ABSTRACT: In contemporary literary culture there is a widespread belief that ironies and paradoxes are closely akin. This is due to the importance that is given to the use of language in contemporary estimations of literature. Ironies and paradoxes seem to embody the sorts of a linguistic rebellion, innovation, deviation, and play, that have throughout this century become the dominant criteria of literary value. The association of irony with paradox, and of both with literature, is often ascribed to the New Criticism, and more specifically to Cleanth Brooks. Brooks, however, used the two terms in a manner that was unconventional, even eccentric, and that differed significantly from their use in figurative theory. I therefore examine irony and paradox as verbal figures, noting their characteristic features and criteria, and, in particular, how they differ from one another (for instance, a paradox means exactly what it says whereas an irony does not). I argue that irony and paradox — as understood by Brooks — have important affinities with irony and paradox as figures, but that they must be regarded as quite distinct, both in figurative theory and in Brooks’ extended sense.
In contemporary literary culture there is a widespread belief, or feeling, that ironies and paradoxes are closely akin. This is due in part to the huge importance that is given to the use of language in contemporary descriptions and estimations of literature. Ironies and paradoxes seem to reflect and embody the sorts of linguistic rebellion, innovation, deviation, and play, that have throughout this century become the dominant criteria of literary value.
The explicit association of irony with paradox, and of both with literature, is often ascribed to the New Criticism, and more specifically to Cleanth Brooks.
Brooks, however, used the two terms in a manner that was unconventional, even eccentric. He seemed to think of irony as a principle of order and unity: not so much a feature of language or meaning as a sort of coherence yoking disparate elements together, rather like Aristotle’s conception of wholeness and integrity in Poetics 8 (Brooks 1951). As for paradox, Brooks seemed to regard it as a quality in language very like Viktor Shklovsky’s defamiliarisation: that is, a deviation from conventional language designed to wrench our perceptions and our thoughts into unaccustomed, and therefore enlightening, pathways. Paradox, in this view, is a device which compensates for the limitations of conventional language, and is thus the only way in which poets can express the unconventional insights that are their stock in trade. Parodox, for Brooks, is not just useful and entertaining, but necessary. “Paradox”, he writes, “is the language appropriate and inevitable to poetry” (Brooks 1949, p.3, my italics).
Brooks was not, of course, the first to say this kind of thing, nor was the New Criticism the first to draw attention to irony and paradox as sources of literary value. The history of Romanticism is filled with similar sentiments, and they are among the factors that define what Romanticism is, or was. It was the first of the rebellions against the Enlightenment, and not least against the ideal of a Cartesian clarity of language. The early Wittgenstein — it is not altogether fanciful, however anachronistic, to see the early Wittgenstein as the last of the Enlightenment philosophers — said that everything that can be thought at all can be thought clearly, and everything that can be put into words can be put clearly (Wittgenstein 1961, 4.116). Nearly a century and a half earlier, Wordsworth had written that the best poetic diction was the language of ordinary men — the very language that, according to Wittgenstein, “disguises thought” (Wittgenstein 1961, 4.002). Keats, far from being impressed by the notion of thinking clearly, said that poets should be content with half-knowledge. Friedrich Schlegel spoke of “the impossibility … of total communication” (Schlegel 1797, 108). It was also Schlegel who explicitly connected irony with paradox: “Irony”, he wrote, “is the form of paradox” (Schlegel 1797, 48). And he it was who made the memorable, if gnomic, remark that irony can be defined as “logical beauty” (Schlegel 1797, 42).
The yoking of irony and paradox together has by now become such a commonplace that it is bound to arouse suspicion, and it is my purpose here to find out whether the connection is justified. The problem arises in part because none of the great works of Classical rhetoric — Quintilian, say, or Aristotle — connected them with one another. In Classical rhetoric they were merely figures of speech, and not especially important ones. If, in our own time, the two terms have acquired different, or extended, meanings, they are none the less derived from an originary source in figurative theory. It is therefore my intention to examine them primarily as figures. This, as we shall see, will also throw light upon the thinking of Cleanth Brooks and anyone else with similar views.
Irony as a figure of speech — verbal irony — has three necessary and sufficient conditions: (i) the speaker’s meaning is partly stated and partly unstated; (ii) the stated and the unstated meanings are in semantic contrast with one another; (iii) the meaning intended by the speaker, and understood by the listener, consists of the stated and the unstated meanings taken together. I will look at each of these in turn.
The first condition is also the best known. The definition of irony as saying one thing but meaning another is at least as old as Quintilian, and in our own time Paul Grice has examined the oddities of this kind of locution in some detail. Grice, however, has also made it clear that irony is a special case of saying one thing and meaning another. Very often irony does not occur at all. One of Grice’s own examples will illustrate this point (Grice 1989, p.32):
A. I am out of petrol.
B. There is a garage round the corner.
B’s reply means more than it says, and could be written in full as, “There is a garage round the corner, which I believe is still open, and you should be able to get petrol there”. The latter part of the reply, the unstated part, is what Grice calls an “implicatum”, and it is characteristic of most implicata that they complement and complete the stated part of an utterance. It is their function to make conversational sense of remarks which would otherwise seem irrelevant or tangential.
Sometimes, however, the unstated part of an utterance, instead of being semantically continuous with the stated part, is in semantic contrast with it. This is the mark of irony, and is the second condition of irony.
There are two relevant kinds of semantic contrast. One is a divergence between sense and reference. If I describe the wearing of a nose-stud as a “revolutionary outrage”, the sense and the reference clearly do not match. Nor do they match if I call a major gun-battle an “altercation”. In the first of these the description connotes more than, and in the second it connotes less than, what is required by the thing to which it refers. Ironies generated in this manner are “ironies of scale”, and in the language of traditional rhetoric they are, respectively, hyperbole and meiosis.
The other kind of semantic contrast occurs whenever the contrast between the stated and the unstated meanings is so complete that one is a negation of the other. The negation can take the form of a contradiction: for instance, Mark Antony’s statement that Brutus is an honourable man is contradicted by his unstated claim that Brutus is not an honourable man. Negation can also take the form of contrariety: for instance, if the statement “It’s a marriage made in Heaven” has the unstated meaning “It’s a marriage made for money”, these two assertions are contraries of one another. Contrariety and contradiction are types of logical opposition, so we can refer to both of these as “ironies of opposition”.
To sum up. I distinguish (i) ironies of scale, subdivided into hyperbole and meiosis, and (ii) ironies of opposition, subdivided into ironies of contradiction and ironies of contrariety. Both ironies of scale and ironies of opposition instantiate a semantic contrast between the stated and the unstated meanings of ironic utterances. Semantic contrast is thus a necessary condition of irony.
We now come to the third necessary condition, and we can best approach it by considering the difference between hyperbole and meiosis on the one hand, and bombast and euphemism on the other. The purpose of the latter is to conceal or disguise the truth. When a man who empties dustbins is called a sanitary engineer, this is designed to conceal or disguise the fact that his job is menial, smelly, and poorly-paid. When a dictator is called Our Great Leader this is meant to conceal his mediocrity and his fear. Both euphemism and bombast may have unstated meanings for at least some people, but their intention is really to minimise and eventually to abolish any unstated meanings. What is actually said is meant to replace what is not said. Their role, as Wittgenstein might have put it, is to disguise thought — and that is, in effect, to change from one thought to another.
It is quite otherwise in the case of hyperbole and meiosis, and, in general, in the case of irony. The purpose of all kinds of irony is to reveal the truth, or, at any rate, to focus our attention on it. It may seem an odd way of doing so — revealing or emphasising a truth by not stating it explicitly, by actually stating the contrary or the contradictory, or over or understating it. Yet that is how irony works, as we well know. Understanding irony requires a complex act of interpretation: not just an interpretation of the words uttered, but also an inference of the unstated meaning, and an understanding of the relation between the two. Many studies of irony suggest that we interpret an irony by mentally setting aside the stated meaning and replacing it by the unstated meaning. This is clearly, and fundamentally, wrong. The ironic statement “Brutus is an honourable man” does not have the same meaning as the non-ironic statement “Brutus is not an honourable man”. They may make the same assertion; they do not have the same meaning. The meaning of an irony is not determined by its truth conditions, but by an interaction between what is stated and what is not. If a slightly deaf Roman turned to his neighbour and asked what Mark Antony had said, and got the reply, “He said that Brutus was not an honourable man”, this would be a distortion, a sort of falsification, of the facts. The deaf Roman would not have been told a lie, but he would have been prevented from understanding what Mark Antony had said.
We now turn to paradox. This is a term that must be used with some care. In philosophy it is a topic that has provoked great interest and some disagreement: disagreement, in part, about how to resolve certain paradoxes, and in part about whether there are different kinds of paradox. In general, however, philosophers tend to give the name “paradox” to any statement which on the surface seems straightforward and innocuous but which turns out, on further examination, to have consequences which undermine some fundamental laws of logic, thought, or language.
Paradox as a figure is just the opposite. Here, the name refers to a statement which on the surface seems false, contradictory, or nonsensical, but which turns out, on further examination, to reveal a hitherto unconsidered truth. It is this that might seem to connect paradox with irony. In both cases, it seems, there is an explicit surface meaning and then a secondary, concealed meaning which is the real point of the utterance and constitutes its truth value.
However, this apparent similarity is misconceived. Let us look at some examples of paradox. George Bernard Shaw and G.K. Chesterton are prolific sources of paradox, but Oscar Wilde is better again. “It is only very shallow people”, he observed, “who do not judge by appearances”. Again, “Those who try to lead the people can only do so by following the mob”. Or, “Man is least himself when he talks in his own person. Give him a mask, and he will tell you the truth”.
Let us consider how we interpret the first of these: “It is only very shallow people who do not judge by appearances”. The initial or surface meaning is the meaning of the words themselves; there is no ambiguity or mystery here. However, what the words state conflicts with the conventional view that shallow people judge by appearances and wise people look beneath the appearance to the reality beneath. (It is interesting to find Platonism alive and well in the conventional wisdom of the man and woman in the street!)
How is the conflict resolved? It is resolved in Wilde’s favour. After a few moments’ thought, we might be led to reflect:
Shallow people do not realise that appearances are as much a part of things as what lies underneath, and that there is no reason to think that they are a less significant part. After all, appearances are produced by the things of which they are appearances. Experience and wisdom tell us that if the appearance is wrong to start with, it is usually the case that everything else is wrong as well.
In other words, we come to accept Wilde and to reject the conventional view. It must be noted that this does not involve the discovery of a concealed or unstated meaning. Rather, it involves an acceptance of the original statement. This is quite different from irony. In the case of irony there is a second, unstated meaning that must be taken together with the stated meaning. In the case of paradox the stated meaning is all that there is. From the point of view of their semantic structure, irony and paradox could hardly be more dissimilar. A paradox means exactly what it says; irony does not.
This explanation of figurative paradox brings us back to Cleanth Brooks. For Brooks, as I have said, “paradox” was the name of a quality very like the defamiliarisation or deautomatisation which we associate with the Russian Formalists and the Linguistic Circle of Prague. It is unlikely that Brooks was aware of these European thinkers when he was writing about irony and paradox, but he was working in the same kind of area. If Oscar Wilde’s paradoxes can be taken as paradigms, then their function is precisely that of alerting readers to hitherto unconsidered truths, and it is a function executed by writing something that flies in the face of conventional wisdom. So too, according to Brooks, Wordsworth wrote in a way that confounded conventional perceptions and beliefs, and replaced them by insights of greater exactitude and depth. If Wordsworth did not use language that is obviously paradoxical in the manner of Oscar Wilde, it is none the less arguable that its role in the economy of its readers’ knowledge and sensibility is much the same.
Irony is a different matter. Of course, I have dealt here only with verbal irony. Other kinds of irony are commonplace in literature — narrative irony in Swift, dramatic irony in Sophocles, conversational and intellectual irony in Plato — and all forms of Romantic irony depend on the supposition that there are some truths that cannot be stated, but which can be vaguely glimpsed or half-known through our encounters with words. If they cannot be put into words clearly, they can at least be hinted at by words, and these half-known truths are in some ways deeper and more important and more personal than other truths.
It is irony in this sense that was meant by Brooks. If I understand him correctly, he wanted to say that a hidden, glimpsed-at, half-known level of meaning produces the wholeness and integrity of a literary work. No matter how disparate, fragmented and circuitous its language and its surface meaning might be, there is a second, unstated layer of meaning which holds it together and gives it sense and coherence. As with verbal irony, the stated and the unstated meanings may dialectically conflict, but ultimately they combine to produce an integrated and meaningful whole.
In some ways, therefore, Brooks’s use of the terms irony and paradox are significantly close to their primary usage in Classical rhetoric and contemporary figurative theory. What is also clear is that there seems to be no justification for taking irony and paradox together. Whether as figures, or in the extended sense of Brooks and others, paradox and irony seem quite distinct. Paradox relies on the clarity and exactness of language; it shows that truth can be expressed by words alone. Irony uses words to point beyond language. Irony shows that there are some truths which, though they cannot be articulated in words, can none the less be expressed by means of words. Irony, like many other figures, is a way of transcending and ultimately extending the limited resources of everyday language, of ensuring that it does not disguise thought but is both the midwife and the medium of thought. Not everything that can be thought at all can be thought clearly, but everything that can be thought at all can be put into words.
– Brooks, Cleanth, 1949. “The Language of Paradox”, in Cleanth Brooks, The Well Wrought Urn. London: Dobson, pp.3-20.
– Brooks, Cleanth, 1951. “Irony as a Principle of Structure”, in Morton D. Zabel (ed), Literary Opinion in America. New York: Harper, 2nd edition, pp. 729-741.
– Grice, Paul, 1989. “Logic and Conversation”, in Paul Grice, Studies in the Ways of Words. Cambridge, MA and London: Harvard University Press, pp.22-40.
– Schlegel, Friedrich, 1797. “Selected Aphorisms from the Lyceum”, in Friedrich Schlegel, Dialogue on Poetry and Literary Aphorisms, translated by Ernst Behler and Roman Struc. London: Pennsylvania State University Press, 1968, pp.121-132.
– Wittgenstein, Ludwig, 1961. Tractatus Logico-Philosophicus, translated by D.F Pears and B.F.McGuinness. London: Routledge.
(This text was presented on the 20-th World Congress of Philosophy – 10 -15 August, Boston, Masachussets, in the section Philosophy and Litterature, on 12 August 1998)