In April 1887 Peirce moved with his second wife, Juliette, from New York City to Milford, Pennsylvania, a small resort town in the upper Poconos. A year and a half later the Peirces moved into a farmhouse two miles northeast of Milford in the direction of Port Jervis, New York. This was to become Peirce's Arisbe, named for a Greek town south of the Hellespont, a colony of Miletus, home of the first philosophers of Greece. (1) The renovation and expansion of the Arisbe house would often preoccupy Peirce during his remaining years. The architectural work of remodeling Arisbe, always with an eye for something vast, would become a living metaphor for his intellectual life.(2)

Starting in the mid-80s with his "Guess at the Riddle," Peirce began to gather his philosophical doctrines together into an integrated system of thought, and with his 1891 Monist article, "Architecture of Theories," he began to attend explicitly to the structural integrity of his system as a whole. One of Peirce's main efforts after 1890 was to reestablish pragmatism, not attended to since his 1877-78 "Illustrations," as an integral component of his systematic philosophy. The integrating structure for his mature philosophy would be a much expanded, though never fully completed, theory of signs. Also prominent in Peirce's later writings is a more dominating form of naturalism that ties the development of human reason unambiguously to natural evolution and that takes on clear religious overtones.

The introduction printed in volume 1 (EP1) is the general introduction for The Essential Peirce as a whole, but no attempt was made to represent Peirce's intellectual development during his last two decades. This special introduction to volume 2 (EP2) is intended to supplement the general introduction by providing a sketch of this period. Peirce's life continues to resist easy characterization—unless cryptically in the claim that he embodied the general maxim he extolled in his fourth Harvard Lecture (sel. 13): "Never say die." There is no doubt that his life was one of much suffering and many defeats, but he never for long lost sight of his purpose: to do what he could to advance human understanding. He knew his own powers, and he knew the mundane truth that knowledge is advanced through scholarly preparedness, insight, humility, and hard intellectual work; and it was no delusion of grandeur for him to realize that he was poised to make contributions no one else could make. The story of Peirce's struggle to redeem his talents is one of the great personal tragedies of our time, but it cannot be told here.(3) These remarks are intended only to provide a unifying structure for the writings in this collection and a vantage point for surveying the grand expanse of a remarkably rich and complicated mind.

One obstacle to a comprehensive understanding of Peirce's thought is the broad range of his intellectual achievements, covering so many of the human and physical sciences; but added to that is the difficulty of determining to what extent he was influenced by his predecessors and peers. Of course, no one can think in a vacuum—thought must necessarily relate to past thought, just as it must appeal to subsequent thought—so it is never cogent to ask about any thinker whether his or her thought was influenced by previous thinkers, but only how and to what extent. To Peirce, this was obvious. Given his upbringing among mathematicians and experimental scientists he learned early that intellectual progress is always relative to knowledge already gained and that any successful science must be a cooperative endeavor. One of the reasons Peirce is so important for the history of ideas is that he approached philosophy in this way, knowing that if philosophy was ever really to amount to anything it would have to abandon the notion that great ideas arise ex nihilo—that one's ideas are wholly one's own. As a result of this understanding, and of his desire to help move philosophy toward a more mature stage of development, Peirce became a diligent student of the history of ideas and sought to connect his thought with the intellectual currents of the past. He also studied carefully the leading ideas of his own time. His debts are extensive—far too numerous to be cataloged fully here—but it could not be too far wrong to say that Aristotle and Kant were his most influential predecessors, with Plato, Scotus, and perhaps Berkeley coming next, although only on the heels of many others such as Leibniz, Hegel, and Comte. With respect to Peirce's scientific, mathematical, or logical ideas, others have to be added, including, certainly, De Morgan and Boole. When one considers how Peirce's thought was influenced by the ideas of his contemporaries one is hard-pressed to settle on a short list. Peirce was very current in many fields of study, due both to his scientifically informed approach and to the fact that he wrote hundreds of book reviews and newspaper reports on scientific meetings and "picked up" ideas along the way. In logic and mathematics, and even in philosophy, aside from predecessors, the influence of Cayley, Sylvester, Schröder, Kempe, Klein, and especially Cantor stands out. Peirce was also responsive to the writings of his fellow-pragmatists, among whom he included Josiah Royce; but he was more influenced by William James than by any other contemporary. Other contemporaries of note were the philosopher and editor, Paul Carus, and the English semiotician, Victoria Lady Welby, whose work on signs ("significs") led her to Peirce, and whose attentive interest in his semiotic ideas encouraged him to develop his theory of signs more fully than he would have without her.

Paul Carus (1852-1919) is a special case. Carus, a student of Hermann Grassmann, has been surprisingly neglected by historians, given his remarkable output as a philosopher and his importance as an editor and critic. He wrote scores of books and hundreds of articles (not only on philosophy) and edited over one hundred issues of the Monist and over seven hundred issues of the Open Court, the two periodical publications of the Open Court Publishing Company.(4) Open Court authors included the classic American quartet, Peirce, James, Royce, and Dewey, and a host of others ranging from Ernst Mach and Bertrand Russell to D. T. Suzuki. Carus was a confirmed monist, as is revealed in the name of his journal, and devoted to the reconciliation of science and religion. He took a special interest in Peirce and for over twenty years, notwithstanding some periods of acrimony, he did more to promote Peirce's philosophy than anyone. Beginning in 1891, Carus published nineteen of Peirce's articles (thirteen in The Monist and six in The Open Court) and many of Peirce's unpublished writings were intended for Carus. The important role played by Carus in Peirce's later life, in particular the fact that after 1890 Peirce wrote most of his best work for the Monist, is what led Max Fisch to call that time Peirce's Monist period.

The writings in the present volume begin in 1893 when Peirce was fifty-four years old, only three years into the Monist period and one year after his forced resignation from the Coast and Geodetic Survey. He had recently delivered a course of lectures on "The History of Science" at the Lowell Institute in Cambridge and was just bringing to a close—one article prematurely—his influential metaphysical series for the Monist (EP1, sels. 21-25). He was at work on "Search for a Method," which was to include a substantially revised version of his 1877-78 "Illustrations of the Logic of Science" (EP1, sels. 7-12), and was about to announce a twelve-volume opus, The Principles of Philosophy, possibly inspired by James's recent success with his Principles of Psychology. Clearly, the opening writings of the present volume arose in the context of an active and ongoing program of research.

For an intellectual profile of EP2, the separate headnotes to the selections might be read consecutively. Although they were not composed to provide a continuous flow of text, they do give an idea of a thread of intellectual development that ties together the writings in this volume. Obviously it is not possible to capture rich full texts, as most of Peirce's are, in short notes, but sometimes a single strand of connected meaning is all that is needed to precipitate more substantial linkings. Building on the headnotes, bearing in mind some of the biographical structures developed in the general introduction in EP1, and also some of the more significant intellectual events of this later period, the following sketch emerges as one way to trace Peirce's development.

In the first selection, "Immortality in the Light of Synechism," written in 1893, Peirce gave an indication of the significance of the argument for continuity that he had planned for a conclusion to his Monist metaphysical series. "I carry the doctrine so far as to maintain that continuity governs the whole domain of experience in every element of it. Accordingly, every proposition, except so far as it relates to an unattainable limit of experience (which I call the Absolute), is to be taken with an indefinite qualification; for a proposition which has no relation whatever to experience is devoid of all meaning." Synechism would guide Peirce's philosophical investigations for the rest of his life. Peirce also signaled his growing conviction that science and religion were closely allied at some deep level.

The following year, in "What is a Sign?" (sel. 2), Peirce explored the relationship between logic and semiotics—even equating reasoning with semiosis. "What is a Sign" is taken from Peirce's unpublished book "How to Reason," also known as "Grand Logic." Elsewhere in that work, Peirce revived the nominalism-realism issue, which he had not dealt with since 1871, and he identified himself, for the first time, as an "extreme" realist.(5) Another year later, in "Of Reasoning in General" (sel. 3), he further developed his semiotic theory of logic elaborating more fully his theory that propositions must always involve two signs, one iconic and the other indexical. These ideas, along with the idea that our success in discovering natural laws is explained by our affinity with nature, would reemerge as key conceptions in Peirce's struggle to rework pragmatism and to account for non-rational human insight. But for a time, he would submerge himself in writing a mathematical textbook called "New Elements of Mathematics,"(6) and also in formal logic, particularly in some elaborate reviews of the recently published volumes of Ernst Schröder's Vorlesungen über die Algebra der Logik.(7)

Near the end of 1896 Peirce took what Max Fisch calls his "most decisive single step" in his progress toward an all-encompassing realism: he accepted "the possible" as a "positive universe" and rejected the nominalist view that the possible is merely what we do not know not to be true.(8) Peirce reported this change of mind in January 1897 in his second Schröder review (CP 3.527) and on 18 March wrote to James that he had "reached this truth by studying the question of possible grades of multitude, where I found myself arrested until I could form a whole logic of possibility" (CP 8.308). With his acceptance of real possibilities—which put Peirce in the Aristotelian wing of the realist camp—Peirce had become what Fisch called "a three-category realist," no longer regarding the potential as what the actual makes it to be, and now distinguishing the generality of firsts from the generality of thirds.

Peirce's embrace of what he would come to call "would-be's" marks a watershed that might be said to separate his middle years from the final period of his intellectual life. This change, in conjunction with his attention to the importance of continuity, would motivate much of the content of his 1898 Cambridge Conferences Lectures. However, the two lectures from that set that are included in the present volume (sels. 4 and 5) were perhaps shaped more by another event: the 1897 appearance of William James's book, The Will to Believe and Other Essays in Popular Philosophy. James had dedicated that book "To my old friend, Charles Sanders Peirce, to whose philosophic comradeship in old times and to whose writings in more recent years I owe more incitement and help than I can express or repay." Peirce was touched, and on 13 March wrote a reflective letter to James expressing his appreciation ("it was a truly sweet thing, my dear William"), and pointing out some ways his thinking had been affected by his experience of "the world of misery" which had been disclosed to him. Although rating "higher than ever the individual deed as the only real meaning there is [in] the Concept," he had come to see "more sharply than ever that it is not the mere arbitrary force in the deed but the life it gives to the idea that is valuable." It is not to "mere action as brute exercise of strength" that we should look if we want to find purpose. Peirce praised James's opening essay, "The Will to Believe," especially for its style and lucidity, but he clearly had reservations. James introduced his essay as an illustration of the continuing concern at Harvard for "vital subjects": it is "a defence of our right to adopt a believing attitude in religious matters, in spite of the fact that our merely logical intellect may not have been coerced."(9) A key point is that "our non-intellectual nature" influences our convictions. "Our passional nature," James wrote, "not only lawfully may, but must, decide an option between propositions, whenever it is a genuine option that cannot by its nature be decided on intellectual grounds." It seems evident that in his Cambridge Conferences Lectures Peirce's great interest in the tensions between theory and practice, and his advocacy of "the will to learn" as a prerequisite to actually learning, were stimulated by James's "The Will to Believe." It is noteworthy that from at least that time on, the role of instinct, or sentiment, as a co-participant with reason in the acquisition of knowledge became a key concern for Peirce, and it would not be long until he came to regard ethics and esthetics as epistemically more fundamental than logic.

Less than six months after hearing Peirce's lectures in 1898, William James traveled to California to address (on 26 August) the Philosophical Union at Berkeley.(10) It was in that lecture, entitled "Philosophical Conceptions and Practical Results," that James publicly introduced the word "Pragmatism." (11) James told his auditors that he would have preferred the name "Practicalism" but that he had settled on "Pragmatism" because that was the name Peirce had used in the early 1870s when he first advocated for pragmatism before the Cambridge Metaphysical Club.(12) James was by this time one of America's most respected intellectuals and his message fell on fertile ground; before long there were a host of pragmatists in the U.S. and abroad. James's acknowledgment of Peirce as the originator of pragmatism increased Peirce's prominence and opened for him an opportunity to bring his distinct views into the growing international debate.(13)

Peirce's second wave of interest in pragmatism is often thought to have started with James's California lecture, but it would be more accurate to say that it began in the early 1890s with the resumption of his research in logic and methodology for his "Critic of Arguments" series for the Open Court, and for his books, "Search for a Method" and "How to Reason." If anything, James's 1890 Principles of Psychology, especially the treatment of the role of inference in perception, probably had more to do with Peirce's return to pragmatism. But it was also about 1890 when Peirce accepted the reality of actuality, or secondness, and then saw clearly that the individual is to be distinguished from the general. It may have been the logical ramifications of that large step toward a more embracing realism, precipitated by his recognition in the mid-80s of the need for both icons and indices for meaningful reference, that led Peirce to begin to rethink the argument of his 1877-78 "Illustrations." Nevertheless, it surely was the increasing popularity of pragmatism that James had spawned in 1898 that led Peirce to resolve to produce a proof that would distinguish his version of pragmatism from popular versions and sanction his as the "scientific" one.

The nineteenth century, after his Cambridge Conferences Lectures, came to a bad ending for Peirce. Between periods of illness and failures to land employment Peirce must have learned more about misery.(14) But he continued to make intellectual progress. On 17 August 1899 he wrote to Carus that "the true nature of continuity . . . is now quite clear to me." Previously Peirce had been "dominated by Cantor's point of view" and had dismissed Kant's definition unjustly. Now he saw that it is best not to try "to build up a continuum from points as Cantor does."(15) He began the twentieth century thinking about great men of science. On 12 January 1901 he published "The Century's Great Men in Science" in the New York Evening Post, noting that "the glory of the nineteenth century has been its science" and asking what it was that has distinguished its great contributors.(16) "Their distinctive characteristic throughout the century, and more and more so in each succeeding generation, has been devotion to the pursuit of truth for truth's sake." He reflected on his own boyhood in Cambridge and on the leaders of the "scientific generation of Darwin," most of whom had passed through his home: "The word science was one often in those men's mouths, and I am quite sure they did not mean by it 'systematized knowledge,' as former ages had defined it, nor anything set down in a book, but, on the contrary, a mode of life; not knowledge, but the devoted, well-considered life-pursuit of knowledge; devotion to Truth—not 'devotion to truth as one sees it,' for that is no devotion to truth at all, but only to party—no, far from that, devotion to the truth that the man is not yet able to see but is striving to obtain." As Peirce's career opportunities dried up he came more and more to regard science and philosophy as devout pursuits.

Fortunately for Peirce, near the end of 1900 James Mark Baldwin hired him to finish the logic definitions after "J" for his Dictionary of Philosophy and Psychology. This work occupied much of Peirce's time in 1901, yet he managed to publish about twenty book reviews and to translate seven articles for the Smithsonian. One of the books Peirce reviewed in 1901 was Karl Pearson's Grammar of Science (sel. 6). An idea Peirce had put forward in his Cambridge Conferences Lectures, that it is illogical to make one's personal well-being "a matter of overwhelming moment," can be seen to be at work in this review. Peirce objected to Pearson's claim that human conduct should be regulated by Darwinian theory and that social stability is the sole justification of scientific research. The human affinity with nature that Peirce had earlier appealed to to explain our success in discovering natural laws (sel. 3), was here explained as resulting from the fact that the human intellect is an outgrowth of the rationality inherent in nature. This was a further rejection of nominalism, which holds that the rationality in nature arises in human reason. Peirce also rejected Pearson's claim that there are first impressions of sense that serve as the starting point for reasoning, and argues that reasoning begins in percepts, which are products of psychical operations involving three kinds of elements: qualities of feelings, reactions, and generalizing elements.

In 1901 in "Laws of Nature" (sel. 7), Peirce reviewed different conceptions of natural law and argued that the typical conception of scientists is that a law of nature is an objective fact—"much more reliable than any single observation." In remarking on the method scientists employ in their "exhumation" of laws of nature, he briefly described a method of conjecture and testing that he would develop in the following selection, "On the Logic of Drawing History from Ancient Documents." In selection 8, Peirce gave one of his most elaborate accounts of the different kinds of reasoning. He drew a distinction between two kinds of deductive reasoning, corollarial, which draws only those conclusions that can be derived from the analysis and manipulation of the premisses as given, and theorematic, which enriches the inference base by adding propositions which were not part of the original premiss set—and "which the thesis of the theorem does not contemplate" (p. 96). Peirce believed this distinction to be the most important division of deductions, and his most important discovery in the logic of mathematics.(17) He also introduced the crucial point he would elaborate in his 1903 Harvard Lectures that "logical criticism cannot go behind perceptual facts"—the "first judgments which we make concerning percepts." Logic cannot criticize involuntary processes. Yet these "first judgments" do represent their percepts, although "in a very meager way."

By mid-1901 Peirce was ready to draw together the many interesting and diverse results he had been achieving into a major book project. The book was to be on logic, but in addition to reflecting his findings on continuity and modality, and his excitement with his progress on a graphical syntax for formal logic, he would incorporate his new discoveries in semiotics and reflect his growing belief that logic is a normative science. The book would be called "Minute Logic" to reflect the minute thoroughness with which he planned to examine every relevant problem. An early draft of the first chapter (MS 425) began with a section entitled "Logic's Promises" and the opening sentence: "Begin, if you will, by calling logic the theory of the conditions which determine reasonings to be secure." Within a year Peirce had drafted and redrafted hundreds of pages, and had finished four large chapters.(18) In July 1902 he prepared an elaborate application asking the Carnegie Institution, presided over by Daniel C. Gilman, to fund his "Logic" which he had reconceived as a set of thirty-six memoirs. His application ran to forty-five pages in typescript, and remains the best single guide to Peirce's system of thought.(19) Even though Peirce received strong recommendations from a powerful group of supporters, including the President, Theodore Roosevelt, and Andrew Carnegie himself, his project was not funded. On 19 June 1903 Peirce's brother, James Mills (Jem) wrote to William James: "Nobody who is familiar with the history of this affair can doubt that the refusal of the Committee is due to determined personal hostility on the part of certain members of the Committee." The matter had dragged on for so long, though, that by the time the rejection was definite, Peirce had already given his 1903 Harvard Lectures and was preparing for his Lowell Institute series—he would never return to his "Minute Logic." Jem wrote to James again on 23 June about the injustice of the Carnegie decision and thanked James for securing the Harvard Lectures for Charles: "I consider that the set of lectures given this Spring at Cambridge and the promise of the Lowell Lectures have saved him from going to ruin. For his fortunes were so desperate, that he could not much longer have resisted forces tending to destroy his bodily health and break down his mind."

The part of the "Minute Logic" included in EP2 is an excerpt from a chapter on the classification of the sciences. In "On Science and Natural Classes" (sel. 9), Peirce described a "natural class" as one "whose members are the sole offspring and vehicles of one idea," and he explained how ideas can "confer existence upon the individual members of the class"—not by bringing them into material existence, but by conferring on them "the power of working out results in this world." Such ideas, Peirce says, when not embodied have a "potential being, a being in futuro." This is Peirce's account of final causation, the power that ideas have "of finding or creating their vehicles, and having found them, of conferring upon them the ability to transform the face of the earth." Such is the power, Peirce believes, of the ideas of Truth and Right. It is in this context that he quotes the famous line from William Cullen Bryant, "Truth, crushed to earth, shall rise again."

In following out this thread of connecting ideas we come to what is probably the single most significant time in Peirce's mature life of ideas, his time in Cambridge in 1903 when he gave his famous "Harvard Lectures," just referred to above, followed not long after by his third series of Lowell Lectures. Peirce had paid close attention to the stream of writings on pragmatism that was gaining momentum and he thought the time had come for him to make a case for a more or less definitive core statement. But making his case or, as he saw it, proving his thesis, was a complicated matter requiring the marshaling of support from all areas of his vast system of thought. Further complicating matters was the fact that Peirce's system had gone through many changes since the 1870s. Among the more significant of those changes, some already mentioned above, was his acceptance of the reality of actuality (secondness) and later of possibility (firstness); his realization that human rationality is continuous with an immanent rationality in the natural cosmos; and his new-found conviction that logic is a normative science, epistemically dependent on ethics and esthetics. For Peirce, pragmatism had become a doctrine that conceptions are fundamentally relative to aims rather than to action per se as he had held in earlier years. To prove pragmatism, then, called for a basic rethinking within the context of a transformed, and still growing, philosophy. That was the task Peirce set out to perform in his 1903 Harvard and Lowell Lectures, and the program he inaugurated that year would guide him for the rest of his life.

In his Harvard Lectures, Peirce built his case for pragmatism on a new theory of perception, grounded in his theory of categories and on results from phenomenology, esthetics, and ethics (sel. 10). He argued that there is a realm of reality associated with each category and that the reality of thirdness is necessary to explain a mode of influence on external facts that cannot be explained by mechanical action alone (sel. 11). He argued that pragmatism is a logical, or semiotic, thesis concerning the meaning of a particular kind of symbol, the proposition, and explained that propositions are signs that must refer to their objects in two ways: indexically, by means of subjects, and iconically, by means of predicates (sel. 12). The crucial element of Peirce's argument, from the standpoint of his realism, involved the connection between propositional thought and perception. To preserve his realism, Peirce distinguished percepts, which are not propositional, from perceptual judgments, which are propositional, and which are, furthermore, the "first premisses" of all our reasonings. The process by which perceptual judgments arise from percepts became a key factor in Peirce's case (sel. 13). But if perceptual judgments are the starting points for all intellectual development, then we must be able to perceive generality (sel. 14). Peirce next argued that abduction shades into perception, so that pragmatism may be regarded as the logic of abduction, and, finally, isolated three key points: that nothing is in the intellect that is not first in the senses; that perceptual judgments contain general elements; and that abductive inference shades into perceptual judgment without any sharp line of demarcation (sel. 15). Pragmatism, Peirce showed, follows from these propositions (sel. 16).

According to Fisch,(20) it was in the Harvard Lectures that Peirce, for the first time, made it clear that his realism was opposed to idealism as well as to nominalism. Peirce's new theory of perception embraced the doctrine of immediate perception, to deny which, according to Peirce, "cuts off all possibility of ever cognizing a relation." That idea was carried forward into the Lowell Lectures, where Peirce continued with his effort to prove pragmatism, making his best attempt so far, according to Fisch.(21) In "What Makes a Reasoning Sound" (sel. 17), the only lecture from the Lowell series that is included in EP2, Peirce made a strong case for objective grounds for evaluating reasonings and argued that with the right method even "a slight tendency to guess correctly" will assure progress toward the truth.

In conjunction with his Lowell Lectures, Peirce prepared a "Syllabus" to be distributed to his auditors. The first part is "An Outline Classification of the Sciences" (sel. 18), showing the normative sciences—esthetics, ethics, and logic—as constituting the central part of philosophy, and giving the order of epistemic and data—support relationships among the sciences that will guide his subsequent research. In "The Ethics of Terminology" (sel. 19), Peirce paused from his central task to elaborate on an issue that had been troubling him since he began working on logic entries in 1900 for Baldwin's Dictionary (and perhaps earlier with his work for the Century Dictionary): the unscientific terminology that prevailed in philosophy. Peirce recognized that philosophy could never abandon ordinary language altogether, for it is essential to understanding common conceptions, but philosophical analysis and progress calls for a specialized vocabulary. That was Peirce's strong conviction, and it explains his frequent resort to neologisms.

It may be that the attention Peirce gave to his classification of the sciences, along with his new-found conviction that logic is coextensive with semiotics, provided the impetus for the remaining two parts of the "Syllabus" that are included in EP2. They introduced a shift to an intensive development of his theory of signs along taxonomic lines motivated by his categories. In "Sundry Logical Conceptions" (sel. 20), Peirce introduced the semiotic trichotomy that divides signs according to whether they are interpreted as signs of possibility, fact, or law: rhemes (here called sumisigns), dicisigns, and arguments. That trichotomy was additional to his long-held division of signs according to whether they represent their objects by virtue of similarity, existential connection, or law: icons, indices, or symbols. In "Nomenclature and Divisions of Triadic Relations" (sel. 21), Peirce introduced another trichotomy that distinguishes signs according to whether, in and of themselves, they are qualities, existents, or laws: qualisigns, sinsigns, and legisigns. With these three trichotomies in place, Peirce was able to identify ten distinct classes of signs. This was the beginning of a rapid development of his formal semiotic theory. There were two other parts of the "Syllabus" that are not included in EP2, one on Peirce's system of Existential Graphs, which Peirce would later choose as the preferred medium for the presentation of his proof of pragmatism, and the other an in-depth treatment of dyadic relations parallel to the treatment of triadic relations found in selection 21.

In the next two selections Peirce shifted his attention from pragmatism and its proof to concentrate more fully on the theory of signs. In "New Elements" (sel. 22), he focused on the abstract mathematical structures necessarily exhibited by sign relations and argued, as he had in "On Science and Natural Classes," that "representations have power to cause real facts" and that "there can be no reality which has not the life of a symbol." And in "Ideas, Stray or Stolen, about Scientific Writing" (sel. 23) Peirce gave one of his most focused accounts of speculative rhetoric, the third branch of his semiotic trivium, which has as its aim to find out "the general secret of rendering signs effective." Peirce made it clear that the range of legitimate semiotic effects (interpretants) includes feelings and physical results, as well as thoughts and other signs. Peirce reiterated a point he had made at least as early as his Harvard Lectures, that nothing can be represented unless it is of the nature of a sign, and he stressed that ideas can only be communicated through their physical effects.

While Peirce was writing about semiotics—and topics outside the scope of this volume (e.g., mathematics and graphical logic)—he had not stopped thinking about pragmatism. On 7 March 1904 he wrote to William James: "The humanistic element of pragmatism is very true and important and impressive; but I do not think that the doctrine can be proved in that way. The present generation likes to skip proofs. . . . You and Schiller carry pragmatism too far for me. I don't want to exaggerate it but keep it within the bounds to which the evidences of it are limited." By this time he was already at work on the first article of another series of papers for the Monist where he would again take up the proof of pragmatism.

Peirce's third Monist series opened with the April 1905 publication of "What Pragmatism Is" (sel. 24). This was to be the first of three papers that would explain in detail Peirce's special brand of pragmatism, give examples of its application, and prove it. Not long into his paper, Peirce paused to deliver a short lesson on philosophical nomenclature—the message being essentially the same as that of selection 19—as a rationale for renaming his form of pragmatism. He chose the name "pragmaticism" as one "ugly enough" to be safe from kidnappers. Peirce lamented that his word "pragmatism" was now met with in the literary journals, "where it gets abused in the merciless way that words have to expect when they fall into literary clutches." He would continue using his new "ugly" word for the rest of the Monist series, and as late at 1909 (sel. 30, p. 457) he used "pragmaticism" because, he wrote, James and Schiller had made "pragmatism" imply "the will to believe, the mutability of truth, the soundness of Zeno's refutation of motion, and pluralism generally"; but he would often revert to his original name, indicating that he may not really have wanted to give it up.

After his excursus into philosophical terminology, Peirce examined the presuppositions of pragmaticism with his proof in mind. One key assumption was that all mental development (learning) takes place in the context of a mass of already formed conceptions, and another was that meaning is always virtual. He also argued for the relevance of all three of the categories of being for his pragmaticism: thought (thirdness) can only govern through action (secondness) which, in turn, cannot arise except in feeling (firstness).

The same year, in "Issues of Pragmaticism" (sel. 25), Peirce restated his pragmatic maxim in semiotic terms, along lines suggested in his sixth Harvard Lecture (sel. 15). He identified the meaning that pragmaticism seeks to enunciate as that of symbols rather than of simple conceptions. The thrust of this article was to articulate his forms of critical common-sensism and scholastic realism, which he regarded as consequences (or "issues") of pragmaticism. He extended his realism to include the acceptance of "real vagues" and "real possibilities," and he pointed out that "it is the reality of some possibilities that pragmaticism is most concerned to insist upon." According to Fisch, pragmaticism had now become pragmatism "purged of the nominalistic dross of its original exposition."(22)

There are a number of manuscript drafts for a third Monist article which indicate that Peirce intended to proceed with his proof along lines he would follow in selection 28. In one of those drafts, "The Basis of Pragmaticism in Phaneroscopy" (sel. 26), he began with an argument from the valency of concepts based in his phenomenology (phaneroscopy) and theory of categories. In another, "The Basis of Pragmaticism in the Normative Sciences" (sel. 27), he focused on the normative sciences, especially on his general theory of signs, as the key to the proof. Peirce pointed out that the pragmaticist will grant that the "summum bonum" consists in a "continual increase of the embodiment of the idea-potentiality" but insisted that without embodiment in something other than symbols, "the principles of logic show there never could be the least growth in idea-potentiality."

Around this time, Peirce was working intensely on the formal structure and systematic interconnections of semiotic relations. His logic notebook (MS 339) in 1905 and 1906 is rife with semiotic analyses and discoveries giving weight to the idea that it was in the context of his theory of signs that he expected to deliver his promised proof of pragmaticism. But when the third article of the series, "Prolegomena to an Apology for Pragmaticism," finally appeared in October 1906, it turned out to be an explication of his system of logical graphs, the Existential Graphs, instead of the expected proof. Peirce had decided that it was by means of the Existential Graphs that he could most convincingly set out his proof, which was to follow in subsequent articles (although it is significantly previewed in this one). Peirce had decided to use his system of graphs for his proof for three principal reasons: it employed the fewest possible arbitrary conventions for representing propositions, its syntax was iconic, and it facilitated the most complete analysis. Peirce worked for years on the continuation of this series, but he never finished it.

It is not known for certain why Peirce was unable to complete his Existential Graphs-based proof, but it is often supposed to have been a consequence of his failure to reach a satisfactory solution to the problem of continuity.(23) It is clear that Peirce expected his argument for pragmatism to also constitute a proof of synechism (see selection 24, p. 335). So it may have been technical problems involving the logic of continuity that kept Peirce from completing this series of papers. Peirce interrupted his efforts to complete this third Monist series with a separate series on "amazing mazes" (two articles of a proposed three were published in 1908-9) in which he developed applications of the Existential Graphs and worked out new definitions of continuity.(24) This mathematical line of thought led Peirce into a number of important technical questions involving probability and modality. By February 1909, Peirce had worked out a matrix method for an extension of the propositional calculus to three values—at least ten years before the similar work of Lukasiewicz and Post.(25) Peirce's acceptance of real possibility had convinced him that the definition of "probability" should include reference to dispositions in addition to frequencies, but even though he tried many alternatives involving the propensity view he was never satisfied that he had got it quite right.(26) For Peirce, this was a matter of considerable importance for pragmatism, because one of the great defects he found with his early theory was the nominalistic appeal to a frequency theory of probability. He also gave up the material interpretation of logical implication.(27)

Among the more entangled and confounding sets of manuscripts in the Harvard collection (the manuscripts acquired by the Harvard Philosophy Department after Peirce's death) is one from 1906-7 in which Peirce attempted to compose a more or less popular account of pragmaticism—but again called "pragmatism"—and to give at least a summary proof (MSS 316-22). Nominally, Peirce was composing a "letter to the editor," initially for the Nation but later for the Atlantic, although Peirce recognized it as a full-fledged article in his correspondence. In the two variants combined in selection 28, Peirce delivered a proof that is probably the one he was intending to give in the Monist before he decided on a more formal approach using his Existential Graphs. The proof in selection 28 is based on Peirce's theory of signs, beginning with the premiss that every concept and every thought beyond immediate perception is a sign, and concluding with the proposition that a final logical interpretant must be of the nature of a habit. This selection provides an illuminating integration of Peirce's theory of signs, including his mature theories of propositions and inference, with his pragmaticism.

It is evident from the refinement of the theory of signs expressed in his remarkable "letter" that Peirce had not given up work on semiotics when he turned to his Existential Graphs for his Monist proof of pragmatism. There may have been a hiatus following his failure to get his "letter" into print, but by August 1908 he was hard at work on the classification of triadic relations (MS 339) and in December he resumed discussion of his theory of signs in correspondence with Lady Welby (sel. 32). Peirce's letters to Lady Welby record, often in summary form, the most advanced theory of signs ever fashioned. The theory as a whole is far too complex to be represented here, although it was lightly sketched in the general introduction in EP1, and a recent book by James Liszka provides an excellent introduction to the system in full.(28) For the thread of intellectual development being pursued here, it is noteworthy that early in 1906 Peirce wrote to Lady Welby that he had found it necessary to distinguish two semiotic objects (immediate and dynamical) and three interpretants (here called "intentional," "effectual," and "communicational"), and he introduced the important conception of the commens, which "consists of all that is, and must be, well understood between utterer and interpreter, at the outset, in order that the sign in question should fulfill its function." On 23 December 1908 Peirce defined "sign" as "anything which is so determined by something else, called its Object, and so determines an effect upon a person, which effect I call its Interpretant, that the latter is thereby mediately determined by the former." He immediately added that the only reason he had inserted "upon a person" into his definition was because he despaired of making his broader conception understood. Over the course of the next few days he laid out his "ten main trichotomies of signs" (eight of them had been quietly given in a single remarkable paragraph on pp. 402-3 of selection 28), the tenth one being the division that expresses the three sources of assurance utterances can have: instinct, experience, or form. This tenth trichotomy would occupy Peirce a great deal during his remaining five years. Peirce's correspondence with William James (sel. 33) repeats many of the same semiotic developments recorded in the letters to Lady Welby, but sometimes more perspicuously and always in a different voice. Modal considerations are more evident in the letters to James. As pointed out above, by 1909 Peirce had made deep advances into modal logic and this is reflected in various ways; for example, in Peirce's emphatic statement that the final interpretant consists in the way every mind "would act," not in the way any mind does act, and also in Peirce's division of semiotic objects into may-be's, actualities, and would-be's.

On 9 April 1908 Peirce received a letter from Cassius J. Keyser inviting him to write an article for the Hibbert Journal. Peirce replied (10 April), outlining ten alternative topics and asking Keyser to choose one. Peirce had written, as his third alternative: "as I believe the Hibbert Journal is favorable to theological discussion, I should willingly treat a little known 'proof' of the Being of God. Properly speaking it is not itself a proof, but is a statement of what I believe to be a fact, which fact, if true, shows that a reasonable man by duly weighing certain great truths will inevitably be led to believe in God."(29) Whether it was Keyser or Peirce who chose the third alternative is not clear, but Peirce spent most of the next three months composing "A Neglected Argument for the Reality of God" (sel. 29).

In that paper Peirce examined the attractive force of the idea of God and concluded that humans instinctively gravitate to it. He contended that belief in God is irresistible to anyone who naturally (through musement) comes to contemplate the possibility of God. The "God hypothesis" appears to be a special kind of abduction (he uses "retroduction" instead of "abduction" in this paper). It arises from a human power of guessing that is analogous to the instincts of animals, and because it recommends itself with unusual force we can take "a certain altogether peculiar confidence" in it as a sign of the truth. Peirce called this his "humble argument" but pointed out that it is not a "proof" because the process leading from the idea of God to belief in God is not a reasoned (self-controlled) development of ideas. Peirce was led to make a distinction between "argument" and "argumentation" that he had not explicitly made before: an argument is "any process of thought reasonably tending to produce a definite belief" while an argumentation is "an argument proceeding upon definitely formulated premisses." An argument, in other words, does not have to be self-controlled. The power of guessing was put forward as "a sort of divinatory power," what Galileo called il lume naturale, and appears to have supplanted Ockham's razor in Peirce's methodological arsenal.

As the conclusion of an "argumentation," the "God hypothesis" must pass through the three successive stages of inquiry: retroduction, deduction, and induction. Peirce devoted nearly half the paper to a discussion of these three stages, but ended up giving only the barest sketch of how they apply in this case. Scientific inquiry requires that any hypothesis be verified by putting its implications to the test of actual experience. The difficulty with the "God hypothesis" is that it is so vague—its object so "infinitely incomprehensible"—that it seems to be impossible to draw any definite implications from its supposed truth. This might appear to fall short of the demands of pragmatism, but, on closer look, one finds that after Peirce embraced the reality of possibility he reconceived the idea of "practical consequences." In his Harvard Lectures he had emphasized that the maxim of pragmatism reaches far beyond the merely practical and allows for any "flight of imagination," provided only that this imagination "ultimately alights upon a possible practical effect." The practical effect that Peirce conceived the "God hypothesis" to "alight upon" is "the self-controlled growth of man's conduct of life." Some scholars wonder whether this weakens the pragmatic maxim beyond recovery—whether, in other words, this opens the way for reinstating into our ontologies all sorts of "beings" that Peirce's earlier pragmatism excluded—but that underscores the fundamental issue raised by this article: whether belief can have any value for the self-controlled growth of the conduct of life if its object is not real.

Peirce's probing of the logic of perception and his reflections on the effectiveness of religious belief, probably along with suggestions that arose from his taxonomic investigations in semiotics, led him in his last years to devote a great deal of thought to "the kinds and degrees of assurance that can be afforded by the different ways of reasoning." The related theory is what Peirce meant by "logical critics," the subject of his intended contribution for a book to honor Lady Welby. That paper, "A Sketch of Logical Critics" (sel. 30), is incomplete, but in the part he finished he made the important point that by "reasoning" we mean a "change in thought" that appeals to a relation between our new cognition (the "conclusion") and "an already existing cognition" (the premiss or premisses) to support our assent in the truth of the conclusion. But not all belief acquisition appeals, in any deliberate sense, to previous cognition, as we saw in the case of perceptual judgments and belief in God. Peirce's conclusion was that knowledge is acquired in two ways, by reasoning, of course, but also by experience. Belief acquired through reasoning must be justified by what preceded it in our minds, but belief gained through experience needs no justification.

In the final article in EP2, "An Essay Toward Reasoning in Security and in Uberty" (sel. 31), Peirce carried further his consideration of the benefits afforded by the different kinds of reasoning—although here again the discussion is left incomplete. This paper, written in October 1913, only a few months before his death, might suggest that he was having doubts about the value of pragmatism. But it would be more accurate to conclude that in his later years Peirce's thought gravitated to ideas and concerns that forced him—or enabled him—to see the limitations of pragmatism. In 1903 he had proclaimed Pragmatism to be "a wonderfully efficient instrument . . . of signal service in every branch of science" (sel. 10). He had recommended it as advantageous for the conduct of life. Now he saw that the appeal of pragmatism was its contribution to the security of reasoning—but there is a price to pay for security. According to Peirce, reasoning always involves a trade-off between security and uberty (rich suggestiveness; potency). Deductive reasoning provides the most security, but it is austere and almost entirely without evocative power. Abduction, on the other hand, is abundant in its uberty though nearly devoid of security. Peirce had come to see that pragmatism has the limitations that come with choosing security over uberty: "[it] does not bestow a single smile upon beauty, upon moral virtue, or upon abstract truth;—the three things that alone raise Humanity above Animality."

Naturalism had grown into a powerful force in Peirce's thought. He had come to believe that attunement to nature was the key to the advancement of knowledge—as it was for life itself—and he thought that the power to guess nature's ways was one of the great wonders of the cosmos. Just as with animals, whose instinct enables them to "rise far above the general level of their intelligence" in performing their proper functions, so it is with humans, whose proper function, Peirce insisted, is to embody general ideas in art-creations, in utilities, and above all in theoretical cognition. But if attunement to nature is the key to the advancement of knowledge, it is at most a necessary condition; it puts thought on the scent of truth, which, to attain, must be won by skilled reasoning. Peirce remained a logician to the end.

This concludes the thread of development chosen here to draw together the separate papers in EP2, but it is only one of many approaches that could have been taken. Peirce's shift to a graphical syntax for his formal logic, with its corresponding emphasis on the importance of icons for reasoning, led to remarkable results in logic and in philosophy that parallels the course of development outlined above. Alternatively, the evolution of Peirce's theory of signs that is evident throughout EP2 might have been more systematically used to mark movements in Peirce's thought through these years. Or one might have expanded on Fisch's account of Peirce's ever-strengthening commitment to realism—or have followed the shifting influence of major thinkers and scientific discoveries on Peirce's thought. These and other approaches could be turned into useful heuristic guides to Peirce's intellectual life in his final two decades. But the growth of his pragmatism and, in particular, the development of its proof, surely represents a strong current running through the period and for much of it probably best represents Peirce's leading idea.

Something more should be said about Peirce's proof of pragmatism—one of the great puzzles for Peirce scholars. Max Fisch characterized it as "elusive" and Richard Robin says it is "unfinished business."(30) When he first claimed publicly in 1905 to have a proof (sel. 24), he said it was "a proof which seems to the writer to leave no reasonable doubt on the subject." Elsewhere he called it a "strict proof" or "scientific proof." We should not accept the pragmatic maxim, Peirce told the auditors of his second Harvard Lecture (sel. 11), "until it has passed through the fire of a drastic analysis." Peirce literally meant to "prove" pragmatism—but in the sense called for by philosophy. Philosophical proofs seek to prove truths, not just theorems (they strive to be sound, not just valid), and must therefore be concerned with establishing the truth of their premisses. Only rarely is the deductive form of a philosophical argument in dispute; the crucial questions almost always have to do with the legitimacy and strength of the premisses. And as with science generally, establishing the relevance and truth of contingent premisses calls for non-deductive forms of reasoning. As a result, proving pragmatism calls for marshaling an appropriate set of assumptions and supportable claims which, as premisses, will entail pragmatism as expressed in Peirce's maxim. In his first Harvard Lecture, to add to the "strictness" of the proof, Peirce deliberately expressed his maxim as a theorem: "Pragmatism is the principle that every theoretical judgment expressible in a sentence in the indicative mood is a confused form of thought whose only meaning, if it has any, lies in its tendency to enforce a corresponding practical maxim expressible as a conditional sentence having its apodosis in the imperative mood." So when Peirce claimed to have a proof of pragmatism, he meant that he could produce what he believed to be a convincing rationale, an argument (or, as he would say in his "Neglected Argument," an argumentation), to demonstrate that the pragmatic maxim, in a given form, strictly follows from a given set of premisses, and, furthermore, that each of the premisses is either a common assumption or can otherwise be shown to be admissible.

When Peirce's efforts to prove pragmatism are understood to be attempts to provide a convincing rationale or argument for the truth of his maxim, it makes sense to suppose that his first proof began to take shape in the early 1870s when he promoted pragmatism among the members of the Cambridge Metaphysical Club. His first published proof, then, would have been the argument of his "Illustrations." This is the view expressed by Max Fisch(31) and it is strongly supported by Peirce himself in his first Harvard Lecture (sel. 10): "The argument upon which I rested the maxim in my original paper was that belief consists mainly in being deliberately prepared to adopt the formula believed in as the guide to action." This belief, in turn, was carried back to "an original impulse to act consistently, to have a definite intention." But this is a "psychological principle" and by 1903 Peirce no longer thought it "satisfactory to reduce such fundamental things to psychology." Besides, as he wrote in the "additament" to his "Neglected Argument" (sel. 29), "I must confess the argument . . . might with some justice be said to beg the question." We might think of this early proof as the proof based on Peirce's theory of belief.

By 1903 Peirce had devoted a great deal of study to scientific proofs and to epistemic support relationships across sciences. By then he was much better prepared to build a proof of pragmatism, and it is clear that he was thinking of "proof" in a more rigorous sense. In his more technical restatement of his maxim for his Harvard Lectures, pragmatism was restricted to conceptions that can be expressed in sentential form. According to the pragmatic maxim, so stated, the meaning of a theoretical judgment expressible in a sentence in the indicative mood (what was originally expressed as "the object of our conception") lies in its tendency to enforce a corresponding practical maxim that takes the form of a conditional sentence (originally, "our conception of effects that might conceivably have practical bearings"). This is the thesis Peirce set out in 1903 to demonstrate. How did he go about it? Roughly by establishing, first, that all intellectual contents amount to theoretical judgments expressible in indicative sentences and, second, that all such judgments fundamentally appeal to imperative practical conditionals. To support the first part, he established: (1) nothing is in the intellect that was not first in the senses, (2) the process by which sensory stimulation rises to perceptual judgment is not subject to self-control, (3) perceptual judgments cannot be called into question and are the first premisses of all our reasonings, (4) perceptual judgments contain general (i.e. interpretative) elements (as in predicates of propositions), and (5) although literally particular, perceptual judgments entail general propositions. Then Peirce argued that (6) the process which results in perceptual judgments is a quasi-abductive process (depending on intellectual habits) which "interprets" percepts as cases falling under practical conditionals (and, therefore in relation to a purpose). This effectively proved his thesis. We might think of this as Peirce's proof of pragmatism based on his theory of perception.

In "Pragmatism" (sel. 28), Peirce shifted the burden of his proof to his theory of signs. He began by developing his thesis along lines he seemed to initially have had in mind for his Monist proof (see selection 26). First he characterized pragmatism as a method of ascertaining the meaning of "intellectual concepts" and he noted that "triadic predicates" are the principal examples (although, in passing, he considered whether there might be non-intellectual triadic relations). He noted that while signs can convey any of three forms of predicates (monadic, dyadic, or triadic), only triadic predicates are properly called "intellectual concepts." Only intellectual concepts convey more than feeling or existential fact, namely the "would-acts" of habitual behavior; and no agglomeration of actual happenings can ever completely fill up the meaning of a "would-be." This line of thought (with many steps left out) led Peirce to his thesis, what he called "the kernel of pragmatism" (p. 402): "The total meaning of the predication of an intellectual concept consists in affirming that, under all conceivable circumstances of a given kind, the subject of the predication would (or would not) behave in a certain way,—that is, that it either would, or would not, be true that under given experiential circumstances (or under a given proportion of them, taken as they would occur in experience) certain facts would exist." He also expressed his thesis in a simpler form: "The whole meaning of an intellectual predicate is that certain kinds of events would happen, once in so often, in the course of experience, under certain kinds of existential circumstances." This is what Peirce set out to prove in 1907.

Peirce's proof, much abbreviated, ran something like this:

1. "Every concept and every thought beyond immediate perception is a sign."

2. The object of a sign is necessarily unexpressed in the sign.

3. The interpretant is the "total proper effect of the sign" and this effect may be emotional, energetic, or logical, but it is the logical interpretant alone that constitutes "the intellectual apprehension of the meaning of a sign."

4. "A sign is anything, of whatsoever mode of being, which mediates between an object and an interpretant; since it is both determined by the object relatively to the interpretant, and determines the interpretant in reference to the object, in such wise as to cause the interpretant to be determined by the object through the mediation of this 'sign.'"

5. The logical interpretant does not correspond to any kind of object, but is essentially in a relatively future tense, what Peirce calls a "would-be." Thus the logical interpretant must be "general in its possibilities of reference."

6. Therefore, the logical interpretant is of the nature of habit.

7. A concept, proposition, or argument may be a logical interpretant, but not a final logical interpretant. The habit alone, though it may be a sign in some other way, does not call for further interpretation. It calls for action.

8. "The deliberately formed, self-analyzing habit . . . is the living definition, the veritable and final logical interpretant."

9. "Consequently, the most perfect account of a concept that words can convey will consist in a description of that habit which that concept is calculated to produce. But how otherwise can a habit be described than by a description of the kind of action to which it gives rise, with the specification of the conditions and of the motive?"

This conclusion is virtually a paraphrase of Peirce's thesis, the "kernel of pragmatism," so it completes his proof. We might think of this as the proof from Peirce's theory of signs. On 10 April 1907, Peirce sent Giovanni Papini a similar, though somewhat fuller, outline and explained that "among all scientific proofs with which I am acquainted [this is] the one that seems to me to come nearest to popular apprehension."(32)

When Peirce began his third Monist series, represented in EP2 in selections 24-27, he probably had something like the above proof in mind, although perhaps something more wide-ranging. The definition of pragmatism as set out in "What Pragmatism Is" (sel. 24) gives some idea of what he was aiming for: pragmatism, he wrote, is "the theory that a conception, that is, the rational purport of a word or other expression, lies exclusively in its conceivable bearing upon the conduct of life; so that, since obviously nothing that might not result from experiment can have any direct bearing upon conduct, if one can define accurately all the conceivable experimental phenomena which the affirmation or denial of a concept could imply, one will have therein a complete definition of the concept, and there is absolutely nothing more in it" (332). Peirce pointed out that to prove this thesis it would be necessary to appeal to a wide range of "preliminary propositions." Don D. Roberts has listed seventeen "premisses" that he thinks are likely to be among the ones Peirce had in mind, and these include "dismiss make-believes," "logical self-control is a mirror of ethical self-control," "an experiment is an operation of thought," "we do not doubt that we can exert a measure of self-control over our future actions," "a person is not absolutely individual," and "thinking is a kind of dialogue."(33)

Midway through his third Monist series, Peirce changed his mind and decided to base his proof on his Existential Graphs. He never completed his graph-based proof, but there are many manuscript pages indicating what he had in mind. In one draft (MS 298) Peirce explained: "You 'catch on,' I hope. I mean, you apprehend in what way the system of Existential Graphs is to furnish a test of the truth or falsity of Pragmaticism. Namely, a sufficient study of the Graphs should show what nature is truly common to all significations of concepts; whereupon a comparison will show whether that nature be or be not the very ilk that Pragmaticism (by the definition of it) avers that it is. . . ."

That proof, as represented in preliminary form in Peirce's 1906 "Prolegomena to an Apology for Pragmaticism" (CP 4.530-72) and in MSS 296-300, is extremely complex. It depends heavily on establishing that the system of Existential Graphs provides a working model of thought and that experimenting with the Graphs amounts to experimenting with concepts themselves. The sweep of issues addressed in the premisses of this proof includes: that the proper objects for investigation in experiments with diagrams are forms of relation; that deductive reasoning is no more certain than inductive reasoning when experimentation can be "multiplied at will at no more cost than a summons before the imagination"; that icons have more to do with the living character of truth than either symbols or indices; that reasoning must be chiefly concerned with forms; that diagrams are icons of the forms of relations that constitute their objects; that members of a collection, taken singly, are not as numerous as the relations among them; that there can be no thought without signs and there are no isolated signs; that every logical evolution of thought should be dialogic; and that thought is not necessarily connected with a brain. This is only a sampling. There is little doubt that the full exposition of Peirce's Graphs-based proof would shed considerable light on the complex network of relationships internal to Peirce's system of thought that support pragmatism, but it is not so clear whether its upshot would be to prove pragmatism or to prove that the system of Existential Graphs is a valid normative logic of cognition—really a "moving picture of thought" as Peirce once said (CP 4.11).

Most of Peirce's arguments for pragmatism, and there are a number that have not been mentioned, seem to be quite straightforward in setting out what is to be proved—the pragmatic maxim as a carefully stated thesis—and in supplying the assumptions and premisses that entail that thesis as conclusion. The intractibility of these arguments usually results from their large number of premisses, ranging over vast sweeps of Peirce's system of thought, and from the difficulty involved in establishing the premisses. But the matter is complicated by the fact that many of the involved premisses require inductive support, and by apparent promises of inductive confirmation for the pragmatic conclusion, which Peirce thought his readers might hesitate to accept because of the overall complexity of the argument and the novel ideas it involved.(34) An important question emerges: What kind of principle is the pragmatic maxim after all? Is it a logical maxim and a regulative principle, or is it a positive truth that can be treated as a scientific hypothesis calling for inductive confirmation? Peirce's treatment suggests that it is both. But as a positive truth informing us how to construe the meaning of conceptions or propositions—signs with intellectual value—how could the pragmatic maxim be confirmed? In criticizing the argument of his 1877-78 "Illustrations," Peirce disallowed any appeal to psychology, and in any case his classification of the sciences shows that the only positive sciences that can legitimately be appealed to are phenomenology and the prior normative sciences (and parts of logic) on which logical methods must rely. Peirce thought the maxim could be tested by using it to analyze familiar intellectual conceptions such as "real," "identity," "sequence," "substance," "time," and "probability," but only after he had established that his logical analyses of those conceptions was neither psychological nor question-begging. That seems to be why he had first to prove that working with his Existential Graphs was "equivalent" to working with conceptions themselves. His proof from the Existential Graphs, then, appears to have been integral to his effort to prove pragmatism inductively. One of the limitations of this approach is that it can never wield demonstrative force, and the argument can always be carried further; but the hope must be that the time will come when further confirmation is beside the point. It is probably this inductive approach that has lent support to the view that Peirce's proof is rather amorphous and perhaps at best a cable with fibers of independent sub-arguments. Overall, it is easy to see why Thompson said that a "real proof" of pragmatism "would amount to a kind of elucidation of most of Peirce's philosophy and formal logic" and why Robin said that "coming to terms with pragmatism's proof" means coming to terms "with the whole Peirce."(35)

When Peirce died in the spring of 1914 he left a lot of important work unfinished. Perhaps most to be regretted is that he was unable to complete his "System of Logic, Considered as Semeiotic," which he hoped would stand for realism in the twentieth century as Mill's System of Logic had stood for nominalism in the nineteenth.(36) As it was, he did leave far more than has since been put to good use. More than fifty years ago, the great American social philosopher, Sidney Hook, wrote of Peirce that "he is just as much the philosopher's philosopher [today], just as much the pioneer of a second Copernican revolution in thought (one more genuine than Kant's) as he was when his meteoric genius first flashed across American skies."(37) It is still true that Peirce is mainly a "philosopher's philosopher." But it may turn out that his pioneering work, perhaps especially his later writings so tightly packed with ideas, will bloom at last into the influential legacy that Peirce in hopeful moments imagined would be his bequest to the future. Perhaps this collection, in spite of its limitations, will contribute to that end.

Nathan Houser

1. See Max H. Fisch, Peirce, Semeiotic, and Pragmatism, K. L. Ketner and C. J. W. Kloesel, eds. (Indiana University Press, 1986), pp. 227-48.

2. Murray G. Murphey, The Development of Peirce's Philosophy (Harvard University Press, 1961; Indianapolis: Hackett Publishing Co., 1993), p. 3.

3. Peirce's life was long neglected and is still obscure. The best accounts can be found in: Fisch, Peirce, Semeiotic, and Pragmatism; Joseph Brent, Charles Sanders Peirce: A Life (Indiana University Press, 1993; revised ed. 1998); and Kenneth Laine Ketner, His Glassy Essence: An Autobiography of Charles Sanders Peirce (Vanderbilt University Press, 1998).

4. Harold Henderson, Catalyst for Controversy: Paul Carus of Open Court, (Southern Illinois University Press, 1993). The Open Court Publishing Company was owned by the Chicago industrialist Edward C. Hegeler.

5. Fisch, Peirce, Semeiotic, and Pragmatism, p. 193

6. Peirce's book was completed but not published in his lifetime. See notes 2-4 to selection 22 (p. 537).

7. Peirce's reviews appeared in the Nation and the Monist; see P620 (CP 3.425-455), P627 (CN 2:132-33), and P637 (CP 3.456-552).

8. Fisch, Peirce, Semeiotic, and Pragmatism, p. 194.

9. This and the quotations that follow in this paragraph are from the opening essay of William James, The Will to Believe and Other Essays in Popular Philosophy (Longmans Green, 1896; Harvard University Press, 1979).

10. For Fisch's full account see Peirce, Semeiotic, and Pragmatism, pp. 283 ff.

11. K. L. Ketner and H. Putnam speculate that James's new-found interest in pragmatism, as well as "Royce's drift toward Peirce's ideas," was a consequence of Peirce's 1898 Cambridge Lectures (RLT 36).

12. Peirce's key anti-foundational arguments had appeared earlier in his 1868 Journal of Speculative Philosophy series; EP1, selections 2-4.

13. According to Murray Murphey, James's lecture put Peirce "in an intolerable intellectual position." Peirce could not now disown pragmatism, but neither could he "embrace it without qualification." Peirce had to come forward with his distinct views (The Development of Peirce's Philosophy, pp. 358-59).

14. See Brent, Charles Sanders Peirce: A Life, ch. 4.

15. Quoted in Eisele's NEM 3:780.

16. This article, as reprinted in the Annual Report of the Smithsonian Institution for Year Ending June 30, 1900 (Washington, D.C., 1901) is published in Philip P. Wiener, ed., Charles S. Peirce: Selected Writings (Dover, 1966), pp. 265-74. Quotations are taken from Wiener's book.

17. Fisch, Peirce, Semeiotic, and Pragmatism, p. 334

18. For an illustration of the logical depth of Peirce's work for this book, see the chapters by Glenn Clark and Shea Zellweger in Studies in the Logic of Charles Sanders Peirce (Indiana University Press, 1997).

19. Peirce's application to the Carnegie Institution (L 75) is available electronically on the Peirce-focused website: http://www.door.net/ARISBE/arisbe.htm.

20. Fisch, Peirce, Semeiotic, and Pragmatism, p. 195.

21. Fisch, Peirce, Semeiotic, and Pragmatism, p. 365.

22. Fisch, Peirce, Semeiotic, and Pragmatism, p. 195.

23. See Fisch, Peirce, Semeiotic, and Pragmatism, p. 365 and Richard S. Robin, "Classical Pragmatism and Pragmatism's Proof" in The Rule of Reason: The Philosophy of Charles Sanders Peirce, Jacqueline Brunning and Paul Forster, eds. (University of Toronto Press, 1997), p. 149.

24. Fisch, Peirce, Semeiotic, and Pragmatism, p. 196.

25. See Fisch, "Peirce's Triadic Logic" (written with Atwell Turquette) in Peirce, Semeiotic, and Pragmatism, pp. 171-83, for details and for further remarks on triadic logic.

26. Fisch, Peirce, Semeiotic, and Pragmatism, p. 196.

27. According to Fisch (Peirce, Semeiotic, and Pragmatism, p. 196), material (Philonian) implication was Peirce's last nominalist stronghold.

28. James Jakób Liszka, A General Introduction to the Semeiotic of Charles Sanders Peirce (Indiana University Press, 1996).

29. Peirce to C. J. Keyser, 10 April 1908 (Rare Book and Manuscript Library, Columbia University).

30. Fisch, Peirce, Semeiotic, and Pragmatism, p. 363, and Robin, "Classical Pragmatism and Pragmatism's Proof," p. 149.

31. Fisch, Peirce, Semeiotic, and Pragmatism, p. 363.

32. Peirce to G. Papini, 10 April 1907 (Papini Archives).

33.Don D. Roberts, "An Introduction to Peirce's Proof of Pragmatism," Transactions of the Charles S. Peirce Society 14 (1978), p. 128.

34. See, for example, MS 300 and Roberts, "An Introduction to Peirce's Proof of Pragmatism," p. 129, for some elaboration.

35. Manley Thompson, The Pragmatic Philosophy of C. S. Peirce (University of Chicago Press, 1953), p. 249. Robin, "Classical Pragmatism and Pragmatism's Proof," p. 150.

36. See MS 640 and NEM 3:875; and Fisch, Peirce, Semeiotic, and Pragmatism, p. 196. Many manuscripts from Peirce's last decade develop logic from the standpoint of semiotics but, perhaps, none more fully than MS 693.

37. Quoted from a tribute solicited by Frederic Harold Young and published by him in Charles Sanders Peirce; America's Greatest Logician and Most Original Philosopher (privately published, 1946), an address delivered in October 1945 to the Pike County Historical Society in Milford, Pennsylvania.