Part-whole relationships
Print Friendly, PDF & Email

Presented at The Developing Group, 3 Aug 2002

“In general the part differs from the whole.  
The part cannot totally contain the whole.  
But it always partially contains the whole.  
The part contains the whole to some degree.”
Bart Kosko, Fuzzy Thinking, p 58.


We have often said that metaphor is fundamental, not only to language but to cognition in general. Metaphor works because we notice a similarity between two different kinds of experience. Generally, the similarity is in the relationship between parts of the experiences.  

There is another, less well known, process that seems to be equally fundamental to language and cognition, that of metonymy.  Metonymy enables us to use one part or aspect of an experience to stand for some other part (or the whole) of that experience. Unlike metaphor which involves two domains of experience, metonymy only requires one. Unlike metaphor which is based on similarity, metonymy requires contiguity, i.e.  ‘closeness’ of association. Most metonymies are so common we never notice them.

Take for example, someone showing you a photograph of the face of a little girl and saying “That’s my daughter.” You smile and never think that you’ve used a metonymic process to comprehend that the face of a person stands for the whole person. Imagine you had been shown a picture of the girl’s foot and they had said “That’s my daughter!” Yet, if someone says “Get your butt over here,” again by processing metonymically, we know exactly what they mean.

In Metaphors We Live By, George Lakoff and Mark Johnson devote a complete chapter to metonymy and define it as:

“Using one entity to refer to another that is related to it.” (p.35)


The ham sandwich is waiting for his check. (= the person who ordered the ham sandwich)

He likes to read the Marquis de Sade. (= the writings of the Marquis)

He’s in dance. (= the dancing profession)

The Times hasn’t arrived at the Press Conference yet. (= the reporter from The Times)

Mrs. Grundy frowns on blue jeans. (= the wearing of blue jeans)

Metonymic concepts like these are systematic in the same way that metaphoric concepts are.  The sentences given above are not random. … Thus, like metaphors, metonymic concepts structure not just our language but our thoughts, attitudes and actions.  And, like metaphoric concepts, metonymic concepts are grounded in our experience.  In fact, the grounding of metonymic concepts is in general more obvious than is the case with metaphoric concepts, since it usually involves direct physical or causal associations. (Metaphors We Live By, p. 39) 

Kinds of metonymy

A special case of metonymy, PART FOR THE WHOLE, is known as is synecdoche:

The automobile is clogging our highways. (= the collection of automobiles)

We need a couple of strong bodies for our team. (= strong people)

There are a lot of good heads in the university. (= intelligent people)

I’ve got a new set of wheels. (= car, motorcycle, etc.)

We’ve got some new blood in the organisation. (= new people)

There are, however, other kinds of metonymy:

“The book is moving right along.” contains an instant of a common metonymy of PRODUCT FOR PROCESS, in which the book, the product of the activity of writing, stands for the activity itself. Without the metonymy of the book for the writing of the book, one would say “the writing is moving right along.” Lakoff and Johnson, Philosophy in the Flesh,  p 203.

A PUNCTUATION MARK stands both metonymically and metaphorically for the meaning of that punctuation mark. For example, a question mark indicates something unknown, as in “He’s a big question mark to me.”Because a period indicates the end of a sentence, the word period indicates the end of what is to be communicated, as in “Be home by midnight — period!” Grammatical morphemes are included as well: “I want this done — no ifs, ands or buts!”. Lakoff and Johnson, Philosophy in the Flesh p. 245.

Other common metonymies given in Metaphors We Live By (pp. 38-39):


He bought a Ford.
He’s got a Picasso.


The sax has the flu today.
The buses are on strike.


Napoleon lost at Waterloo.
A Mercedes rear-ended me.


Exxon has raised its prices again.
You’ll never get the university to agree to that.


The White House isn’t saying anything.
Wall Street is in a panic.


Remember the Alamo.
Watergate changed our politics.

[JL: To my way of thinking these other kinds of metonymies still have a PART-WHOLE relationship. From a practical point of view we can spot them all by simply considering whether the word actually stands for something bigger, without needing to be concerned about the other categories.]

On the relationship between metaphor and metonymy

From Lakoff & Johnson, Metaphors We Live By (p. 36-37):

They are “different kinds of processes.  Metaphor is principally a way of conceiving of one thing in terms of another, and its primary function is understanding.  Metonymy, on the other hand, has primarily a referential function, that is, it allows us to use one entity to stand for another. But metonymy is not merely a referential device. It also serves the function of providing understanding. 

For example, in the case of the metonymy THE PART FOR THE WHOLE there are many parts that can stand for the whole.  Which part we pick out determines which aspect of the whole we are focusing on.  When we say that we need some “good heads” on the project, we are using “good heads” to refer to “intelligent people”.  The point is not just to use a part (head) to stand for a whole (person) but rather to pick out a particular characteristic of the person, namely, intelligence, which is associated with the head.  The same is true for other kinds of metonymies. …

Thus metonymy serves some of the same purposes that metaphor does, and in somewhat the same way, but it allows us to focus more specifically on certain aspects of what is being referred to.  It is also like metaphor in that it is not just a poetic or rhetorical device.  Nor is it just a matter of language.  Metonymic concepts (like THE PART FOR THE WHOLE) are part of the ordinary, everyday way we think and act as well as talk.  

For example, we have in our conceptual system a special case of the metonymy THE PART FOR THE WHOLE, namely, THE FACE FOR THE PERSON.  For example:

She’s just a pretty face.
There are an awful lot of faces out there in the audience.
We need some new faces around here

From Raymond W. Gibbs Jr. (‘Process and Products in Making Sense of Tropes’ in Metaphor and Thought, edited by Andrew Ortony.) pp. 258-259:

Metaphor is based on similarity whereas metonymy expresses simple contiguous relations between objects, such as part-whole, cause-effect, and so on.

… These two figurative types can be distinguished because the connections made between things are different in each case.  In metaphor, there are two conceptual domains and one is understood in terms of the other.  For instance, when a boxer is compared to a cream puff, as in “The boxer was a cream puff,” two separate conceptual domains are contrasted (athletes and food) and the fighter is viewed as similar to a pastry in being soft and easy to devour.  Metonymy involves only one conceptual domain in that the mapping or connection between two things is done within the same domain.

Traditional rhetoric defines metonymy as a figure of speech wherein the name of one entity is used to refer to another entity that is contiguous to it.  This process of transferred reference is possible [because of] a referring function.  Thus, referring to a baseball player as a glove, as in “We need a new glove at second base,” uses a salient characteristic of one domain (the glove part of the baseball player) to represent the entire domain (the player).  When the two things being compared form a part-whole relationship (that is, when glove is part of the whole baseball player), the metonymic expression is often referred to as synecdoche.

… Many [metonymic] models depend on conventional cultural associations (e.g. “Let’s not let Iraq become another Viet Nam”) which reflect the general principle that a thing may stand for what it is conventionally associated with. This principle limits the use of metonymy to only certain relationships between entities. For example, we can use the name of any well-known creative artist to refer to the artistic creations of the artist as in “Does he like Hemingway?” or “I saw a Jasper Johns yesterday” but not any product can be referred to by the name of the person who created the product.  I could hardly say “Mary was tasty” meaning by Mary the cheesecake that Mary made. 

… Any given instance of a referring function needs to be sanctioned by a body of beliefs encapsulated in an appropriate frame.  Thus, one widespread belief in our culture is that the distinctive value of a work of art is due uniquely to the genius of the individual who created it.  But we do not normally believe that such a relationship always holds between a cake and the person who baked it.

Metonymic models

George Lakoff, Women, Fire and Dangerous Things, pp. 77-79:

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 it to stand either for the thing as a whole or for some other aspect or part of it. … The kind [of metonymic models] of most interest for [categorization] are those in which a member or subcategory can stand metonymically for the whole category for the purpose of making inferences or judgements.

[JL: In Symbolic Modelling we make use of an equivalent process when we use an example of a behaviour for the whole class of behaviour. But then every example is an example of metonymy because an example is standing for the whole class of experience of which that example is a part.]

In general, a metonymic model has the following characteristics:

    • There is a “target” concept A.
    • There is a conceptual structure containing both A and another concept B.
    • B is either part of A or closely associated with it in that conceptual structure.  Typically, a choice of B will uniquely determine A, within that conceptual structure.
    • Compared to A, B is either easier to understand, easier to remember, easier to recognise, or more immediately useful for the given purpose.
    • A metonymic model is a model of how A and B are related in a conceptual structure; the relationship is specified by a function from B to A.

Categorization using metonymy

George Lakoff, Women, Fire and Dangerous Things, pp. 84-88:

Social Stereotypes:

A politician is conniving, egotistical, and dishonest.
A bachelor is macho, and dates a lot of different women.
The Japanese are industrious, polite and clever.

Typical examples:

Robins and sparrows are typical birds.
Apples and oranges are typical fruits.
Saws and hammers are typical tools.


The ideal husband is a good provider, faithful, strong, respected, attractive. [Whereas] the stereotypical husband is bumbling, dull, pot-bellied.


The Cadillac of vacuum cleaners.
The Michaelson-Morley experiment is the paragon of physics experiments
[He’s another David Beckham.]

[JL: Another way of thinking about metonymic categories is to consider them instances of “reference-point reasoning” (Rosch). That is we start from a commonly recognised reference point — a part — and use this to infer something about the whole category.]

On part-whole relations in general

Gregory Bateson:

A human being in relation with another has very limited control over what happens in that relationship.  He is a partof a two-person unit, and the control which any part can have over any whole is strictly limited … The contrast between part and whole, whenever this contrast appears in the realm of communication, is simply a contrast in logical typing.  The whole is always in a meta-relationship with its parts.  As in logic the proposition can never determine the meta-proposition, so also in matters of control the smaller context can never determine the larger. (Steps to an Ecology of Mind, p 267.)

There is that hierarchy of differences which biologists call “levels.”  I mean such differences as that between a cell and a tissue, between tissue and organ, organ and organism, and organism and society.  These are the hierarchies of units or Gestalten, in which each sub unit is a part of the unit of the next larger scope.  And, always in biology, this difference or relationship which I call “part of” is such that certain differences in the part have informational effect upon the larger unit, and vice versa. (Steps to an Ecology of Mind, p. 458)

“There are no parts.” (Capra)  But “Without differentiation of parts, there can be no differentiation of events or functioning.” (Mind and Nature, p. 99)

[A] form of very primitive coding … is part-for-whole coding.  For example, I see a redwood tree standing up out of the ground and I know from this perception that underneath the ground I shall find roots, or I hear the beginning of a sentence and know at once from that beginning the grammatical structure of the rest of the sentence and may very well know many of the words and ideas contained in it.  We live in a life in which our percepts are perhaps always the perception of parts, and our guesses about wholes are continually being verified or contradicted by the latter presentation of other parts. It is perhaps so, that wholes can never be presented; for that would involve direct communication. (Mind and Nature, pp. 121-122)

Ken Wilber, Sex, Ecology, Spirituality:

This is a book about holons — about wholes that are parts of other wholes, indefinitely.  Whole atoms are parts of molecules; whole molecules are parts of cells; whole cells are parts of organisms, and so on. Each whole is simultaneously a part, a whole/part, a holon and reality is composed, not of things nor processes, nor wholes nor parts, but of whole/parts, of holons. (p. viii)

Tenet No. 1: Reality as a whole is not composed of things or processes, but of holons.  Composed, that is of wholes that are simultaneously parts of other wholes with no upward or downward limit.  To say that holons are processes instead of things is in some ways true, but misses the essential point that processes themselves exist only within other processes.  There are no things or processes, only holons.

Since reality is not composed of wholes, and since it has no parts — since there are only whole/parts — then this approach undercuts the traditional argument between atomism (all things are fundamentally isolated and individual wholes that interact only by chance) and wholism (all things are merely strands or parts of the larger web or whole).  Both of those views are absolutely incorrect.  There are no wholes, and there are no parts.  There are only whole/parts.

This approach also undercuts the argument between the materialist and the idealist camps.  Reality isn’t composed of quarks, or bootstrapping hadrons, or subatomic exchange; but neither is it composed of ideas, symbols, or thoughts.  It is composed of holons. (p. 35)

Freud referred to the primary process, which is governed by two general laws, that of displacement and that of condensation. In displacement, two different objects are equated or “linked” because they share similar parts or predicates (a relation of similarity).  In condensation, different objects are related because they exist in the same space (a relation of contiguity: a lock of hair of a great warrior “contains” in condensed form the power of the warrior.  There is another way to categorize this.  The former is metaphor, the latter is metonymy. (p. 217)

These two figures of speech, based on similar agency (metaphor) and similar communion (metonym) — or simply similarity (agency) and contiguity (communion) — are, as linguists have pointed out, the most basic holons of linguistic communication.

Interestingly, these two basic figures of linguistic cognition are so fundamental that Jakobson found, in the two major forms of childhood aphasia (loss of power to understand speech), that in one the child loses the capacity to understand metaphor, but grasps metonym quite well, whereas in the other the child cannot understand metonym but perfectly grasps metaphor.  Jakobson actually called them “similarity disorder” and “contiguity disorder” …

Given that, for Freud, the primary-process cognition operates primarily with metaphor and metonym, and given that neurosis is basically a failure to outgrow the primary process in certain important ways, it is a small step to Lacan’s formulations, under the banner of structuralism, in which symptom is a metaphor and desire a metonym. … Thus, as Lacan uses the words, displacement is a metonymy that marks the nature of the subject’s desire, which is primarily a lack; condensation is a metaphor that traces the repressed meaning of desire through sliding chains of signification, found in symptoms.  Same fundamental figures of speech (metaphor and metonym), which I accept, but tied to a ‘single boundary’ theory of ‘lack’ that I reject entirely. (pp. 589-590)

[JL: Lacan uses them the opposite way round!]

Finally, Cei Davies Linn has written in a newsletter:

Metonymy definition: A figure of speech that substantiates the name of a related object, person or idea for the subject. For example, crown for monarchy or Shakespeare for the Works of Shakespeare.

Clinical implication: Unlike metaphor, metonymy has a “horizontal” motif of moving across time [Jakobson]. Metonymy is rather like a hologram, a part of something represents the whole. When working with metonymy, a large expanse of time needs to be considered. For example: a client’s experience of depression is experienced in terms of a black cloud. If there is a predisposing family history of depression, instead of deepening the black cloud, as the therapist would in quadrant two, this black cloud belongs to previous generations of depression and will be addressed by asking quadrant four questions. Quadrant four questions pull the information, the black cloud, back in time to the first owner of the cloud and the original situation that caused it. The original situation is then healed of imposing the cloud on subsequent generations.

[JL: I have to say, I am not at all clear how Cei links metonymy with quadrant four pulling back. Any ideas?]

Print Friendly, PDF & Email
body * { color: inherit !important; }