PRESUPPOSING

The notions of "presupposing" and of contextual implication, which we shall compare and contrast in what follows, have come to play increasingly prominent roles in the philosophical literature of the English-speaking world since the 1940s. This development is not accidental but arises from the stress the twentieth century put upon analysis as a fundamental mode of philosophical inquiry. The notions of presupposing and of contextual implication play both negative and positive roles within this general orientation. Negatively, they are devices that contemporary thinkers employ in order to minimize the tendency of philosophers and other reflective persons to view the world in terms of oversimplified conceptual models. Positively, they function as instruments in the dissection and ultimate understanding of certain human activities, especially those that involve the efforts of human beings to communicate with one another, as in promising, stating, saying, implying, a task that, some philosophers feel, is hindered or obstructed by the natural disposition of reflective individuals to subsume such activities under excessively simple descriptions. The appeal to the notions of presupposing and of contextual implication has thus served to widen—and at the same time to make more accurate—our conceptions of the circumstances in which human communication takes place. This entry will describe the history (all of it relatively recent, of course) of the major developments that have taken place with regard to these subjects, and will in this way attempt to bring out their essential features.

Similarities and Differences

It is no simple matter to show why presupposing and contextual implication are two separate concepts, since the differences between them are subtle. Most writers have, in fact, not discriminated between them, in part because both notions are slippery but also because they have similar functions. Their similarities may be elucidated as follows. If we distinguish between what a person explicitly states, or asserts, when he utters certain words in certain circumstances and what he (or perhaps his statement) implies, then the concepts of presupposing and of contextual implication belong to the latter category rather than to the former. This crude distinction must be refined further, however, for the sense of "implies" that is being marked out here is not that of logical implication in any of the various senses of that term—for example, the sense involved in saying that "X is a husband" implies "X is married." Indeed, both presupposing and contextual implication are to be contrasted with logical implication.

The kinds of implications that fall into this category may be indicated by simple examples. In saying "alas!" in certain circumstances, I am normally taken as implying that I am unhappy. But I am not taken to be asserting that I am unhappy, as I would be if I were to utter the words "I am unhappy." Or, to vary the example, when a person says, "All my children are now in college," he is normally taken to be implying that he has children (although not to be asserting that he has), and his auditors are justified in making this assumption. Or again, when one says in such sorts of contexts, "Smith has just gone out," he implies, or his words imply, that he believes or knows that Smith has gone out, and those to whom he is speaking are justified in assuming that he does. That the sense of "implication" expressed by these examples is not that of logical implication may be illustrated by the observation that there is no formal contradiction in asserting "All my children are in college, but I have no children" or in asserting "Smith has gone out, but I don't believe he has." Indeed, in standard systems of mathematical logic, the first statement is true whenever the speaker has no children, and the second is true whenever Smith has gone out but the speaker does not believe he has.

Sentences like "All my children are in college, but I have no children" and "Smith has gone out, but I don't believe he has" thus satisfy the rules of logical syntax and, indeed, the rules for correct English. Yet they fall upon the ear as decidedly odd. If employed at all in everyday speech, they would occur only in unusual circumstances—"I don't believe he has" might be whispered as an aside to a confederate, for example. But except for situations like this, they would be perplexing things to say. What, then, is the source of their oddity, given that they do not involve any formal mistake?

It is now generally agreed that the oddity we feel upon hearing such sentences stems from a disparity between the conditions we assume will have been satisfied whenever someone is trying to communicate with another and the utterances we expect will be employed in those circumstances. In effect, this is to say that certain assumptions, or presuppositions, that communicating human beings make in the everyday give-and-take of verbal intercourse, assumptions that thus form the ground of such intercourse, fail to hold or are violated in such circumstances.

Talk about presuppositions and talk about what is contextually implied by a speaker's words thus have in common a reference to the background conditions normally expected to obtain when an utterance is made. If stating and asserting are conceived of as elements constituting part of the foreground of the situation in which communication takes place (that is, as activities that bring an item of information into the immediate focus of attention), then presupposing and contextual implication may be thought of as elements constituting part of the background of the situation (that is, as factors that remain implicit unless they are otherwise articulated but that nonetheless are essential factors in communication). Part of the task that faces the student of informal logic is to specify what these conditions are, how they contribute to the background that makes communication possible, and what sorts of relations exist between them and the utterances that occupy the foreground during the transmission of information.

Let us then call the concepts referring to such conditions background concepts. Because such concepts play covert roles in daily discourse and because their functions are remarkably similar, it is not surprising that many writers have failed to discriminate between them. But not all writers have blurred the distinction. Isabel C. Hungerland is one notable exception. In her important paper "Contextual Implication" (Inquiry, Vol. 4, 1960, pp. 211–258), she writes, "The relation (presupposing) defined by Strawson is not that of contextual implication…. The relation between the two may be indicated as follows: When S presupposes S ′, a speaker in making the statement S, contextually implies that he believes that S ′" (p. 239). Following Mrs. Hungerland's suggestion and overlooking the many subtleties a full treatment of the subject would demand, we may say that the key distinctions that mark off the one notion from the other are those of scope: Neither the conditions subsumed under the two notions nor the range of entities to which the notions apply are in all cases the same.

Presupposing is a concept referring to those conditions that must be satisfied before an utterance can count as a statement, or if "statement" is so defined that statements need be neither true nor false (see P. F. Strawson, "Identifying Reference and Truth-Values," in Theoria 30 [2] [1964]), then presupposing applies to those conditions that must be satisfied before statements can be either true or false. Contextual implication, on the other hand, is a concept that applies to those conditions that must be satisfied before an utterance can count as "normal" in the circumstances in which it is made—that is, it applies to those beliefs a speaker has when he makes the utterance he does in certain circumstances and which rule out that he is lying or deliberately deceiving someone. The range of entities thus referred to by the concept of presupposing is either the class of statements as such or the class of those statements that are either true or false, whereas the range of entities referred to by the notion of contextual implication is the class of beliefs held by the speaker (and, derivatively, by his auditors).

Examples may be invoked at this point to illuminate the above remarks. Suppose during the course of a conversation I say, "The store on the corner sells such goods," not realizing that there is no longer a store on the corner. My remark in this circumstance is neither true nor false; as R. G. Collingwood puts it, the question of its truth or falsity "does not arise." For it is a presupposition of my using that utterance to make a statement (that is, an utterance that can be either true or false) that there be such a store. We may say in such a case that it is a condition of the truth or falsity of the remark that the store exist. But I may well believe that there is such a store, and in making the remark, I imply that I have this belief at the time of my utterance. One of the conditions for the normality of the remark (that is, that I was not lying) is that I had this belief at the time of saying what I did. We may say therefore that the conditions determining the normality of the background from which my remark issued and the conditions determining the background from which a statement would have issued are different conditions. It is this sort of difference in the background conditions that determines the difference between the concepts of presupposing and of contextual implication.

History of Contextual Implication

The genesis of the notions of contextual implication and of presupposing differs considerably. As a philosophical subject, contextual implication, under another name, has a longer traceable history in the modern period than does presupposing. The history of contextual implication is mainly connected with developments in moral philosophy, especially with efforts to give a correct analysis of the use of moral language. In G. E. Moore's Ethics (London, 1912), for example, we find the following comments:

There is an important distinction, which is not always observed, between what a man means by a given assertion and what he expresses by it. Whenever we make any assertion whatever (unless we do not mean what we say) we are always expressing one or other of two things—namely, either that we think the thing in question to be so, or that we know it to be so." (p. 125)

In the subsequent history of moral philosophy the distinction referred to by Moore became the key distinction invoked by those authors who espoused the emotive theory of ethics. According to advocates of this doctrine, the sorts of utterances used in moral contexts ("That's good," "Stealing is wrong") are not being used to make assertions and hence are neither true nor false, as both naturalists and nonnaturalists had assumed. The primary use of such utterances is to express the attitude or the feelings of the speaker toward whatever he is talking about and to arouse comparable attitudes in the auditor. The later history of contextual implication is deeply concerned with the import of this distinction, and the main works in which it is discussed, sometimes critically, are Language, Truth and Logic by A. J. Ayer (London, 1936); The Philosophy of G. E. Moore, edited by P. A. Schilpp (Evanston, IL, 1942), pp. 540–554; Ethics and Language by C. L. Stevenson (New Haven, CT, 1944); Ethics by P. H. Nowell-Smith (Harmondsworth, U.K., 1954); The Emotive Theory of Ethics by Avrum Stroll (Berkeley, CA, 1954); The Logic of Moral Discourse by Paul Edwards (Glencoe, IL, 1955); and "Contextual Implication" by Isabel Hungerland (see above). Various formulas are proposed by some of these writers.

Nowell-Smith says, for example, "A statement p contextually implies q if anyone who knew the normal conventions of the language would be entitled to infer q from p in the context in which they occur " (Ethics, p. 80). According to Hungerland all such early attempts to characterize the relation that obtains between what a speaker expressly asserts and what he implies suffer either from vagueness or from mistakenly thinking that the relation is a special case of inductive inference. Her own contention is that it is neither vague nor a case of inductive inference, but is, rather, the presumption that in a situation of communication, acts of stating are normal. She thus likens contextual implication to the juridical principle that a man is presumed to be innocent until proved guilty, a principle that is not arrived at inductively, by surveying the evidence, but which serves to place the onus of proof in a legal contest upon the prosecution. As she puts it, "Contextual inference (if we wish to use the word) is a matter, rather, of a communal assumption in the absence of evidence to the contrary, that, in a situation of communication, acts of stating are normal" (p. 233). Her view is that contextual implication depends upon three factors: (1) The presence of a stating context (since the question of a man's believing what he says does not arise in a nonstating context); (2) the presumptions of normality (that is, that within a stating context the implication holds only if the presumptions are principles of communication); and (3) rules for the correct use of an expression (that is, whether belief is implied when a man says p will be in part determined by rules for the correct use of p ).

History of Presupposing

Unlike contextual implication, the notion of presupposing has its genesis in logical theory, especially in those developments involving alternative accounts of Bertrand Russell's theory of descriptions and of the so-called square of opposition. The writer most closely identified with both of these matters is P. F. Strawson of Oxford University. He has dealt with the theory of descriptions in his papers "On Referring" (Mind, 1950), "Presupposing" (Philosophical Review, 1954) and "Identifying Reference and Truth-Values" (see above) and in his book Individuals (London, 1959; Ch. 8 especially). In Introduction to Logical Theory (London, 1952) Strawson considers both the theory of descriptions and the square of opposition.

In the works that deal only with the theory of descriptions, Strawson rejects Russell's analysis of sentences containing definite descriptive phrases (that is, phrases of the form "the so and so" used in the singular in English). According to Russell, the analysis of a sentence like "The queen of England is beautiful" contains in part an assertion to the effect that the queen of England exists. Strawson argues, cogently, that this statement is not an explicit part of what is asserted by "The queen of England is beautiful" but is presupposed by a speaker who would use such a sentence in normal circumstances to make a statement. In Introduction to Logical Theory, Strawson goes on to define the statement "S presupposes S ′" as follows: "The truth of S ′ is a necessary condition of the truth or falsity of the statement that S " (p. 175).

This characterization has been objected to by various writers, including David Rynin, who points out that when "necessary condition" and "truth or falsity of the statement that" are interpreted in the ordinary, truth-functional way, the definition has the paradoxical consequence that all presupposed statements are true. Rynin's demonstration is that (S ⊃ S ′) and (−S ⊃ S ′), but (S ∨ −S ); therefore S ′. Avrum Stroll has also suggested that Strawson's account suffers from the difficulty that if "The king of France no longer exists" is used to make a true statement, then by Strawson's criterion one who employs it thereby presupposes the existence of the king of France. It is now generally agreed that neither Russell's nor Strawson's analysis does full justice to all uses of sentences in everyday English containing "the" phrases in the singular. But regarded as proposals for the development of explanatory models for subparts of everyday discourse, each has considerable merit. In this interpretation Strawson's doctrine belongs to the logical tradition of analyzing descriptive phrases initiated by Gottlob Frege in "Über Sinn und Bedeutung" (1892) and supported by David Hilbert and Paul Bernays in their Grundlagen der Arithmetik (Berlin, 1934; Vol. I, p. 384) and by Rudolf Carnap in Meaning and Necessity (Chicago, 1947; pp. 33–42).

Strawson has also argued that if universal statements ("All my children are in college") are interpreted as presupposing the existence of the items mentioned by the subject term, paradoxes stemming from modern symbolic interpretations of the square of opposition can be eliminated without affecting the logical relations that one intuitively feels ought to hold between the elements of the square. This matter is persuasively discussed by S. Peterson in "All John's Children" (in Philosophical Quarterly, 1960).

Presupposing in Metaphysics

The notion of presupposition plays an important role in various metaphysical constructions, including Collingwood's An Essay on Metaphysics (Oxford, 1940) and Michael Polanyi's Personal Knowledge (Chicago, 1958). Collingwood distinguishes (Chs. 3–4) between absolute and relative presuppositions, arguing that the former are neither true nor false and that metaphysics is the science that ascertains what these absolute presuppositions are. His view is that absolute presuppositions form the basis of the civilizations developed at various times in history and the ground of the science developed in such civilizations. When a civilization changes, its presuppositions change and are succeeded by others. According to this view, metaphysics is therefore a branch of the historical sciences.

See also Ayer, Alfred Jules; Carnap, Rudolf; Collingwood, Robin George; Entailment, Presupposition, and Implicature; Frege, Gottlob; Hilbert, David; Moore, George Edward; Questions; Russell, Bertrand Arthur William; Strawson, Peter Frederick.

Bibliography

Anscombe, G. E. M. An Introduction to Wittgenstein's Tractatus. London: Hutchinson, 1959. Ch. II.

Asher, Nicholas, and Alex Lascarides. "The Semantics and Pragmatics of Presupposition." Journal of Semantics 15 (1998): 239–300.

Baker, A. J. "Presupposition and Types of Clause." Mind 65 (1956): 368–378.

Bar-Hillel, Yehoshua. "Analysis of 'Correct' Language." Mind 55 (1946): 328–340.

Black, Max. Models and Metaphors. Ithaca, NY: Cornell University Press, 1962. Ch. 4.

Black, Max. Problems of Analysis. Ithaca, NY: Cornell University Press, 1954. Chs. 2–3.

Cavell, Stanley. "Must We Mean What We Say?" Inquiry 1 (1958).

Collingwood, R. G. An Autobiography. London: Oxford University Press, 1939. See especially pp. 66–76.

Donagan, Alan. The Later Philosophy of R. G. Collingwood. Oxford: Clarendon Press, 1962.

Frege, Gottlob. "Über Sinn und Bedeutung." Zeitschrift für Philosophie und philosophische Kritik 100 (1892): 25–50. Translated by Max Black as "On Sense and Reference." In Translations from the Philosophical Writings of Gottlob Frege, edited by P. T. Geach and Max Black. New York: Philosophical Library, 1952.

Geach, P. T. "Russell on Meaning and Denoting." Analysis 19 (1959): 69–72.

Grant, C. K. "Pragmatic Implication." Philosophy 33 (1958): 303–324.

Grice, Paul. "Presupposition and Conversational Implicature." In Radical Pragmatics, edited by Peter Cole. New York: Academic Press, 1981.

Griffiths, A. P. "Presuppositions." Analysis 15 (1955).

Hall, Roland. "Assuming: One Set of Positing Words." Philosophical Review 67 (1958): 52–75.

Hall, Roland. "Presuming." Philosophical Quarterly 11 (1961): 10–21.

Hampshire, Stuart. "On Referring and Intending." Philosophical Review 65 (1956): 1–13.

Hancock, Roger. "Presuppositions." Philosophical Quarterly 10 (1960): 73–78.

Hart, H. L. A. "A Logician's Fairy Tale." Philosophical Review 60 (1951): 198–212.

Kemeny, John G. A Philosopher Looks at Science. Princeton, NJ: Van Nostrand, 1959. Ch. 3.

Langford, C. H. "The Notion of Analysis in Moore's Philosophy." In The Philosophy of G. E. Moore, edited by P. A. Schilpp. Evanston, IL: Northwestern University Press, 1942.

Lewis, David. "Scorekeeping in a Language Game." Journal of Philosophical Logic 8 (1979): 339–359.

Llewelyn, John E. "Collingwood's Doctrine of Absolute Presuppositions." Philosophical Quarterly 11 (1961): 49–60.

Llewelyn, John E. "Presuppositions, Assumptions, and Presumptions." Theoria 28 (1962): 158–172.

Lycan, William. "The Myth of the 'Projection Problem for Presupposition.'" Philosophical Topics 15 (1987): 169–175.

MacIver, A. M. "Some Questions about 'Know' and 'Think.'" Analysis 5 (1937–1938).

Nelson, Everett J. "Contradiction and the Presupposition of Existence." Mind 55 (1946): 319–327.

Nowell-Smith, P. H. "Contextual Implication and Ethical Theory." PAS, Supp. 36 (1962).

O'Connor, D. J. "Pragmatic Paradoxes." Mind 57 (1948): 358–359.

Quine, W. V. "Meaning and Inference." In From a Logical Point of View. Cambridge, MA: Harvard University Press, 1953. See especially pp. 164–166.

Rynin, David. "Donagan on Collingwood on Metaphysics." Review of Metaphysics 18 (1964).

Sellars, Wilfrid. "Presupposing." Philosophical Review 63 (1954): 197–215.

Shwayder, D. S. "Self-Defeating Pronouncements." Analysis 16 (1956).

Shwayder, D. S. "Uses of Language and Uses of Words." Theoria 26 (1960).

Soames, Scott. "Presupposition." In Handbook of Philosophical Logic, edited by D. Gabbay. Dordrecht: Kluwer, 1989.

Stalnaker, R. C. "Pragmatic Presupposition." In Pragmatics: A Reader, edited by S. Davies. New York: Oxford University Press, 1991.

Strawson, P. F. Introduction to Logical Theory. London: Methuen, 1952.

Strawson, P. F. "On Referring." Mind 59 (1950): 320–344. Reprinted in Logico-Linguistic Papers. London: Methuen, 1971.

Strawson, P. F. "A Reply to Mr. Sellars." Philosophical Review 63 (1954): 216–231.

Stroll, Avrum. "The Paradox of the First Person Singular Pronoun." Inquiry 6 (1963): 217–233.

Urmson, J. O. "Parenthetical Verbs." In Essays in Conceptual Analysis, edited by A. Flew. London: Macmillan, 1956.

Wolterstorff, Nicholas P. "Referring and Existing." Philosophical Quarterly 11 (1961): 335–349.

Avrum Stroll (1967)

Bibliographic assistance by J. Ornstein (1967)

Bibliography updated by Benjamin Fiedor (2005)