Lewis, Clarence Irving (1883–1964)

views updated

LEWIS, CLARENCE IRVING
(1883–1964)

Clarence Irving Lewis, the American epistemologist, logician, and moral philosopher, was born in Stoneham, Massachusetts, and educated at Harvard University (AB, 1906; PhD, 1910). He taught at the University of California from 1911 to 1920 and at Harvard from 1920 until his retirement in 1953; after 1930 he was the Edward Pierce professor of philosophy. He delivered the Carus Lectures in 1945 and the Woodbridge Lectures in 1954.

Lewis was a student and critic of modern extensional systems of logic and developed a modal logic based on the notion of strict implication. In epistemology and ethics, he was a pragmatic Kantian.

Lewis internalized within himself the great dialogue on knowledge and reality which began with René Descartes and continued with the British empiricists, Immanuel Kant and the German idealists, and the American pragmatists. It may be said that this tortuous development, both in its long history and in the intellectual life of Lewis, is the attempt of the modern mind to achieve consistency and adequacy in its conceptual foundations.

The basic commitments of any philosopher, whether formulated or not, concern the nature and modes of knowledge; they not only determine what is philosophically problematic for him but also determine how intelligibility can be achieved. Lewis modifies the classical certainty theory of knowledge, which maintains that knowing is an infallible state of mind. He contends that it does not make sense to talk about knowledge where there is no possibility of error. Knowing, according to him, is an assertive state of mind that is subject to appraisal as correct or incorrect by virtue of its relationship to what it is about, and also subject to appraisal as justified or unjustified in terms of its grounds or reasons. Thus the apprehension of a sensory given, or, in other words, the occurrence of an appearance, the classical paradigm of empirical knowledge, is not regarded by Lewis as knowledge, for there is no possibility of error. The apprehension of the appearance and its existence are indistinguishable.

Yet Lewis's departure from the tradition is not great. He, too, insists that at the foundation of our knowledge structure there must be certainty and that this is found in knowledge of sensory appearances. This certainty, however, does not reside in the apprehension of the given. Sensory appearances may be linguistically reported in "expressive" language, which denotes and signifies only appearances. Although there can be no error in the apprehension of a given appearance, it is possible to tell lies about it. Therefore, such reports are statements with truth-values. But still there is no knowledge, for no judgment is made in which the person could be in error. Knowledge is born at the level of what Lewis calls "terminating judgments," which are of the form "'S being given, if A then E,' where [in expressive language] 'A ' represents some mode of action taken to be possible, 'E ' some expected consequence in experience, and 'S ' the sensory cue." For example, there being a red patch in my visual field, if I seem to turn my head to the left, the red patch moves to the right. Such a judgment is not merely the apprehension of a given, or the linguistic expression of such. It embodies a prediction that the red patch will be displaced to the right if the specified condition is fulfilled, which Lewis contends is conclusively verified by the occurrence of the mentioned appearances.

Thus Lewis locates certainty in verified terminating judgments, which are about sensory appearances. Furthermore, he claims that all knowledge about the world is grounded in and derived from such certainties. Although this is more sophisticated than the traditional empiricist's account, it comes to much the same subjectivistic conclusion, namely, that the direct objects of knowledge are subjective and private, and therefore falls heir to all the problems of modern subjectivism. Lewis's major works are devoted to the central and toughest of these problems: how to make intelligible, from within these epistemological commitments, empirical knowledge of the objective world; a priori knowledge, including mathematics, logic, and philosophy itself; and value claims and normative judgments.

Empirical Knowledge of the Objective World

The paradigm of empirical knowledge for Lewis is the verified terminating judgment. It alone can be conclusively verified. All other empirical judgments are nonterminating. They may be shown to be probable but cannot be established with certainty. The probability value they have is conferred upon them by the verification of terminating judgments that they entail. Therefore, a necessary condition for a nonterminating judgment to be confirmable in any degree, and thus meaningful, is for it to entail terminating judgments.

Any statement that purports to be about objects other than appearances, such as physical objects, is nonterminating, and insofar as it is confirmable and therefore meaningful, it entails terminating judgments, which are about appearances only. It would seem that the full meaning of such a statement would be expressible in the terminating statements entailed by it and that, since these statements are about appearances only, the physical-object statement itself would really refer only to appearances. This would be phenomenalism.

Lewis resists this conclusion. He gives two arguments for realism. The first is that although a physical-object statement is intensionally equivalent to an inexhaustible set of terminating statements and the terms in the latter refer only to appearances, the terms in the physical-object statement genuinely denote physical objects. Thus we have two sets of statements, phenomenalistic and physical-object statements. For each physical-object statement there is a set (although inexhaustible) of phenomenalistic statements intensionally equivalent to it. By confirming the phenomenalistic set we confirm its equivalent physical-object statement with the same degree of probability. Yet the two are about radically different kinds of objects, and from knowledge of appearances we derive knowledge of physical objects.

This argument turns upon his theory of meaning. Lewis distinguishes four modes of the meaning of terms: (1) denotation, "the class of all actual things to which the term applies" (for example, the denotation of "man" is the class of all actual men, past, present, and future); (2) comprehension, "the classification of all possible or consistently thinkable things to which the term would be correctly applicable" (for example, the comprehension of "man" includes not only actual men but those who might have been but were not, like the present writer's sisters, since he has none); (3) signification, "that property in things the presence of which indicates that the term correctly applies, and the absence of which indicates that it does not apply" (for example, the property "rationality" is often regarded as included in the signification of "man"), and (4) intension, which consists of (a ) linguistic intension or connotation, all other terms which must be applicable to anything to which the given term is applicable (for example, "animal" must be applicable to anything to which "man" is applicable); and (b ) sense meaning, the criterion in mind, an imagined operation "by reference to which one is able to apply or refuse to apply the expression in question in the case of the presented, or imagined, things or situations" (for example, the sense meaning of "kilogon" is the imagined operation of counting the sides of a plane figure and the completion of the operation with the count of 1,000). Since he regards "propositions," statements with the assertive factor extracted (for example, "Mary's baking pies"), as terms, these modes of meaning apply to them as well. He further distinguishes between the "holophrastic" meaning of a statement, its meaning as a whole, and its "analytic" meaning, the meaning of its terms.

His argument is that although the holophrastic intensional meaning of a physical-object statement is the same as that of a set of phenomenalistic statements, the physical-object statement and its corresponding set of phenomenalistic statements are different in their analytic denotive meaning, the former denoting physical objects and the latter appearances.

Lewis rightly maintains that any two expressions that have the same intension have the same signification. Yet if a term denotes a physical object, it must signify a physical-object property. Therefore such a term could not have the same signification as a phenomenalistic term. Hence it seems that a physical-object statement could not have the same intension as a set of phenomenalistic statements.

Lewis senses this difficulty and seeks to avoid it by speaking of intension, in the form of sense meaning, as "that in mind which refers to signification." Appearances are said to signalize objective properties or states of affairs. Yet he gives no account of how this is possible for beings who can apprehend only appearances. How can appearances, as simple occurrences, be signs of anything other than other appearances? It would seem that the only way out of this subjectivistic trap is to regard appearances not as simple occurrences or objects of apprehension but as intentional in nature, as experiences of physical objects that embody truth claims about them which can be assessed as true or false on the basis of their consistency or lack of it with the claims of other experiences.

Lewis's second argument for realism turns upon the interpretation of "if … then …" in terminating judgments. He regards it as a contrary-to-fact conditional, that is, he claims that the truth of the conditional as a whole is independent of the truth-value of the antecedent and therefore may be significantly asserted when the antecedent is known to be false. Therefore, since it does not express a logical relation of entailment or a truth-functional relationship, it must express a real connection, perhaps causality, that holds between the facts or states of affairs located or referred to by the antecedent and the consequent of the conditional sentence. Belief in a real world, he maintains, is belief in such contrary-to-fact conditionals.

It is not clear how this is an argument for realism. Why must independent physical objects be assumed to account for the contrary-to-fact character of terminating judgments? Why couldn't the "real" connection hold between kinds of appearances?

Furthermore, if terminating judgments are to be interpreted in the manner of the contrary-to-fact conditional, does this not compromise their conclusive verifiability? It would seem to introduce an element of generality that would transcend any specific sequence of subjective experiences. In fact, Lewis himself, for other reasons, held that no terminating judgment of the form "S being given, if A then E " is strictly entailed by a physical-object statement. The most we can say, he concluded, is "If P [physical-object statement], then when presentation S is given and act A is performed, it is more or less highly probable that E will be observed to follow." Since the statement is inconclusive, it seems that he has given up the terminating character of "terminating" judgments.

Lewis has not, it seems, made a convincing case for realism from within his phenomenalistic foundations. Some have concluded that it is impossible to do so and that the only way out of phenomenalism is to abandon the subjectivistic starting point itself.

A Priori Knowledge

The a priori disciplines, namely, mathematics, logic, and philosophy, were the stronghold of classical rationalism. They were regarded as yielding knowledge, grounded in rational intuition, about the essential and necessary structure of the world. Empiricists, for the most part, claim that such knowledge is only intralinguistic, that it consists of analytic truths, which are said to be uninformative about the world.

Lewis subscribes to the view that all "a priori truth is definitive in nature and rises exclusively from the analysis of concepts." Unlike many empiricists, however, he is not content with merely characterizing a priori knowledge as analytic. For him, concepts, their logical relations, and their relation to the data of sense and the structure of the world are highly problematic. He regards concepts, logical relations, and a priori truths arising from them as the peculiar characteristics of mind. He sets them in contrast with the given data of sense experience, which he regards as brute fact, unlimited and unaffected by the conceptual structure. But these givens would be unintelligible without the a priori criteria of classification provided by mind, criteria which are involved not only in talk about things but even in the experience of objects. Thus, the necessary connections of concepts are embedded in perception, and analytic truths, far from being trivial and only intralinguistic, formulate the a priori structure of the world as experienced and known.

Our basic conceptual structure, and thus our a priori truths, are not fixed and eternal. They consist of deep-seated attitudes grounded in decisions that are somewhat like fiats in certain respects and like deliberate choice in others. There is nothing in our conceptual structure that is not subject to change in the face of continuing experience. This includes such basic decisions as the decision that whatever is to count as real, in contrast with the hallucinatory, must stand in causal relations with other real things. Even the laws of logic, "the parliamentary rules of intelligent thought and action," are subject to change. The only test applicable is pragmatic, the achievement of intelligible order with simplicity, economy, and comprehensiveness in a way that will be conducive to the long-run satisfaction of human needs. Thus, Lewis holds to a pragmatic theory of a priori truth but not of empirical truth.

Philosophy, according to Lewis, is a reflective, critical study of mind and its a priori principles as found in "the thick experience of everyday life," and thus in "the structure of the real world which we know." Although it studies what is implicit in experience, it is analytic and critical in method rather than descriptive. Its function is not only to formulate the conceptual structure built into experience and thought but to sharpen and to correct it. Thus philosophical claims may be analytic in character, like "There is an intelligible order in the objective world." Lewis takes this statement to be analytic on the ground that an intelligible order is an essential mark of the objective world. Whatever lacks a certain minimum order is only subjective, private experience, like dreams and hallucinations. Philosophical claims also may be critical and revisionary, recommending some change in our categorial attitudes, such as "Only the physical is real."

Lewis's theory of the a priori places the conceptual framework between two sets of givens, the presentations of sense, to which concepts apply to yield empirical knowledge, and the values in terms of which the a priori structure is pragmatically tested. It seems that both sensory experiences and values would have to be free of a priori assumptions in order to serve the function ascribed to them. This is a difficult doctrine to maintain.

Value Claims and Normative Judgments

The ultimate test of the a priori conceptual framework, according to Lewis, is "the long-run satisfaction of our needs in general." It would seem that value judgments would have to be independent of the conceptual framework that is being pragmatically tested if the test is to be clear-cut and not beg the question. But obviously this would be impossible in the case of basic issues. Although Lewis does not face the problem in these terms, he may be said to blunt the criticism by locating values among sense presentations and by invoking unavoidable imperatives that would be operative in any conceptual framework.

Value, in its most primitive sense, has to do with sense presentations. It is not so much a specific phenomenal quality as a mode or aspect of the given, namely, the given as gratifying or grievous. The only thing that is intrinsically good is liked or wanted subjective experience. In addition to the immediately found intrinsic value of an experience, it may be said to have contributory value by virtue of the contribution it makes to the total value quality of the conscious life of which it is a part. Such a life, he contends, is not simply a sum of its parts. So the contributory value of an experience is quite different from its intrinsic value. Objects of experience are said to be extrinsically good or bad according to their capacity to produce experiences which are satisfying or unpleasant.

Thus, for Lewis, value knowledge is a form of empirical knowledge. There are both terminating and nonterminating value judgments. The former are subjective statements of intrinsic and contributory value; the latter are objective statements about extrinsic values. Judgments of right and wrong, however, are not empirical in character. They are determinable only by reference to rules or principles that refer to values in their prescriptions. He regards the basic rational imperative to be so to think and so to act that later you will not be sorry. The only way this can be achieved is for decisions to be guided by objective knowledge rather than merely by the affective quality of immediate experience. In the area of morals, this requires that we respect others as the realities we know them to be, "as creatures whose gratifications and griefs have the same poignant factuality as our own; and as creatures who, like ourselves, find it imperative to govern themselves in light of the cognitive apprehensions vouchsafed to them by decisions which they themselves reach, and by reference to values discoverable to them."

Any attempt to prove the validity of such principles can only appeal to an antecedent recognition of them. They must be recognized by all who make decisions, all who think and act. Genuine skepticism with regard to judgments of right and wrong, good and bad, would be impossible, for on such a basis even doubt itself would be meaningless.

The question remains: Is the conceptual framework in which normative and value knowledge is formulated pragmatically testable, and, if so, just what could such a pragmatic test amount to? If it is not so testable, then it would seem that in the end Lewis is not a pragmatist after all.