“Peirce's Conception of Architectonic and Related Views”

Practically all books and not a few articles on Peirce point out the fact that the American philosopher highly commended the parallel drawn by Kant between a philosophical doctrine and a piece of architecture.^* This coincidence is understandable: Peirce's idea of a philosophic architectonic looks to be anything but a casual remark. The editors of the Collected Papers had already a keen perception of the central rôle played by Peirce's idea when they included some of his thoughts on the topic as a foreword to his strenuous and never dismissed efforts toward a classification of the sciences. In this respect they followed Peirce's own consciousness of the importance of a philosophic architectonic when he prefaced five of his most significant metaphysical articles with some further comments on it. We might therefore conclude by saying that Peirce, no matter how little systematic he himself turned out to be at the end, was undoubtedly haunted by the idea of system. But this conclusion, even if correct, is scarcely enlightening. It remains to be seen what kind of system Peirce had in mind and whether some related views may help to shed additional light on his philosophical attitude.

It is not therefore my purpose to discuss whether Peirce's idea of architectonic lies at the very core of his philosophy, as contended, at least implicitly, by Mr. Feibleman, or whether it expresses only one side, the transcendentalist side of it, as argued by Mr. Goudge. In other words, I do not propose myself to look at Peirce as a systemic philosopher, as was so ably done by Mr. Weiss, or to consider Peirce's systematic leanings as an occasional mood as assumed by Mr. Buchler. I suspect that all these excellent interpreters of Peirce's thought might very well argue indefinitely and always find a sufficient reason, namely, a pertinent quotation, for them to remain entrenched in their respective positions. These various possibilities of interpreting Peirce are due certainly to the fact that the American philosopher provided enough conflicting statements on what philosophy is, or rather should be, to permit interminable and, let me add, not necessarily unfruitful discussions on his "real position." But they arise also from the fact that all these interpreters have placed more emphasis on the question of whether or not Peirce possessed a system than upon the question of the kind of system he had in mind when proposing his philosophic architectonic or when in various types of systematic construction, most of them, as it is known, led by a persistent trichomania. My aim is not to give an interpretation of Peirce's system or to reach the conclusion that such an interpretation cannot be given, but merely to examine his theory about systems.

The task is not easy because, to begin with, the word 'system' is a highly ambiguous one. A contemporary logician, for example, will understand by this word something quite different than, say, a classical philosopher. And even if we drop, for the sake of simplification, the logician's idea of formal system—an idea, let it be said in passing, to which Peirce contributed a great deal through his researches on formal logic—and restrict our meaning to the philosophical one, we possess as many ideas of what a system is as there are great systematic philosophies. Baffled by this variety, Peirce took two different courses in his investigations on the idea of system. On the one hand, he was concerned with the problem of what is a system. On the other hand, he seemed anxious to provide a classification of systems. The first problem was dealt with by means of an elaboration of Kant's comparison between a philosophic system and a piece of architecture. The second one was tackled by applying his own doctrine of categories to several outstanding systems of the past. Now, these two different ways of dealing with the problem of philosophic systems are the source of a great deal of confusion. For Peirce was not always eager to make the proper distinction between research on the conception of system and research on different kinds of systems. He seemed to forget that while the former research belongs to the methodology of philosophy, the latter one follows from history or from psychology. In order to avoid useless complexities we shall introduce a further restriction to the present paper. Our analysis of Peirce's architectonic will exclude, therefore, both its most abstract aspect—the doctrine of logical and mathematical formal systems—and its most concrete aspect—the doctrine of the classifications of historical systems. Thus circumscribed, our topic will appear perhaps as a very minor one. Yet, only at the price of narrowing it down to its essentials shall we be permitted to gain a certain clarity on it.

What type of system had Peirce in mind when reviving the Kantian conception of a philosophic architectonic? Before attempting an answer, it will prove convenient to examine Kant's suggestion itself. We shall see very soon that while it contains something that is relevant to Peirce's conception, it would be a grave mistake to proceed hastily and assume that all of the elements of Kant's architectonic are again present in Peirce. The comparison of both ideas will help to understand what Peirce himself, after all, had often acknowledged: that his ideas, on architectonic or on any other fundamental issue, were at the same time a continuation of Kant and a reaction against Kant.

As it is well known, Kant introduced his conception of the architectonic of pure reason in the transcendental doctrine of method and, therefore, at that stage of his thought in which he attempted a rational justification of metaphysics. The architectonic was defined as the art of constructing a system, for knowledge was not accepted as a science unless it possessed a systematic unity. According to Kant, knowledge cannot remain in a rhapsodic state, for then it could not advance the ends of reason. Thus a system was defined as "the unity of various cognitions under one idea." Thanks to this idea the system becomes an organism and ceases to be a mere aggregate. Thus a system can grow from within, without necessarily changing its proportions. In order to obtain such a result, it is necessary to provide a scheme, but a scheme projected in accordance with an idea, namely, from the standpoint of the highest aim of reason. Otherwise, the unity achieved will be a mere technical, but never an architectonic, unity. A technical unity is obtained from the observance of similarities; an architectonic unity, from an idea providing the possibility of the scientific whole. It should be noted that Kant does not propose such an idea as a result of an arbitrary decision of an individual. He is perfectly aware of the fact that in the concrete elaboration of a science, the so-called rhapsodic stage may be inevitable. This inevitability is, however, only psychological; when considered from the point of view of its structure, it will always be clear that, unless it turns out to be a mere aggregate, a science has been outlined according to a definite, although germinal, plan of arrangement. This is due to the fact that the scheme or germ of all lies in reason. But it is also due to the fact that reason is not something alien to and superimposed upon the human mind, but the root of the human mind itself. Reason is the form and substance of human legislation. The essential aims of science coincide with the essential aims of human reason because they are both different aspects of the same reality: the reality of man—or, if you prefer, of the transcendental subject—as the legislator of the universe.

It does not even matter that such a legislator exists nowhere. In harmony with Kant's famous idea of science as an infinite process, the prototype of this legislator—the philosopher—is a being in the making rather than an already existing entity. Legislative power is, therefore, an idea residing in the mind of every man. For that reason, the idea of legislation to which philosophy ultimately boils down is a cosmical conception. Scholastic conceptions are always partial; only the cosmical conception of philosophy is complete. Philosophic architectonic is thus the expression of the ultimate destiny of man, so that what appears at the beginning as a bold theoretical speculation becomes at the end a postulate of practical reason. The famous Kantian jump from theoretical reason to practical reason is no jump at all, for the preeminence of the latter already appears in the frame of the former. That Moral philosophy occupies a superior position in the Kantian system is a well-known fact. But that such a superior position had been proclaimed by Kant while he was still busy with theoretical speculative reason deserves to be better known than it is even by the most fastidious commentators of the German philosopher.

I have dealt with Kant's conception of architectonic in a more extensive way than a short paper would permit, because it seems to me that not a few discussions of the relationship between Peirce and Kant have failed to notice the complexities of their conceptions about architectonic. When considered in full Kant's and Peirce's ideas in this respect strike us as being at the same time very similar and quite dissimilar. It will be rewarding to point out their resemblances and differences before proceeding to an appraisal of Peirce's idea of a philosophical system.

It is unquestionable that Peirce would agree—or rather did agree—with Kant in the postulate that philosophy must by planned and that this planning is by analogy equivalent to the planning of a piece of architecture. It is also unquestionable that both thinkers would hail the idea that "happy thoughts which have accidentally occurred" to the philosophers, as Peirce puts it, are of little help for the development of a comprehensive philosophy. These thoughts may be "pitchforked" in volumes which make easy and pleasing reading. They will never produce, however, a conception covering the whole of reality unless they follow a plan previously laid down by their authors. There are some other respects in which Peirce and Kant completely agree. For instance, that philosophical activity must be deliberate and, whenever possible, highly conscious, that the arbitrary and the individualistic are prejudicial and, last but not least, that Philosophy must be like a building capable of sheltering everybody and not only a few technical-minded philosophers. Yet, when we come to a closer examination of the roots of the agreement we notice that they persist only in so far as the expressions used by both philosophers are based upon vague analogies and, to a certain extent, upon metaphors. Peirce in particular is often fond of such metaphorical usage. He seems even to view philosophy as something almost concrete, made up of solid blocks of stone that the philosophers of innumerable ages polish, combine and distribute—or redistribute—according to a certain blueprint provided by a man of genius—Aristotle or perhaps Hegel. He speaks of the architect's soul, of messages with which an age is charged and which the architect delivers to posterity. He writes about composition, contrasting the easy problems of composition raised by painting with the ponderous questions of distribution of elements posed by building-construction. He tells us about houses made up of papier mâché, that are built with the help of an interesting, but weak idea, as contrasted with houses made up of granitic stuff, capable of withstanding the storms of time because built on the basis of ideas that are not only appealing and subtle, but also sound. He even speaks of a synchronism between the different periods of medieval architecture and the different periods of logic. If Kant had been inclined to the casual way of talking in which Peirce sometimes indulges, he would have undoubtedly hailed with enthusiasm these charming descriptions of architectonic. Unfortunately, his approval would not go very far. As soon as the metaphorical level was abandoned, sharp disagreements would arise between the two philosophers.

The first disagreement is important enough to ruin whatever Kantianism was left in Peirce's philosophy. It concerns the faculty which is supposed capable of mapping out the whole system of a science, namely, of giving a priori the plan of it. Although Kant does not mention specifically any given faculty, only one remains possible on the basis of his assumptions: imagination. Now, the word 'imagination' has various meanings. Two of them seem to be predominant. On the one hand, we can speak of imitative imagination. On the other hand, we can talk about productive imagination. The former type of imagination limits itself to reproducing and occasionally to combining already existing concrete images; the latter type proceeds to arouse general images susceptible of being filled out with actual perceptions. To use Kant's vocabulary, imitative imagination is phantasy, whereas productive imagination is a facultas imaginandi, in the proper sense of this expression, an Einbildungskraft. It is well known the rôle that this facultas imaginandi plays in Kant's theory of the schematism of categories. In fact productive imagination turns out to be a condition a priori of the possibility of unification of diversity in the field of knowledge. I shall not therefore ponder further over this problem. It will suffice to point out that this rather obscure doctrine of imagination is based upon an assumption that permeates the whole of Kantian epistemology: the assumption that understanding is a spontaneous activity, capable of achieving that most controversial aim of all the idealist-minded philosophers, namely, transcendental synthesis. The remnants of epistemological realism that were still operative in the transcendental Esthetics soon became crushed under the powerful impact of an unbridled idealism. Now, in order to build the architectonic unity of the sciences Kant gave free rein to creative imagination, to such an extent that the philosopher—the supreme architect of reason—became the creator assuredly not of the universe, but of the frame of the universe. The fact that the philosopher, according to Kant, is not an artist, must not deceive us in this respect in view of the peculiar meaning given by the German thinker to the word 'artist.' For an 'artist' is 'only' a man who occupies himself with conceptions, while a philosopher is a man who legislates on conceptions. It is true that when reaching this stage of his analysis Kant had to acknowledge that the title of philosopher can be granted to nobody, for its perfection lies only in its idea and not in its concrete fulfilment. None the less, according to Kant, a philosopher is a philosopher only in so far as he is capable of behaving in accordance with that high prototype of knowledge that makes of him a living image of God himself.

It is hardly necessary to point out that such a speculation is entirely alien to Peirce's thought even if we raise to a maximum his metaphysical tendencies and water down to a minimum his undeniable naturalistic and empiristic leanings. Between Kant and Peirce only a common basis of agreement remains: that philosophy has, at least sometimes, a programmatic character and that to be busy with blueprints and tables of contents is an activity that cannot be lightly dismissed by a true philosopher. Beyond this all resemblances peter out until they vanish completely. Now, the main source of disagreement lies in the entirely different, nay opposite ideas that the two philosophers had about the meaning of the word 'system.' A few words on this point will therefore be needed before proceeding further in our analysis.

From what has been said, Kant's idea of a system appears to be quite clear: a system is something completed in itself, no matter how much time mankind needs in order to grasp it entirely. It will perhaps be argued that Peirce is at times quite close to such a postulate of completeness. When he proposed "to erect a philosophical edifice that shall outlast the vicissitudes of time" and announced that his undertaking was "to make a philosophy like that of Aristotle, that is to say, to outline a theory so comprehensive that, for a long time to come, the entire work of human reason, in philosophy of every school and kind, in mathematics, in psychology, in physical science, in history, in sociology, and in whatever other department there may be, shall appear as the filling of its details," he seemed to be most faithful to the aforementioned ideal of completeness. Yet, even in this passage—assuredly the most unbridled of all passages written by Peirce about architectonic—a few restrictions appear that belie any hopes that Peirce had yielded to speculative idealism. To begin with, Peirce stresses the temporal condition of the systematic undertaking. Furthermore, he speaks of a very broad outline, solid and unshakable enough not to be easily undermined, but also vague and rough enough to permit enlargement and, if necessary, modification. If that was not sufficient to convince the reader that Peirce's philosophic architectonic is anything but a universal legislation, he adds a few though convincing lines about that most permanent feature of his thought: fallibilism. To find out and not to prescribe is, avowedly, the main aim of Peirce's philosophy, not only when he speaks, to use again Mr. Goudge's interpretation, the empiricist and naturalist language, but also when he indulges in the transcendentalist vooabulary. Thus to present a philosophic blueprint to be filled up by the centuries, is not to postulate a system that is, or will be, complete, but to foster a system that is not, and can never be, completed. 'System' means in Peirce's philosophy a body of compositions which must unceasingly and actively be tested. At no moment should a system be looked at, as Peirce puts it, with a "vacant eye."

Many reasons might be adduced to explain why Peirce's conception of a philosophic architectonic is an open rather than a closed one. It will suffice to point out two of them.

First, a system applies to the whole of the sciences. But sciences are not rigid schemata; they are living historical entities. For this reason, no abstract definition can ever be given of a science. In this respect, the famous distinction proposed by Peirce between laboratory and seminary philosophies must be taken at its face value. In a laboratory, science is an object for inquiry; in a seminary, it is at most a subject of dissection. In a laboratory, science is a living organism; in a seminary, a dead issue. Like the universe, science is permeated with continuous growth. Now, if architectonic is a blueprint for the sciences, it must not be a schematic problem, but an outline capable of following the meanderings and windings of science. The great success of Aristotle's philosophy was due precisely to the faithfulness of the Greek philosopher to this postulate of flexibility. And "the secret of Hegel" consisted in nothing but in his idea of growing continuity. It is therefore no longer necessary to interpret Peirce's statement, "My whole method will be found to be in profound contrast to that of Hegel: I reject his philosophy in toto," as an expression of naturalism, and his statement, "My philosophy resuscitates Hegel, though in a strange costume," as a profession of faith in transcendentalism. For the "strange costume" means only the disguise adopted by a philosopher who, while believing in growth and continuity, firmly assumed that these conditions are to be tested in a laboratory and not in a seminary.

Secondly, Peirce never admits that the building blocks of a philosophic architecture occupy their fixed and never changing places in the whole. In a similar way in which a sign which in a given logical system appears as primitive can appear in another system as defined, any of the conceptions out of which a philosophic architecture is composed can appear either as primitive or as defined. If mathematical axioms themselves have been discredited as axioms, namely, as self-evident truths, and have turned out to be postulates or conditions for the development of hypothetico-deductive systems, there is no reason why the so-called metaphysical axioms must remain unquestioned. But to assert that there are no metaphysical axioms is tantamount to assuming that the building blocks of philosophic architecture are interchangeable. Thus Peirce's speculative architectonic may very well fit into a laboratory philosophy and therefore remain faithful to that "first rule of reason": not to block the road of inquiry.

The expression 'related views' in the title of this paper appears in the plural. It should be expected therefore that some other conceptions of architectonic besides Kant's ought at least to be mentioned. I shall proceed to do so in a most sketchy way.

To the best of my knowledge, only six philosophers besides Kant and Peirce have introduced the word 'architectonic' into their philosophies: Aristotle, Leibniz, Lambert, Wronski, Warrain and Bronstein. Wronski, Warrain and Bronstein are irrelevant for our purpose, not only because their conceptions of architectonic could not have been known by Peirce, but also because they are in many senses of an entirely different character. Aristotle's conception is too general to be included as a precedent. As for Leibniz, his use of the word 'architectonic' is tied up with his doctrine of final causes. Although I would not be in the least surprised to find that Kant drew his idea from Leibnizian sources, he obviously worked it out in such a way as to make it almost unrecognizable to a faithful Leibnizian. At any rate, it would be far-fetched to suppose a direct relationship between Leibniz and Peirce in this respect, for the American philosopher, who declared Leibniz to be a great and singular genius and "one of the minds that grow," did not seem to have realized the Leibnizian precedent. Only Lambert remains as a possible subject of inquiry. An interesting subject, indeed. For Peirce might have found in Lambert a doctrine of architectonic that bears some resemblance to his own efforts toward combining bold speculation with strong empiricism. Lambert's architectonic is certainly comparable to Leibniz-Wolffian ontology. But whereas this ontology was concerned with the realm of thought as a realm of mere possibilities, Lambert continued in his two-volume work on Architectonic the same efforts initiated in his Neues Organon and tried very hard to develop the "the doctrine of truth," the "doctrine of reality," and "the doctrine of thought" in such a way as to present a blueprint for all possible sciences. How to relate logical concepts with metaphysical concepts, mathematical truths with factual truths, analysis with experience and observation, were the great aims of this much too neglected philosopher. Needless to say, he indulged in all kinds of unwarranted assumptions, so that his approach turned out to be far too speculative for the taste of any, however moderate, empiricist. Yet, we can discover in his analyses a serious attempt to find out a way of organizing the system of the sciences without necessarily having recourse, as Kant had, to transcendental legislation and without losing sight of the fact that no metaphysical truth can subsist unless filled up, so to speak, with the stuff of experience and scientific observation. Now, if Peirce had a consistent ideal throughout his life, it was certainly the one that Lambert strenuously maintained. I do not for a moment suggest that there is any direct influence of Lambert on Peirce in the above respect. Although Peirce—whose historico-philosophical erudition was truly overwhelming—was familiar with some of Lambert's doctrines and even referred on one occasion to Lambert's Arcitektonik, we have no reason to suppose that he wanted to go beyond the specific issues that prompted him to bring in the name of the Alsatian thinker: the nature of the so-called geometrical axioms, or the use of Eulerian diagrams. However, if we speak of related views and not of direct influences, Lambert's conception of architectonic appears to me as more related to Peirce's than any of the others, not excluding the Kantian. It would not be entirely useless for an understanding of Peirce's thought to carry this merely hinted at analogy a little further.

* The main texts on which the present investigation is based are: C.S. Peirce, Collected Papers, 1.1, 1.2, 1.3, 1.7, 1.13, 1.14, 1.40, 1.41, 1.42, 1.44, 1.126, 1.127, 1.129, 1.130, 1.135, 1.141, 1.176, 1.177, 1.178, 1.179, 1.232, 1.234, 1.368, 4.27, 4.28, 4.29, 4.353, 5.343, 5.382 (note), 5.392 (plus note), 6.7, 6.8, 6.9; I. Kant, Kritik der reinen Vernunft, A-832-A 851, B 860-B 879; G. W. Leibniz, "Tentam Anagogicum," Werke, ed. Gerhardt, VII, 273; Aristotle, Eth. Nic., I 1 1094 a 14, 25, VI 8 1141 b 22, 25, VII 11 1152 b2; B. Bornstein, apud Studia Philosophica, I (1935); 446; J. M. H. Wronski, Architechtonique de l'Univers, II, 1936; F. Warrain, L'Armature métaphysique, 1925; J. H. Lambert, Anlage zur Architektonik oder Theorie des Einfachen und Ersten in der philosophischen und mathematischen Erkenntnis, 2 vols., 1771.—The writings on Peirce referred to are: J. Feibleman, An Introduction to Peirce's Philosophy interpreted as a System, 1946; Th. A. Goudge, The Thought of C. S. Peirce, 1950; P. Weiss, "The Essence of Peirce's System," The Journal of Philosophy, XXXVII (1940), 253-264; J. Buchler, Charles Peirce's Empiricism, 1939; the same, "The Accidents of Peirce's System," The Journal of Philosophy, XXXVII (1940), 264-269.