attempts at humor, bad writing, or worse, sheer cognitive error, Gibbs encouraged authors to write anything they wanted in connection with mixed metaphor, allowing them to create this diverse body of work, using a variety of empirical findings and theories of metaphor. By doing so, he hoped to bring mixed metaphor to center stage, encouraging the audience of the metaphor community to build and expand this particular section of discourse.

He divides the volume into three sections, each composed of four chapters. The first part of the volume examines whether mixed metaphors cause problems at all. Following that, the second part examines how and why mixed metaphors are used. Finally, the third part looks at ways in which mixed metaphors are realized in different types of discourse.

Chapter 1: “A view of mixed metaphor within a conceptual metaphor theory framework” by Zoltán Kövesce (pp. 3-16)

Several metaphor scholars insist that conceptual metaphor theory (CMT) cannot support the existence of mixed metaphors, since once a conceptual metaphor is activated, only a homogeneous metaphor can follow. In this chapter, Kövesce argues that CMT accurately predicts the existence of mixed metaphors in natural discourse. Building on the work done by Kimmel (2010), in which Kimmel argues that the comprehensibility of mixed metaphor can be thoroughly understood through close analysis of clausal structure, Kövesce seeks to address some unexplained issues, specifically why we use mixed metaphors, why they are so common, and how we comprehend them. He proposes that “all (target domain) concepts consist of a number of different aspects” and that these aspects are conceptualized using a large variety of source domains. For instance, FAMILY includes a variety of aspects such as parents, children, child-raising, and many more. Each of those aspects employs different source domains traditionally and can be explained through other non-traditional domains as well. As such, mixed metaphors provide a richer understanding of the concept we’re trying to explain by acknowledging the different aspects of that domain. Our comprehension of discourse that is communicatively and stylistically neutral, which Kövesce argues must of necessity be mixed, is largely processed without difficulty due to low degree of activation involved with the different concepts. He urges the continued exploration of this hypothetical model that is compatible with both CMT and the comprehension of mixed metaphor.

Chapter 2: “Mixed metaphor from a discourse dynamics perspective: A non-issue?” by Lynne Cameron (pp. 17-30)

In this chapter, Cameron continues to build on her model of discourse dynamic perspective, or “how people use [metaphors] in the flow of situated text and talk,” specifically examining the frequency of mixed metaphors juxtaposed with the infrequency of difficulty making sense of them. She particularly mentions that her findings relate to spontaneous oral discourse, which is distinguished from writing in that the text is not visible when spoken, and comprehension is a quick, almost immediate process. Specifically, Cameron chooses to analyze ‘metaphor clusters’, portions of oral discourse with a significantly higher number of metaphors than the talk surrounding the cluster. She states that “speakers explore and develop ideas as they speak about them.” This premise illustrates that even when a speaker starts with a clear idea of what they want to say, putting that idea into words and seeing those words’ effects on the audience can produce adjustments in the moment, which creates shifts in metaphor that produce mixed metaphors. Cameron’s use of discourse dynamics is specifically to combat generalizations in CMT, preferring instead to emphasize the socio-cultural elements of language in use, which she argues eliminates any ‘problem’ with mixed metaphor and instead insists that multiple metaphors are “necessary and inevitable.”

Chapter 3: “Why mixed metaphors make sense” by Cornelia Müller (pp. 31-56)

Müller argues that the use of mixed metaphor in speaking, writing, and acted gestures is not only completely reasonable but beneficial. Through the use of mixed metaphors, makers of discourse are able to demonstrate a flexibility of understanding through shifts to different parts of metaphorical mappings. While this type of dynamic understanding of metaphor is not favored in studies of style and rhetoric, Müller claims that mixed metaphors are based on “the foregrounding of uncommon aspects of meaning,” that their active intentionality belies the simplistic approach traditional methods take in treating mixed metaphors as problematic. Very often, mixed metaphors are what Linguistic Metaphor Theories refer to as ‘dead’ metaphors, metaphors whose meanings have been conventionalized and do not simultaneously contain the original literal meaning and the figurative one. Along with other proponents of CMT, Müller points out that these dead conceptual metaphors govern our cognition. She goes beyond CMT, however, in suggesting that we consciously remap uncommon meanings on mixed metaphor, rather than as a kind of subconscious entailment to the activated metaphor. In this process, although the different source domains for the metaphors may clash, hidden meaning can be activated which in turn will lend itself to semantic similarity. In effect, Müller’s dynamic view coordinates with Cameron’s discourse dynamics, but posits a more active control of metaphoricity.

Chapter 4: “Tackling mixed metaphors in discourse: Corpus and psychological studies” by Julia E. Lonergan and Raymond W. Gibbs, Jr. (pp. 57-74)

In this chapter, Lonergan and Gibbs examine the corpus of the “Block that Metaphor!” column from The New Yorker. First, they analyzed the composition of excerpted mixed metaphors, finding that most had been employed in other kinds of discourse and partially confirming their hypothesis that “people interpret mixed metaphors as being meaningful and coherent because of their abilities to engage in elaborate reasoning about the source domains explicitly mentioned in the texts.” In a second study, they describe what is perhaps their most important finding, namely that people consistently give the same reading of phrases considered problematic due to the mixed state of their domains. This is accomplished through an integration of the various metaphors into a coherent whole via assumptions made by the audience. In conclusion, Lonergan and Gibbs assert that mixed metaphor moves beyond CMT to include more extensive inferences to a variety of source domains in the audience’s social and cultural knowledge.

Chapter 5: “Mixed metaphor: Its depth, its breadth, and a pretense-based approach” by John Barnden (pp. 75-112)

In this chapter, Barnden presents an overview of his AI implementation of the ATT-Meta approach to metaphor he developed with Mark G. Lee. He uses the model to handle a variety of types of mixed metaphor, such as chaining, which he calls serial mixing, parallel mixing (different metaphorical views on same target), and combinations of the two. What he calls parallel mixing is the traditional idea where A is both B and C at roughly the same time. His inclusion of serial mixing is to allow for situations where A is B and B is C at roughly the same time. This model uses the nesting of pretence worlds to examine discourse and different patterns of mixed metaphor. In these pretence worlds, modeled users pretend that everything said is literally true, using inference to build more complex mappings of metaphors run through the program. By doing so, Barnden argues that these deep inferences based on assumptions of the source topic and understanding of its domain allow for metaphor processing. This runs counter to the traditional view that metaphor understanding is derived from parallel mappings between a source and target.

Chapter 6: “Mixed metaphor is a question of deliberateness” by Gerard Steen (pp. 113-132)

In this chapter, Steen argues that although the clash of images elicited by the use of mixed metaphor is often called poor writing, these metaphors are usually deliberately constructed for specific rhetorical purposes. At the same time, Steen is not insisting that most of this metaphor use is conscious, but rather recognizing the goal-directed nature of language that is typically unconscious, though in the cases he studies, still deliberate. In this way, Steen uses deliberate to describe metaphors that have a specific rhetorical purpose in their context. As listeners and readers recognize this deliberateness of purpose, they are able to infer rich understandings of the message being presented. Although there might be non-deliberate uses of metaphor that appear mixed, Steen suggests that paying heed to the possible deliberateness is an integral part of a three-dimensional approach to metaphor that includes language, thought, and communication.

Chapter 7: “When language and cultures meet: Mixed metaphors in the discourse of Spanish speakers of English” by Fiona MacArthur (pp. 133-154)

MacArthur examines a type of mixed metaphor that she labels “hybrid” metaphor, where a user employs conceptual understandings from a foreign language in their attempts at communication. Although this can occur with native speakers, it is far more common among non-native speakers of a language who are attempting to get across an idea which they are struggling to communicate in their non-native language. MacArthur starts by refuting Pesmen’s (1991) argument that mixed metaphors are condemned because discourse is intended to show coherence in the culture and world view of the speakers, an inherent characteristic of an autonomous closed system. Specifically MacArthur points out that languages and cultures are most definitely not autonomous closed systems, especially when viewed in the context of foreign language learners. In the early stages of language learning, metaphor is largely absent from the new speaker’s vocabulary, what little there is consisting of prepositional metaphors rather than those based in nouns and verbs. However, when the demands of fluency require speakers to do things like state their personal views, they begin to produce metaphors, usually based on those found in their native tongue. Perhaps the most important finding MacArthur relates is the fact that ad hoc meaning created when listeners focus on meaning instead of content demonstrates that strict adherence to conventional forms is not entirely necessary for proficiency as competent understanding can be achieved between listener and speaker. What’s more, in a global world where English is used as a vehicle for international communication, insistence on conventional metaphors is not as effective as fomenting greater understanding of metaphor use in general to increase comprehension of these hybrid metaphors.

Chapter 8: “The ‘dull row’ and the ‘burning barbed wire pantyhose’: Complex metaphor in chronic accounts of pain” by Charles Charteris-Black (pp. 155-178)

Charteris-Black examines how those who experience chronic pain are able to lend credibility to and elicit greater understanding of their pain via the use of mixed metaphor. Charteris-Black prefers to use the term ‘purposeful metaphor’ to describe what is called deliberate metaphor by Steen earlier in this collection. The use of this type of metaphor lends greater control and power to the person in pain, who otherwise feels imprisoned by their illness. Specifically, mixed metaphors describe pain or illness as being out of control, liberating the speaker from what guilt they may feel about being in their situation, as they simultaneously work toward repeated and extended metaphors as they assert control. Further, these strong feelings are able to communicate publicly what is otherwise an extremely private experience, by creating an embodied simulation as described by Gibbs (2006) and Semino (2010).

Chapter 9: “We drank with our eyes first: The web of sensory perception, aesthetic experiences, and mixed mappings in wine reviews” by Carita Paradis and Charles Hommerberg (pp. 179-202)

Paradis and Hommerberg explore how mixed metaphor is almost expected in the discursive practice of wine reviews as a way to evoke the mixed, sensory experience associated with the initial wine tasting. The use of a variety of source domains to describe a variety of different sensations (taste, sight, smell, touch) helps the writer move from that initial sensory experience through the process of thinking about it and finally into the realm of language. This is important as their rhetorical purpose is to activate kinesthetic and sensorial experiences for their readers. Beyond that, the domains of metaphoric imagery employed reflect the source culture and help create a sense of belonging for readers.

Chapter 10: “A corpus-based study of ‘mixed metaphor’ as a metalinguistic comment” by Elena Semino (pp. 203-222)

In this chapter, Semino conducts a study of 141 occurrences of the expression ‘mixed metaphor’ in the Oxford English Corpus, specifically analyzing the way they’re employed in different genres and grammatical forms, and the different rhetorical purposes for their creation. After recognizing a variety of deliberate uses of mixed metaphor in which the user stated a rhetorical purpose, ranging from negative assessments to displays of humor or creativity, Semino questions whether or not mixed metaphor is even a “viable and operationable technical term.” She points this out because although the vast majority of cases fit the ‘folk’ concept of mixed metaphor as negative, a minority instead humorously point out witty uses of metaphor that are effective, not negative. One particular conclusion of note is that Semino points out that we are more aware of the contrast between metaphors that are mixed when they are similar to one another. In so doing, she doesn’t intend to dismiss the conclusion by other authors in the book that close proximity does not necessarily limit understanding, but rather to point out something that must be taken into account when studying comprehension and mixed metaphors.

Chapter 11: “Mixing in pictorial and multimodal metaphors?” by Charles Forceville (pp. 223-240)

After a thorough examination of examples of non-linguistic mixing of domains, Forceville argues that these uses are much more conscious and deliberate than linguistic constructions, and are typically employed for important functional and aesthetic purposes, and can especially be used to quickly grab an audience’s attention. Beyond that, he argues that these instances of mixed metaphor can be best explained using conceptual blending theory. However, Forceville eschews the term mixed metaphor and its negative semantic impact when describing these pictorial instances, preferring instead ‘multiple source domain metaphors,’ as in all instances these constructions are so artfully done as to not seem allied to the often unintentional mixed metaphors found within verbal clauses. As pictures do not have grammar, Forceville finds there is little to be gained through analysis of a term that depends on grammar.

Chapter 12: “Extended metaphor in the web of discourse” by Anita Nasiscione (pp. 241-266)

Nasiscione specifically examines the stylistic use of lexical metaphor and metaphorical phraseological units ranging back to Old English. Her conclusion is that extended metaphor is a regular element of thought, an interrelationship of metaphor and metonymy that provides stylistic and semantic coherence to the text in which it is found. This naturally implies that the idea that mixed metaphor is impermissible is incorrect, as extended metaphor is by nature mixed and the study of extended metaphor is essential in understanding meaning construction. If each extended metaphor consists of a base metaphor and a series of sub-images, those sub-images with different domains will invariably add to the understanding of the extended metaphor while making it mixed.


In this volume, Gibbs is able to bring together a variety of established metaphor scholars to create a survey of current research in mixed metaphors via focused articles related to those scholars’ work. Although the authors come from the metaphor community, their work is of interest to linguists in a variety of other specializations, especially since each of these scholars uses techniques and theories from different areas of linguistics to advance their own theories. As the first major edited volume to specifically focus on mixed metaphor, this book serves to provide reference and access to the wide-ranging work of these scholars to those who might not otherwise delve into the scholars’ monographs.

Overall, the articles coordinate with each other, providing counterpoints to each other’s arguments and fleshing out the evidence for their various conclusions. One point at which this coordination fails is in terminology. While the choice of terms is of significant importance in a field focused on language, the use of different terms for the same thing can make it difficult for the reader to connect some of the articles without a sort of mental gymnastics. For example, where Steen and many of the authors, including the editor, refer to deliberate metaphors, Charteris-Black opts for the term ‘purposeful metaphors,’ perhaps obscuring the strong connections between his work and Steen’s. While allowing for changing terminology in an evolving field, greater care could be taken within an edited volume to ease understanding of the reader and provide greater coherence to the work.

Most impressive from the volume is the individual authors’ insistence on their work as a starting point for research in mixed metaphor, rather than a conclusion. How can we further develop conceptual metaphor theory while accounting for mixed metaphor? How does this research change views of comprehension in language acquisition? What is the value of our current terminology with respect to mixed metaphor? These are just a few of the questions asked by the authors that provide starting points for future research.

This volume is a valuable reference for anyone interested in the purpose and structure of metaphors. Although at times the specifics of an author’s argument may become too technical for a lay reader, the conclusions are succinct and well-written and could even provide an entry point for those with less knowledge of metaphor theory.


Gibbs, R. W.J. (2006). Embodiment and cognitive science. New York: Cambridge University Press.

Kimmel, M. (2010). Why we mix metaphors (and mix them well): Discourse coherence, conceptual metaphor, and beyond. Journal of Pragmatics, 42, 97-115.

Lee, M.G., & Barnden, J.A. (2001). Reasoning about mixed metaphors with an implemented AI system. Metaphor and Symbol, 16(1&2), 29042. Doi: 10.1207/S15327868MS1601&2_3

Pesmen, D. (1991). Reasonable and unreasonable worlds: Some expectations of coherence in culture implied by the prohibition of mixed metaphor. In J.W. Fernandez (Ed.), Beyond metaphor: The theory of tropes in anthropology (pp. 213-243). Stanford, CA: Stanford University Press.

Semino, E. (2010). Descriptions of pain, metaphor, and embodied simulation. Metaphor and Symbol, 25, 205-226. Doi:10.1080/ teachin�


Eric “C” Heaps is a PhD candidate at Indiana University where he focuses on theatre translation, investigating methods for translating theatre texts and performance not only from language to language but from culture to culture in the eventual production of the text. He does so using tools from translation studies, somatic studies, and cognitive linguistics, among other fields. In the past few years he has translated Gil Vicente’s ''Auto da Barca do Inferno,'' Augusto Boal’s ''Torquemada,'' and three plays by Lucienne Guedes Fahrer: ''A Recusa da Flor,'' ''Eu Não Esperava Tanta Humanidade,'' and ''Vendaval.''

Page Updated: 13-Apr-2017