Murray Murphey's The Development of Peirce's Philosophy, a seminal work, energized Peirce scholarship. Now, approximately 50 years later, he has produced yet another work, C. I. Lewis: The Last Great Pragmatist that promises to do for Lewis what his earlier work did for Charles Peirce.

The two philosophers, Lewis and Peirce, have been linked ever since Lewis returned to Harvard in 1920 in part to edit the Peirce manuscripts that had a few years before come into Harvard's possession. He spent two years on that project before quitting it to devote full time to his own career. Intellectually, the two philosophers had much in common. Later in life, Lewis said that Peirce was the philosopher who influenced him most. My intention, however, is to concentrate on what divides them with a view toward raising questions for further clarification and exploration.

The received view of Lewis' epistemology is that it is foundationalist, that is, Cartesian. The metaphor of knowledge employed is that of an edifice supported by a foundation of basic beliefs, where "basic" means a belief some measure of whose justifiability is non-inferential, that is, a belief at least part of whose justification is not derivable by way of non-deductive inferential relations. By "Cartesianism" I refer to a "justification" theory and more directly to a tradition whose adherents in addition to acknowledging epistemologically basic (self-justified) beliefs, also affirm the sensory character of those beliefs as well as their non-transcendence of immediate experience. Not surprisingly, I'm concerned less with specific aspects of the Cartesian rationalistic program than with the empiricistic emendation of it.

That Lewis fits into the Cartesian landscape seems quite clear. He meets the sensory requirement by accepting the immediate data [End Page 45] of sense--called "the given" or hard data or other locutions--as primary. He does deviate from the immediacy requirement to the extent that his theory of empirical knowledge requires inductive inference from past experiences. Only the memory of past experience is actually given. For all empirical beliefs, their credibility depends upon prima facie credibility of what is remembered and upon mutual support or congruence.1 There is the appearance of paradox: memory generally is trustworthy but not all memories are. Lewis' Cartesianism runs headlong into Peirce's anti-Cartesianism.

Descartes adopted a procedure by which he fully expected to uncover those epistemologically basic propositions upon which philosophy could be reconstructed. To gain clarity, Descartes said, we must break up our complex ideas into their simple parts in respect to which there is no discernable lack of clarity. Forced to distinguish between what is clear and what merely seems clear, he fell back upon the intuitions of a scrutinizing mind. Today, many philosophers reject this classical theme of knowing as the unclouded contemplation of the object, questioning whether sheer contemplation has anything to do with cognition. This transformation of outlook is, in part, due to Peirce's efforts. Even Hume, who played such an important role in the construction of empirical knowledge, permitted knowledge of the relations of ideas by intellectual inspection. Peirce questioned whether any domain was open to intuition, denying that there are strictly necessary truths or analytical propositions as traditionally understood. Appealing to the principle of non-contradiction, he held, won't do, for what is demanded is a criterion for its application.

Arguing against a sensationalistic theory of belief, Peirce seized yet another opportunity to support the public character of belief. He doesn't deny that there is a feeling of belief and hence admits, along with others, that there are sensational beliefs. What he denies is that the sensational, or feeling, element is exclusive. Noting that we can have beliefs which are either sensational or not, Peirce makes the distinction between sensational and active beliefs, whereby an active belief he means nothing other than an incipient judgment upon which one is prepared to act. On this view, one need not be conscious of beliefs in order to have them. That is, beliefs are not appearances that move into or out of consciousness. Rather, they are the manifestation of habits, which are propensities or dispositions to act. Only in a Pickwickian sense does Peirce say that beliefs are habits.

For the sensationalist, the privileged position of the individual or self is maintained. For him it makes sense to say, for example, "Apart from appearances, I really do believe this" or "How can you know what I believe?" or simply, "I know what I believe." But for Peirce, beliefs, as forms of behavior, belong to the public domain. With no intention of denying the inner life, Peirce does deny that the individual is in a privileged position with respect to it.

In short, Peirce sought to replace the concept of self-evident truth, privately ascertainable, with that of hypothesis, publicly testable. Human [End Page 46] knowledge, for him, consisted of a network of beliefs such that for the most part each belief supports every other belief, with the whole corresponding to an objective reality. There are no first beliefs, unique starting points for investigations that are irrefutably such.

Peirce confronts Descartes specifically in a number of places in his writings, but perhaps most crucially in two early essays entitled "Questions Concerning Certain Faculties claimed for Man" and its sequel "Some Consequences of Four Incapacities," both of which first appeared in the Journal of Speculative Philosophy (1868). I shall concentrate immediately upon the first of these essays, and later upon portions of the second. To fully appreciate how devastating Peirce's attack on Cartesian intuitionism is in the first of the two essays requires a full exposition of that essay, and in precisely the order in which Peirce takes up his "Questions," for his replies to them are intended to remove progressively a buttress of the intuitionist position. But this is not possible here. So, let his response to the last question suffice.

The last argument for intuitionism is as follows: "It would seem that there is, or has been, [a cognition not determined by a previous cognition]; for since we are in possession of cognitions, which are all determined by previous ones, and these by cognitions earlier still, there must have been a first in this series" (5.259).2 In short, if there isn't a first cognition in the series of cognitions, then nothing could be known, but in point of fact, something is known.

Peirce resorts to an analogy in his reply. Permit an inverted triangle to be gradually lowered into water, so that the surface of the water will make a horizontal line across the triangle. By lowering the triangle further, another horizontal line will be made across the triangle, with the second line standing higher across the triangle than the first. Let these horizontal lines represent different cognitions, so that the horizontal line below the other horizontal line represents one cognition determining another. In the analogy, the apex of the triangle will represent the object external to consciousness which determines both these cognitions. Now, for someone to say that there must be an absolutely first cognition, one would have to be saying, according to the analogy, that there must be an absolutely first line below which no other line could be made. But then, one would be mistaken, because for any line across the triangle as many lines as one pleases can be assigned below the given line. It is simply not the case that there must be a first line. It is equally not the case that there must be a first cognition.

The logical difficulties of this paradox are, Peirce observes, identical with Zeno's Achilles paradox. Whatever can be said of the Achilles paradox will apply as well to the case of cognitions determining other cognitions. So that if one denies motion, then one must deny the process of one cognition determining another. Say that points and lines are fictions, then say the same of cognitions and judgments. Peirce's point is that the transcendental argument he has been considering has no special merit. It no more proves that cognition is impossible than the Achilles paradox proves that there can be no motion. [End Page 47]

Peirce's fascination with the continuum was to continue. It would lead him beyond his early nominalistic views into various speculative ventures and to a doctrine called "Synechism," the principle thesis of which is that everything existent is continuous. Synechism and fallibilism are joined by the principle of continuity, since the fallibilistic thesis affirms that inquiry is ongoing or continuous. For Peirce "the principle of continuity is the idea of fallibilism objectified" (1.171­2).

Peirce's questioning the possibility of getting behind percepts to the sense impressions which presumably constitute the percepts led him to conclude that these impressions are "hypothetical creations of nominalistic metaphysics" (6.492). Moreover, even if the existence of such impressions is assumed, it is not in these impressions that experience consists, for "by experience must be understood the entire mental product" (ibid.). The new position would have the percepts "constitute experience proper" (2.142).

Once the meaning of "percept" is established as "the entire mental product" or "the total content of immediately present awareness," then the meaning of "perceptual judgment" and "perceptual fact" can be fixed, in part, as "the intellect's description or characterization of the percept." Although both perceptual judgments and perceptual facts have interpretive functions, they are distinguishable by the fact that perceptual judgments are so swiftly formed as seemingly to be uncontrollable, whereas perceptual facts, or descriptions of percepts, require voluntary effort of some kind (cf. 2.141). Perceptual facts are not forced on our attention in precisely the way perceptual judgments are.

It is the problem of describing percepts that lies behind Peirce's introduction of the new conception of perceptual fact. In his words:

Hundreds of percepts have succeeded one another while I have been setting down these sentences. I recognize that there is a percept or flow of percepts very different from anything I can describe or think. What precisely that is I cannot even tell myself. It would be gone long before I could tell myself many items; and those terms would be quite unlike the percepts themselves. In this thought, there would always be effort or endeavor. Whatever is the product of effort might be suppressed by effort, and therefore is subject to possible error. I am forced to content myself not with the fleeting percepts, but with the crude and possibly erroneous thoughts, or self-informations, of what the percepts were.


What we must content ourselves with, then, is what is remembered; namely, the intellect's description of the evidence of the senses called "perceptual facts" (2.141). The testimony of a witness to an occurrence will consist of recollections which constitute perceptual facts, because (a) the "facts" presented are totally unlike the event witnessed and (b) the "facts" themselves were not literally forced upon the witness, but are the product of reflection and some effort on his part. [End Page 48]

In contrast, "a judgment asserting in propositional form what a character of a percept directly present to the mind is" is called a "perceptual judgment" (5.54; italics mine). The perceptual judgment is the first judgment a person makes as to what is before his senses (5.115). Unlike the perceptual fact, the perceptual judgment seems quite beyond control and not subject to criticism. But, as in the case of a perceptual fact, it "bears little resemblance to the percept" (ibid.). The percept, after all, is "an image or moving picture or other exhibition" (ibid.); a judgment is an act of forming a proposition, together with an act of assenting to it.

For example, you look at something and say, "it is red." Well, I ask you what justification you have for such a judgment. You reply, "I saw it was red." Not at all. You saw nothing in the least like that. You saw an image. There was no subject or predicate in it. It was just one unseparated image, not resembling a proposition in the smallest particular.


The upshot of the discussion thus far has not been merely to deny intuitive knowledge but also to affirm the thesis that propositional claims, all knowledge in effect, are, in principle, fallible. This thesis, for which his arguments against intuitionism may be regarded as preparatory, Peirce labeled "fallibilism."

All perceptual judgments and all perceptual facts are fallible, but the perceptual judgments, as distinguished from the perceptual facts, are indubitable. Why? The perceptual judgments, unlike the perceptual facts, are forced upon us. They transcend all possibility of criticism:

If I judge a perceptual image to be red, I can conceive of another man's not having the same percept. I can also conceive of his having this percept but never having thought whether it was red or not. I can conceive that while colors are among his sensations, he shall never have had his attention directed to them. Or I can conceive that, instead of redness, a somewhat different conception should arise in his mind; that he should, for example, judge that this percept has a warmth of color. I can imagine that the redness of my percept is excessively faint and dim, so that one can hardly make sure whether it is red or not. But that any man should have a percept similar to mine and should ask himself the question whether this percept be red, which would imply that he had already judged some percept to be red, and that he should, upon careful attention to this percept, pronounce it to be decidedly and clearly not red, when I judge it to be prominently red, that I cannot comprehend at all. An adductive suggestion, however, is something whose truth can be questioned or even denied.


As to the indubitable inferences, they, too, are indubitable in the sense in which they are acritical (cf. 5.440). Now Peirce would refuse to apply the term "reasoning" to the determination of one belief by a second belief along with the consciousness that the first belief is an effect of the second belief, but where there is no understanding how or why it is so. It is required that [End Page 49] "... in reasoning we should be conscious, not only of the conclusion, and of our deliberate approval of it, but also of its being the result of the premise from which it does result, and furthermore that the inference is one of a possible class of inferences which conform to one guiding principle" (5.441). The class of mental operations which possesses these properties is called reasonings. But a belief may be consciously determined by another belief without an awareness that the inference is governed by a guiding or leading principle. "Such a process should be called, not a reasoning, but an acriticalinference" (5.441), and the example which Peirce gives of this kind of inference is the well-known "cogitoergosum."

If fallibilism is Peirce's answer to the dogmatist, then his defense of the indubitability of some propositions might count as the answer to the outright skeptic. For the sake of convenience, allow me to label Peirce's collective views on indubitability "credibilism" and also to set aside, for the moment, the question of the consistency of the two "isms."

Peirce's defense of credibilism involves an attack upon Descartes, but now from another side. Descartes had proposed a method whereby the inquirer doubts everything inclination would have him believe in the expectation of reaching beliefs which survive the experiment and which, in consequence of their survival, serve as instances of self-sufficient knowledge of the truth. Peirce's objection to this program emanated from the fact that the doubts the inquirer invoked are not genuine, that a mere act of will, without positive reason to disbelieve, cannot create doubts which are. "We must begin" Peirce said in his essay "Some Consequences of Four Incapacities," "with all the prejudices which we actually have.... Let us not pretend to doubt ... what we do not doubt in our hearts" (5.225). Required is, as Peirce put it: ". . . an external origin, usually from surprise" (5.443).

Common sense indubitables serve as ultimate premises of a sort. They cannot be ultimate premises strictly unless, of course, Peirce either forgot or deliberately chose to repudiate the views expressed in the first of the Journal of Speculative Philosophy articles of 1868 in which it was maintained that all knowledge is inferential and that the series of inferences produced was in fact infinite, since there could be no cognition not determined by a previous cognition. So if common sense indubitables stand first or at the foundations of inquiry, then clearly it cannot be because they themselves are not the product of an inferential process, but because they are not the product of conscious inference, which is precisely the reason why they are not subject to criticism. When Peirce writes about "prejudices," he is writing about prejudices that we actually have. The mass of cognitions to which he refers are those already formed. From a psychological standpoint, these prejudices and that mass are inferable, presumably from an infinite series of subconscious inferences. But let us not confuse the epistemology of justification with the psychology of knowing but understand that claims concerning an infinite regress of justification do not prevent one from accepting as logically foundational special beliefs and the propositions or feelings these beliefs are about. [End Page 50]

The two positions of credibilism and fallibilism are complementary. Once it is understood that indubitability stands for "freedom from genuine doubt," there is no inconsistency in holding that all propositions which are presently indubitable are, in principle, fallible. As Peirce said: the philosopher "fully acknowledges that even upon completion of any given review of his own beliefs it may be that some of his indubitable beliefs may be proved false" (5.451).

The juxtaposition of credibilism and fallibilism was, indeed, formally accomplished by Peirce in 1905 in the second of three articles for the Monist. Concerned there primarily with the investigation of the unquestioned beliefs which underlie inquiry and the grounds upon which they are subject to criticism, he proposed a doctrine which he called "Critical Common-sensism" whose purpose was to provide a synthesis of common sense and criticism, the culmination of the effort to find middle ground between foundationalism and coherentism, opposite ends of a philosophical spectrum. It appears that there is an unbridgeable gap between Lewis and Peirce. But more about this later.

The story doesn't end with Peirce. Roderick Firth, Lewis' successor at Harvard, sought and failed to produce what he hoped would be the definitive defense of Cartesian epistemology. Coincidentally, his Harvard colleague, Willard Van Quine, was moving toward a position currently referred to as holistic pragmatism. The background was Quine's attack on the analytic/synthetic dualism. Quine's target, among several, was Lewis. Noteworthy is his contrasting appraisal of Peirce's empiricism, namely that it followed a course of development parallel to his own, including shifts of focus from ideas to words via beliefs, from terms to sentences, and from sentences to networks of sentences, each shift marking an advance. Also acknowledged approvingly is the holistic character of Peirce's empiricism. 3 Quine drew the line when it came to first philosophy, to which he believed Peirce was committed.

Morton White takes holistic pragmatism a step further by adding normative ethical principles to the mix that also includes descriptions, the whole tested empirically. But Peirce is already there because for him logic, broadly conceived to include scientific methodology, is normative. It is classified with ethics and aesthetics and is dependent upon both. But White's ontology of actualities runs counter to Peirce's more elaborate metaphysics that Peirce labels "scientific."

Earlier Peirce made the resolution of doubt by belief the motivating force. Later he asked whether this was motivation enough. The last version of his pragmatic maxim makes explicit reference to a loftier ideal that doesn't consist in acts per se but "in that process of evolution whereby the existent comes more and more to embody those generals which ... [are] said to be destined, which is what we strive to express in calling them reasonable" (5.4). Individual acts are valued in the way they further this development. The reformulation avoids his early psychologizing, substitutes [End Page 51] realistic convictions for nominalistic ones, and articulates a higher ideal for human beings than personal satisfaction.

At the same time Peirce provides motivation as well as direction for exploring anew the question of being. That exploration yielded his three universal and irreducible but distinctly hypothetical categories, each with its own monadic, dyadic or triadic character and with its own brand of realism. The important point is that the categories are genuinely hypothetical once their empirical credentials are established in virtue of their relevancy for every item of experience. Indeed, they may be the most empirical of hypotheses. Metaphysics could be and, in fact, should be an observational science, differing from the special sciences only by the breadth of its observational base.

On the issues raised concerning normativity and metaphysics, where does Lewis stand? It is known that he believed that logic was normative. Indeed all knowledge for him was normative. Moreover, he held that pragmatism was tied up with the realism question. Frequently he declared that there were natural or real connections among facts. Beyond this the picture is blurred. One might have guessed that his modal logic would lead naturally to a modal realism, possibly along the lines of Peirce or of other modal logicians more contemporary by way of anticipation. Instead there appears to be a general reluctance to engage in metaphysical speculation of any kind. He once remarked that his attempts to do metaphysics invariably turned out to be epistemology. This reluctance is borne out in his response to Victor Lowe, a Whitehead disciple, who chides Lewis for his failure to take metaphysics seriously. Lewis wrote:

My sense of the ultimate I do not put into words. There is a favorite haunt, in an almost uninhabited wilderness, to which I go as often as I can. I hope there will always be such wilderness spots remaining and I hope there will always be, included in philosophy, a kind of literature which I could not by any possibility write.4

What do we make of this? It is open to several kinds of interpretation. In his essay "The Categories of Natural Knowledge," Lewis reveals more straightforwardly that he once had a lively interest in the middle period works of Whitehead when Whitehead was preoccupied with the philosophy of the natural sciences, propelled by what Peirce thought of as scientific metaphysics.5 Lewis, a non-scientist, unlike Peirce and Whitehead, was limited in his efforts to deal in depth with the subject matter of those works. One senses, however, that dealing in depth with the scientific material is what he might have wanted to do had he the proper scientific training. We now know that the metaphysical views of Peirce and Whitehead are similar in many ways. It is intriguing to think that Lewis found something in Whitehead that he wasn't able to find in Peirce.

Let me conclude on another speculative note by drawing attention to what Susan Haack, a distinguished philosopher in her own right, has to say [End Page 52] about Lewis, a philosopher for whom she has enormous respect. She calls Lewis a sometime "proto-foundherentist," an epistemological position that combines in some fashion foundationalism and coherentism, two positions normally taken to be in opposition to each other. What she has in mind is not simply an alternative position situated half-way between the other two, but a radical revision, a reconstruction, of epistemology itself. As she makes clear, she has Peirce's critical common-sensism in mind.6

But how plausible is Haack's claim about Lewis and proto-foundherentism, an admittedly unattractive name for an attractive stance? The problem is some of Lewis' signature themes, his entrenched apriorism, for example, upon which I dwelt. The answer is for the Cartesian Lewis to be put aside in favor of the Lewis who is more openly pragmatic and who thinks of himself as a contrite fallibilist in the manner of Peirce. In truth there is much the two have in common, not the least of which is heeding the clarion call to seek the truth, the hypothetically ideal upshot of a disinterested but passionate pursuit, backed up by a robust realism and by a common-sensical conviction of the orderliness of the universe. Surely there is enough here to give the claim the benefit of the doubt and to license further exploration of the question.

Richard S. Robin
Mount Holyoke College


1. Congruence functions for Lewis as concordance does for Peirce, except that Peirce operates on a larger scale. The aim for both is to bolster credence beyond what formal logical consistency achieves.

2. All Peirce references are to The Collected Papers of Charles Sanders Peirce, eds., C. Hartshorne and P. Weiss, Volumes I—VI (Cambridge, Mass.: Harvard University Press, 1931–1935).

3. W. V. Quine, "The Pragmatists' Place in Empiricism," in Pragmatism: Its Sources and Prospects, eds. R. Mulvaney and P. Zeltner (Columbia, S.C.: University of South Carolina Press, 1981), 21–39.

4. C. I. Lewis, The Philosophy of C. I. Lewis, ed. P. A. Schilpp (LaSalle, Illinois: Open Court, 1968), 655. For Victor Lowe's criticism, see "Lewis' Conception of Philosophy" in The Philosophy of C.I. Lewis, 23–59.

5. C. I. Lewis, "The Categories of Natural Knowledge," Collected Papers of Clarence Irving Lewis, eds. J. Goheen and J. Mothershead, Jr., (Stanford, CA: Stanford University Press, 1970), 113–147.

6. Susan Haack, "Reflections of a Critical Common-sensist," Transactions of the Charles S. Peirce Society, Spring 1996, 359–373. [End Page 53]

Additional Information

Print ISSN
Launched on MUSE
Open Access
Back To Top

This website uses cookies to ensure you get the best experience on our website. Without cookies your experience may not be seamless.