Appendix III

Further Reflections on KSP1

 

 

1. Three New Maps of Kant’s System

          This Appendix consists of four sections. In this first section, I suggest three new maps that can be used to clarify the complex relationships between the three standpoints and the four per­spectives in Kant’s System. In the next section, I respond to several scholars who have criticized or pointed out weaknesses of various sorts in KSP1. The third section makes up for a serious omission in KSP1 by providing a textual analysis of Kant’s use of the term ‘architectonic’. And AIII.4 supplies some minor corrections I shall make to KSP1, if/when a second printing or revised edition is published.

          A question that might arise about my account of Kant’s System of Perspectives is: precisely how are the standpoints related to the perspectives, and why are there three of the former, yet four of the latter? Is this merely an arbitrary device, or is it grounded in the logical structure of reason’s architec­tonic unity? There is indeed such a logical ground. The perspectives are, as it were, second order applications of the standpoints to themselves. In other words, the perspectives arise as a direct result of the connections between a given standpoint and each of the three standpoints (including itself). There are exactly four possible connections in each case, as follows:

 

          1. A standpoint’s relation to itself (= the transcendental perspective).

          2. A standpoint’s relation to opposing standpoint A (= the logical perspective).

          3. A standpoint’s relation to opposing standpoint B (= the empirical perspective).

4. A standpoint’s relation to the relation between standpoints A and B (= the hypo­thetical perspective).

 

          These complex inter-relations can be mapped quite simply and precisely onto three inter­secting circles, arranged in such a way that they describe eight discrete spaces, as shown in Figure AIII.1. The three questions that give rise to the 3LAR are specified to the right of the figure. By using each circle to repre­sent one standpoint (with the --- com­ponent representing transcendent reality, which does not play any constitutive role in the System), the rela­tion­ship be­tween the stand­points and the perspectives comes into full view. The unshaded por­tion of each map in Figure AIII.2 represents a given stand­point. When the term that represents the stand­point itself (as defined by the three questions given in Figure AIII.1) is put in brackets, the four remaining two-term components in the circle constitute a 2LAR. The arrows show how the system generated by each standpoint starts with transcendent reality as un­knowable (and so, empty [---]), and proceeds by four

 

 

Figure AIII.1: Three Intersecting Circles as a 3LAR[1]

 

Figure AIII.2: The Standpoint-Perspective Relation

 

successive stages to construct a rationally accept­able picture of reality as we can know it (in its fullness [+++]).[2] Thus, the four arrows in the first two figures—(a) and (b) in Figure AIII.2, corresponding to the first two Critiques—depict the four transformations that take place when we move clockwise around the perimeter, starting from the top of a standard 2LAR cross

(usually labeled ++, but at the outset of a twelve-step system, assumed to be empty—see the (0) at the bottom of Figures III.6-7).

          That the third standpoint does not access transcendent reality directly, but only through the mediation of the other two standpoints, corresponds nicely with Kant’s division of the third Critique into two discrete parts. The first part explores the judicial standpoint’s access to the transcendent via the practical stand­point, while the second explores its access via the theoretical standpoint. The third diagram [Fig. AIII.2(c)] accurately depicts this twofold entry by show­ing the two ways of accessing transcendent reality (---) from the third circle. The reason for this is that sys­temj has no direct access to the thing in itself, but can approach it only through one of the other two sys­tems, and that in so doing it serves as a

Figure AIII.3: An Architectonic Justification for Kt7’s Two Parts[3]

synthe­sizing bridge between systemt and sys­temp. Figure AIII.3 further depicts the closer association Part I of Kt7 has with Kt4 and Part II with Kt1. This new map nicely illus­trates how the need for two parts in Kt7 is rooted in the essential architec­tonic re­lationship be­tween a 1SLR and a 2LAR, with the 2LAR’s impure components aris­ing out of the 1SLR’s x term by extending the arrows beyond the x. The first-level distinction assumed here is between nature-based (-) and freedom-based (+) stand­points, while the second level contrasts subjective (-) and objective (+) forms of finality.

 

2. Responses and Rejoinders to Critics of KSP1

Anthony Perovich

          Long before KSP1 appeared in print (and before splitting the work into four volumes), I wrote a brief paper giving a synopsis of my intent to put forward a new interpretation of Kant. Three years after this relatively brief article [Pa89] was published, the same journal published a response by Anthony Perovich [Pe92]. The response was largely skeptical, rais­ing concerns about the lack of support provided for the claims I put forward, and about ambiguities in my account of Kant’s various ‘perspectives’. In a rejoinder [Pa94a] I explained that the article was intended as a ‘manifesto’, not as the actual demonstration of the positions sketched. Interestingly, the alternative line of interpretation suggested by Perovich turns out to be remarkably close to the position defended here in KSP2. Since these articles are pub­lished elsewhere and are now somewhat outdated, I shall not discuss them any further here.

 

Jennifer McRobert

          In a generally accurate review of KSP1, published in Canadian Philosophical Reviews [Mc94], Jennifer McRobert raises two criticisms that merit a brief re­sponse. First, she says KSP1 ‘reduces the roles of such faculties as imagina­tion and sensibility in Kant’s epistemol­ogy.’ This simply is not true: the proper role of sensibility as the governing faculty for the entire first stage of Kant’s theoretical system is clearly emphasized in KSP1:VII.1-2.A [see e.g., Fig. VII.3]; and my strategy of not assigning imagination to any one stage, but re­gard­ing it as a ‘floating faculty’ [KSP1:60n] is not meant to diminish its impor­tance, but to highlight it! On the latter point, McRobert later claims I omit imagination altogether from my ‘framework of interpretation’. Although the footnote she cites [60n] could give this mis­taken impression, the fact is that imagination’s crucial role of linking each of the four stages is explicitly ac­knowledged at numerous points in Part Three [see e.g., 213,223-4,239-41].

          The one point McRobert makes on this issue that is fully justified and worth pursuing (in a revised edition of KSP1) is that the imagination’s role in ‘the Typic in the second Critique’ could have been brought out more fully in Chapter VIII. For there is indeed a suitable place for the imagination in step seven of systemp. In the meantime, let it suffice to say that I recognize the imagination as a crucial part of Kant’s System, but did not see it as a focal point of my own study. I would hope the interpretation developed in KSP1 can serve as a frame­work for those who wish to explore this aspect of Kant’s philosophy in greater depth. (The interested reader would do well to consult Ma90c along these lines.)

          McRobert’s second criticism, also mentioned twice, is more significant. She casts a dark shadow over ‘[t]he grade of textual support’ in KSP1 by citing as ‘typical’ a quotation she claims is taken out of context. In KSP1:160 I quote a passage from Kt43:222(82), where Kant says ‘faith ... provides ... a clear light to enlighten philosophy itself’; whereas I treat this as Kant’s own view, McRobert protests that it should be read as ‘part of a hypothetical objector’s complaint’ against Kant’s position [Mc94:120]. While it is true that the statement in question appears at the close of a paragraph devoted to a possible objection, there is nothing in that passage indicating Kant would reject the quoted claim. On the contrary, the very force of the objection depends on the assumption that Kant would want to affirm this and the other views mentioned in that paragraph. These other views—viz., that God exists as a world-governor, that natural beauty is a good way of demonstrating a divine Author of the world, and that Christian morality is superior to Epicureanism—all refer to positions Kant explicitly defends elsewhere; so the most natural interpretation of the fourth claim (the one I quoted) would be that Kant also wishes to affirm it in spite of the objection. McRobert’s attempt to cite an example of my alleged tendency to quote out of context therefore fails.

          Near the end of her review, McRobert again laments ‘that Palmquist so often fails to deem it necessary to back controversial interpretations with sub­stan­tial arguments and ade­quate textual support’, adding that it is unclear ‘whether his book is intended as an exposition of or as a redeployment of the critical philosophy.’[4] The answer, quite simply, is both! The best ‘interpreta­tion’, I believe, consists of both exegesis and eisegesis. The two factors are complementary, not mutually exclusive: exegesis without eisegesis is like a dead corpse; eisegesis without exegesis is like a disembodied spirit. With this in mind, a fair-minded reader cannot avoid recognizing that many points of my interpretation are well-grounded in clear and straightforward textual references. Herein lies its weight of conviction—a weight that cannot be ig­nored—though being admitted­ly imperfect, I may have inadvertently taken some quotes out of context. At the same time, a mere exposition of Kant’s theory inevitably contains gaps that can be filled only with a reconstruction. My attempts to reconstruct are so thorough­going (applying as they do to Kant’s entire System and lifelong philosophical de­velopment) that they are bound to be sketchy on points of minute detail. Such points, when raised at all, are intended more to be thought-provoking than to convince all readers by their logical rigor. They are not intended to have the force of incontrovertible proof, but to lift the reader to the heights from which a vision of Kant’s overall architectonic idea can be clearly seen.

 

Susan F. Krantz

          After giving a brief but accurate synopsis of KSP1, Susan Krantz’s book review in The Review of Metaphysics [Kr94] raises some ‘problems’ about my use of the term ‘perspective’ and its relation to ‘the system of graphic models’ [420]. She argues that, although I ‘may be right in some other cases’, my claim that various forms of der Gebrauch can be regarded as perspectival equivalents [see KSP1:II.3.C] is misleading: it transforms the ‘dynamic connota­tions’ of a faculty’s use into a ‘static image’. While admitting that the strategy of represent­ing perspectival relationships in terms of ‘graphic models’ lends clarity and consistency to the interpretation, she asks, in a rhetorical tone: ‘Should not a translator or interpreter be wary when his or her version begins to look much “clearer” than the original?’

          In response to this charge let me first point out that my inclusion of der Gebrauch in the list of perspectival equivalents discussed in KSP1:II.3 was not indiscriminate. Variants of this term occur 427 times in Kt1. Before compiling Table II.3, I examined each of these occurrences individually, in both German and English, selecting only 256 occurrences (60%) for inclusion in the list of perspectival equivalents. The primary determining factor was context. Without a doubt, Krantz is right that der Gebrauch has dynamic connotations; indeed, sometimes these connotations do not in any way imply the presence of a per­spective. My claim was that the dynamic connotations of the specific passages I selected do imply (or ‘serve as a signpost, indicating’ [KSP1:50]) the presence of an identifiable perspective. My decision to translate the selected passages with the term ‘perspective’ was defended by examining four typical examples [51-3], since it was not feasible to present my rationale for all 256 cases. In so doing, I did not mean to suggest that ‘perspective’ is the most accurate transla­tion (though for some of the other perspectival equivalents, I believe it is), but only that using it in my quotations will help to highlight and clarify certain implica­tions that really are there in the text. In other words, my expressed purpose was to make explicit what often remains implicit in Kant’s text.[5]

          The doubts raised by Krantz’s rhetorical question are of a deep, herme­neu­tic nature. Clearly, a translator ought to be ‘wary’ of imposing such an inter­pretive bias onto the transla­tion (though to some extent even this may be inevitable). The translator’s basic task is to present the text as accurately as pos­sible to the reader, who must then attempt to make sense out of the ambi­guities and innuendoes. Translators differ, however, in how far they are will­ing to go in ‘clearing up’ such ambiguities by translating the text more clearly than it was written. In any case, my reason for pointing out various perspecti­val equiva­lents in KSP1:II.3 was not to argue that previous translators have been wrong, but to suggest that certain (admittedly loose) translations can be used in the service of my interpretation. (Note that the word ‘interpretation’ in the subtitle of KSP1 is not without significance.) In other words, the perspecti­val termi­nology should be regarded in most cases as a hermeneutic device, not as a claim about the literal meaning of certain German words. When this is un­der­stood, the ‘wariness’ should fall away. For what is the task of an inter­preter, if not to understand the original author better that he understood him­self—just as Kant himself claimed with respect to Plato [see Kt1:370]?

          Interpreters should be wary when their attempted clarifications produce a conclusion that diverges from the original intentions or underlying ‘spirit’ of the author they are interpreting. This is a factor I always try to keep in mind. Thus, KSP1:I.3 specifically argues that Kant had a discernible preference for metaphorical thinking in general and geometrical metaphors in particular. This, taken together with Kant’s undeniable love of architectonic, provides ample evidence that he would not have objected to the use of geometrical models in clarifying the perspectival relationships in his System.

 

Jack Chun Ka-Cheong

          In his 1995 doctoral thesis at University of Toronto, entitled Kant’s Objective Explana­tion of Our Objective Knowledge (especially Chapter 2, ‘The Transcendental Turn’), Jack Chun takes issue with several points in my in­ter­preta­tion. His comments indicate that in KSP1:V and elsewhere, I used ‘transcendent starting-point’ in a potentially misleading way. Chun takes my view to be that the thing in itself serves somehow as a ‘transcendent object’ in systemt, akin to the false (positively noumenal) way of viewing the ideas in the Dialectic. But by ‘transcendent starting-point’, I mean something very different from ‘transcendent ob­ject’. The former is rather a way of designating the need for some meta-Critical point of refer­ence, from which the entire Copernican Perspective can be viewed and in some sense justified.

          This has important implications for the proper understanding of my view of the role of faith in Kant’s System. My claim in KSP1:V that faith has a theo­retical as well as a practical application for Kant can indeed give the impression that faith actually operates within systemt. But as the foregoing paragraph should have made clear, this impression was unintentional. For my actual position is that faith serves as a justification for transcendental reflection at the outset of the entire System. It is, as it were, the key that opens the door to the ‘house’ of Critical philosophy. Once we are inside, the key serves no purpose, except of course, when we think about leaving the house for one reason or another—just as Kant does in the Dialectic of each Critique, where faith, or some form of hypothetical reasoning akin to faith, comes back into the picture.

          Chun does raise a very important point with respect to the limitations of my claim. He points out that the thing in itself is most definitely not the starting-point of any of Kant’s arguments. The refutation of the skeptic in particular must start from what the skeptic admits, and this must be our experience of phenomena. So Kant always begins with the appearance. If my claim about the thing in itself is regarded as implying that it serves as the logical start­ing-point of his actual arguments, then such a claim will appear obviously flawed. But my actual intent is only to argue that the thing in itself is the architectonic starting-point of Kant’s System. (On the relative importance of systemization versus argumentation, see KSP1:I.1).

 

Christopher L. Firestone

          As a graduate student who also served as my teaching assistant during the 1995-96 aca­demic year, Chris Firestone worked in close contact with me, developing his interpretation of Kant and applying it to various aspects of Kant’s philosophy. In his subsequent publications and doctoral research, he has been developing his own version of a perspectival way of inter­preting Kant. Occasionally this has led him to take issue with specific points in my interpretation. Originally I intended to include a response to his criticisms and revisionary suggestions in this Appendix. However, due to limitations of space—the publisher thinks this book is already too long (and is probably right!)—and in order to give Firestone a fair chance to develop his views before I criticize them, I have decided not to offer a response at this point. Moreover, most (if not all) of the work in question relates to Kant’s theology and philosophy of religion, whereas my focus here is on criticisms of KSP1.

 

Paula Manchester (née Hunter)

           As my work on the present volume was approaching completion, Paula Manchester con­tacted me by email to inform me of her interest in Kant’s conception of philosophy as ‘architectonic’. After an interesting series of initial exchanges, we began to read each other’s work. Her doctoral dissertation, Architectonic and Critical Philosophy [Hu81], is a well-researched examination of the historical use of the term ‘architectonic’, from the classical period to Kant, within Kant’s own writings, and from Kant’s time to the first half of this century. Along the way, she offers some suggestions as to what she thinks is central to Kant’s own view of architecton­ic. Had I come across this valuable work while I was writing KSP1, I would surely have done a better job of defending my own interpretation of Kant’s position. In our subsequent correspondence, Manchester has convinced me that KSP1 sorely lacks a clear account of the passages where Kant explains what he means by ‘architectonic’. I have there­fore devoted the next section of this Appendix to that task. For now I shall briefly explain the position defended in Hu81 and how it calls into question certain aspects of my ap­proach. Although it obviously does not contain any direct criticisms of KSP1, Manch­ester (rightly) regards her treatment of the subject as superior to and implicitly critical of mine.[6]

          Manchester’s thesis begins with a chapter introducing Kant’s concep­tion of architec­ton­ic, including a fascinating review of how Kant-scholars have interpreted it from Kant’s day down to the present. After pointing out that the term is typically regarded as part of an architectural metaphor, she laments:

 

The reduction of the term Architektonik to a vague metaphor has been so predomi­nant that the specifically philosophical meanings of the term as it appears in the history of philosophy ... go unmentioned in discussions about Kant’s presumed use of it. Hence, the philosophical issues that motivated Kant to write a chapter in the Critique called ‘The Architectonic of Pure Reason’ have never been brought to light. [Hu81:3]

 

Her aim in Hu81 is to fill this gap by examining these historical roots and by focusing primary attention on the Architectonic chapter in Kt1. On the latter point, she repeatedly emphasizes that Kant ‘thinks of it as an “art” similar to that of teaching’ [3]—especially ‘the teaching of the scientific’[7]—and as a form of ‘commanding’ [4]. After examining Kant’s use of these and related phrases, Chapter I reviews the interpretations of architectonic advanced by G.A. Will, Arthur Schopenhauer, Benno Erdmann, Erich Adickes, and Norman Kemp Smith.

          After tracing in Chapter II the development of Kant’s conception during various periods of his life, Hu81’s third chapter examines the history of the term ‘architecton­ic’. Manchester carefully examines how Plato, Aristotle, Aquinas, Leibniz, Wolff, and Lambert used the term. A surprising range of possible meanings is revealed, some very close to the conventional interpretation adopted in KSP1 (that ‘Kant believed all architectonic patterns to be based on the structure of formal logic’ [KSP1:359]) and others very close to Manchester’s preferred interpretation. She provides no direct evidence that Kant was influenced more by any of these philosophers than by the others, though she does re­peatedly assume that the ones closer to the position she prefers were the ones who influenced him. Hu81 concludes in Chapter IV with some reflections on how Kant’s architectonic is founded on Critical principles, but admits in the end [187] that no attempt has been made ‘to explain the critical doctrines founding Kant’s conception in terms of their agreement or disagreement with the history of “architectonic.”’

          In our extensive email correspondence Manchester made it abundantly clear to me that she regards my position as hermeneutically unsound, a virtual figment of my own imagina­tion that could not be defended from Kant’s actual usage. Her weightiest criticism, that I totally ignored the background influences on Kant’s thinking, is undoubtedly correct. In writing KSP1, as well as the present volume, I made a conscious decision to eliminate certain types of work from my research, and historical background studies have as a result been low on my list of priorities (with the main exception being the influence of Swedenborg, discussed in II.2-4 and AII.2-4). However, this alone does not necessarily condemn my interpre­ta­tion to the graveyard of total subjectivity. On the contrary, the task of deciding which predecessors’ views were the ones that influenced Kant is, in the absence of any explicit word from Kant on this matter, itself a matter of interpretation that is inevitably subjective to some extent. My response to Manchester’s valuable historical survey is that it opens up the new possibility of a debate concerning Kant’s influences, but does not in itself provide any conclusive (non question-begging) answer.

          Manchester’s other main objections to my interpretation, that it sees Kant’s architectonic as bound up with reason’s need to impose logical patterns onto the empirical data and that it treats ‘architectonic’ as part of an architectural metaphor, are less weighty. On the former point, her main evidence for reject­ing such an approach comes not from Kant’s use of ‘architectonic’, but from her belief that this term is connected for Kant to a distinction he makes between the ‘school concept’ of philosophy and the ‘world concept’ [see Kt1:867]. She points out that for Kant ‘the Schulbegriff of philosophy is satisfied with a sys­tematic unity which is merely logical, while the Weltbegriff seeks a system which teleologically relates those cognitions properly called philosoph­i­cal to the essential ends of human reason’ [Hu81:15]. Later she goes so far as to say that the phrase ‘logical architectonic’ is ‘a contra­diction in terms’ because of its ne­glect of the Weltbegriff-Schulbegriff distinction [29; s.a. 94-5]. What she fails to recognize is that Kant does not say the ‘world concept’ repudiates the for­mer’s logical unity; rather, his point is that mere logical unity must be raised to the status of a higher, teleological unity by imposing on it ‘the idea of the whole’. Only then does phi­losophy deserve to be called ‘architectonic’. Kant’s point, in other words, is not that architec­tonic unity is not rooted in logical pat­terns, but rather that it raises these patterns to a new level by inter­preting them as manifestations of a single unifying idea. ‘Logical architectonic’ is therefore not a contradiction in terms, though ‘merely logical architectonic’ would be.

          On the second issue, Manchester never clearly explains what it is about being an archi­tect that is so disagreeable. It is apparently nothing more than a bias she holds against the text. For many of the features she readily admits to being essential to the art of architectonic —features such as ‘“leading,” “conducting,” “directing,” and “lawgiving”’ [Hu81:86]—are, in fact, tasks that could be regarded as describing an actual architect’s function in relation to the builder(s). Yet she neglects such similarities, without providing any supporting rationale.

          This bias is particularly noteworthy in Hu81’s summary of the view advanced by Will, the first interpreter of Kant’s architectonic. Will thought ‘that Kant conceived of Architek­tonik as a kind of “philosophical architecture,” and ... that the “system” executed in the transcendental doctrine of elements represented “Kant’s architectonic”’ [37]. Manchester suggests that, if these claims had been challenged, this ‘would have gone a long way to clarifying Kant’s conception of architectonic.’ But this is an extremely weak argument. Will’s interpretation, published in 1788, would surely not have gone unchallenged during the last 16 years of Kant’s life, had it been as disastrously misconceived as Manchester claims it was. Why did Kant remain silent? Perhaps it was because Will had properly understood his intended meaning![8] After all, he is the only interpreter Manchester acknowledges who ever praised Kant for his architectonic, treating it ‘as the instrument of Kant’s success’ [39]. That Kant may have approved of Will’s interpretation is a highly significant possibility, because Will’s interpretation prefigures my own in some remarkable ways: ‘For Will the “system” of the “fine and artful philosophical Architekt” consisted of the frame of “fourfolds and threefolds” found in the table of categories, the schemata, the classification of the basic propositions, and finally, the metaphysical treatment of natural science’ [38]. Unfortunately, instead of taking Will’s position as an opportunity to have a closer look at Kant’s text, Manchester responds dismissively [38-9]: ‘It is inexplicable ... how Will can claim that this represents the execution of the plan described in the Architectonic since none of this is even mentioned there by Kant.’ In the next section, I shall demonstrate that this claim is incorrect.

          Because her alternative account of Kant’s architectonic consists mostly of a few key phrases (especially ‘art of systems’ and ‘teaching of the scientific’) repeated over and over, with no clear and detailed elaboration of what Kant actually means, Manchester’s conclusions are highly questionable. She claims, for instance, that Kt1’s infamous ‘problems in presentation are not rooted in Kant’s Architektonik. Quite to the contrary, Kant is rarely—if ever—“architectonical” at all’ [Hu81:40]! She defends this sur­pris­ing claim on the grounds that a connection between the Doctrine of Elements and the Architectonic chapter ‘is neither obvious nor discoverable on the surface of the textual arrangement of the doctrines presented’ [41]. And ‘in any case’, she in­sists [42], ‘the organizational structure that Kant describes in the Architectonic as indicative of the “art of systems” in no way resembles ... an architectural scaffold, or any building that could be built up out of materials no matter how firm the foundation or ingenious the plan of its architect.’ That these claims are unjustified will be shown in AIII.3, where I shall demon­strate how the text itself supports an interpretation that bases Kant’s architectonic on the 4x3=12 pattern of the categories. Manchester acknowl­edges that Kant’s language in the Meta­physical Deduction suggests that he may be thinking there of his architectonic; but the prob­lem, she claims (falling back now on the textbook account of Kant) [43], ‘is that the idea of the whole which led to the discovery of the twelve [logical functions] has never really been satisfactorily uncovered by anyone other than Kant.’

          The solution to such a problem is not to throw up one’s hands and aban­don all hope of finding the architectonic unity Kant held so dear; in so doing, Manchester inadvertently joins forces with Kemp Smith and all the other commentators whose views she so forcefully criticizes.[9] The solution is rather to search harder and more carefully than Kant himself (being the original pioneer) was able and to lay bare the structure of reason’s architectonic form in all its detail. This is what I have attempted to do in KSP1, especially Chapter III, at least as it re­lates to Kant’s System. I plan to provide a fuller explanation of reason’s architectonic form, independently of its application to Kant’s System, in a book entitled The Geometry of Logic.[10]

          Manchester’s hermeneutic relies so heavily on external evidence (i.e., on her recon­struc­tion of historical uses of the term ‘archi­tectonic’) that the meaning of Kant’s text is some­times grossly ignored. She tends to impose assump­tions on the text that are supported by little or no inter­nal evidence. The best example is her insistence that Kant was fundamen­tally opposed to any conception of architectonic as related to the tasks of building or architec­ture. That this was not the heart of its classical meaning in the writings of Plato and Aristotle, and that it only later came to be regarded in this way, she indisputably demonstrates in Chapter III. However, in Chapter II she totally ignores masses of evidence in Kant’s texts (some even quoted in close proximity to her denials of their existence [see e.g., Hu81:72-5,93-4]) wherein he uses language that is obviously related to the standard architectural metaphor.[11] At the same time she adopts the questionable assumption that analyzing Kant’s use of the term ‘world’ is the best way to discern the development of his view of architec­tonic. This is due to her conviction that Kant has identified architectonic with the ‘world con­cept’ of philosophy in the Architectonic chapter of Kt1. While it is true that he mentions this distinction in that Chap­ter, he does not do so in any of the paragraphs where the term ‘architec­tonic’ actually occurs. Manchester’s preoccupa­tion with the implica­tions of Kant’s ‘world-concept’ of philosophy, while intrinsically interesting and helpful in understanding Kant’s general approach to philosophy, have led her on a ‘wild-goose chase’ that takes her far from anything Kant himself explicitly treats as essential to architectonic.

          When her discussion takes such side-tracks, Manchester does have some interesting things to say about Kant’s development. But she never comes close to justifying her allergy to all talk of building or architecture. Kant’s complaint against past philosophers is not that they wrongly regarded philosophy as a kind of ‘building’, nor that they naively compared the philosopher to an ‘architect’; rather, his reproach is that those who built systems did so with­out a plan that would guarantee unity and completeness, and that instead they acted like archi­tects who design merely according to whim, rather than according to reason’s essential ends.[12]

          In making such dogmatic denials of the architectural aspects of Kant’s conception of architectonic, Manchester does herself a disservice. Her approach is highly instructive in demonstrating that this is not all Kant intended by the term, that it goes far deeper and must encompass (for example) a philosopher’s effort to direct his or her thinking (or that of the reader) towards reason’s essential ends. What she fails to grasp is that the architectural meta­phor is meant to assist in achieving that goal, not detract from it. By never explaining why the architectural anal­ogy is so dangerous, she cuts herself off from the riches of its possible applications; likewise, by never explaining how architectonic can survive without logic, she puts herself in a position of being unable to explain how architectonic could ever reveal to us an idea of the whole that helps reason to realize its own ends.

 

3. Textual Review of Kant’s Conception of Architectonic

          In this section I shall review how Kant uses the term ‘architectonic(ally)’ in his primary systematic works (i.e., Kt1-Kt10), paying special attention to any evidence that Kant may have intended it to have a logical or architectural character. The term appears in Kt1 only 14 times, nine of these being in Chapter III of the Doctrine of Method, entitled ‘The Architec­tonic of Pure Reason’. Before looking at this key chapter, let us examine each of the five references that occur earlier in the book, one in the Introduction and four in the Doctrine of Ele­ments. These will provide indispensable clues for understanding Kant’s subsequent, more sys­tematic treatment of the topic.

          The first occurrence appears in the last numbered section of Kt1’s Introduction, entitled ‘The Idea and Division of a Special Science, under the Title “Critique of Pure Reason”’ [24], where Kant explains the technical mean­ing of the term ‘transcendental’ for the first time [25]: ‘I entitle transcendental all knowledge which is occupied not so much with objects as with the mode of our knowledge of objects in so far as this mode of knowledge is to be possible a priori.’ Transcendental philosophy, as the systematic exposition of such concepts in the form of a ‘doctrine’ or ‘organon’ [25-6], cannot be accomplished until preliminary groundwork is performed by a critique. The purpose of the latter ‘is not to extend knowledge, but only to correct it’ [26]. In this context the second paragraph of this section begins as follows:

 

Transcendental philosophy is only the idea of a science, for which the critique of pure reason has to lay down the complete architectonic plan. That is to say, it has to guarantee, as following from principles, the completeness and certainty of the structure in all its parts. It is the system of all principles of pure reason. [27]

 

Kant goes on to explain that the com­pleteness of the plan is to be derived from the Critique’s ‘enumeration of all the fundamental concepts that go to constitute ... pure knowledge’ [27]. That he is referring here to the table of ‘the logical function of the understanding in judgments’ [95, e.a.], and to the table of categories derived therefrom, is quite clear from the context. About no other part of his philosophy does Kant emphasize its completeness so adamantly; so it is not without significance that Kant uses ‘complete’ and its cognates no less than fourteen times in this section of the Introduction.[13]

          This first passage says nothing explicit about what Kant means by the term ‘architec­tonic’. Kant’s grammar alone tells us, however, that ‘architectonic’ is a type of ‘plan’ and that this type of plan can be made ‘complete’. Furthermore, we learn that laying down such a plan is Kt1’s central task and that this plan will guarantee that, once transcendental philosophy is fully elaborated, its structure will be both complete and certain. Kant’s use of words like ‘plan’ and ‘structure’ would seem, at first sight, to support the view that ‘architectonic’ is being used as part of a metaphor wherein the philosopher’s work is being compared to that of an architect. The difference, as is clear from the context, is that the philosopher is designing a plan not for a physical structure but for a logically structured set of a priori concepts, complete and certain in their relation to each other and to all human knowl­edge.

          On its own, this initial reference would not be enough to build an interpretation on; but in conjunction with the next four occurrences of ‘architec­tonic’ and with Kant’s many other references to the terms ‘plan’ and ‘structure’, the tentative suggestions made above can be con­firmed. The second occurrence appears in the introductory section of the Dialectic, entitled ‘The Ideas in General’. The section’s main purpose is to distinguish Kant’s use of the term ‘idea’ from Plato’s. Near the climax of this discussion [Kt1:375] we read: ‘If we set aside the exaggerations in Plato’s methods of expression, the philoso­pher’s spiritual flight from the ectypal mode of reflecting upon the physical world-order to the architectonic ordering of it according to ends, that is, according to ideas, is an enterprise which calls for respect and imita­tion.’ Here ‘architectonic’ is regarded as a mode of ordering ‘according to ends’ or ‘ideas’ and is directly contrasted with ‘ectypal’. Earlier in the same paragraph [374] Kant uses the cognate term, ‘archetype’, to refer to Plato’s view that the soul contains divinely predetermined ideas that ‘are the original causes of things.’ So we now know that Kant regards ‘architectonic’ as a teleological method of using ideas to impose order onto the physical world and that he has at least a qualified sympathy with Plato’s employment of it.

          The third and fourth occurrences of ‘architectonic’ appear together in a paragraph towards the end of Section 3 of ‘The Antinomy of Pure Reason’. The relevant portions of the paragraph (omitting the comments relating solely to the antinomies) are as follows:

 

Human reason is by nature architectonic. That is to say, it regards all our knowl­edge as belonging to a possible system, and therefore allows only such principles as do not at any rate make it impossible for any knowledge that we may attain to combine into a system with other knowledge.... Since ... the antithesis thus refuses to admit as first or as a beginning anything that could serve as a foundation for building, a complete edifice of knowledge is, on such assumptions, altogether impossible. Thus the architectonic interest of reason—the demand not for empirical but for pure a priori unity of reason—forms a natural recommendation for the assertions of the thesis. [Kt1:502-3]

 

Here we find a number of new claims about ‘architectonic’, as well as some reaffirmations of hints mentioned earlier. First, architectonic is not merely a fabrication of some philosopher’s imagination; it is part of the very nature of reason itself. Second, to say something is architectonic is to say it is systemati­cal­ly unified or unifiable. Third, once again we see Kant using language that suggests he has in mind an architectural metaphor: reason’s architectonic system is compared to a ‘building’ that has a ‘foundation’ and stands as a ‘complete edifice’. Finally, and most significantly, the phrase set off by dashes clarifies that ‘architectonic’ provides an a priori unity, not a unity that has been derived from empirical sources. Since Kant regards analytic a priori as the status of formal logic and synthetic a priori as the status of transcendental logic, we can now conclude with virtual certainty that Kant thinks architectonic performs its unifying function by means of logic (in one of its two forms).

          In its fifth occurrence [Kt1:736] ‘architectonic’ merely appears in a list of the four main divisions of the Doctrine of Method. Nevertheless, the context is informative. The sentence immediately preceding the list states that the Doctrine of Method concerns ‘the determination of the formal conditions of a complete system of pure reason’ [735-6]. It examines the form of systemt, whereas the Doctrine of Elements provides the material or content. This tells us that ‘architectonic’ relates in some way to a philosophical system’s formal structure, rather than to the details of its arguments or theories. The sentence immediately following the list notes that a more common term for the four topics in question is ‘practical logic’ [736]. The remainder of the paragraph calls into question the appropriateness of the term ‘practical’, but not the term ‘logic’. Kant’s criticism of the term boils down to the fact that the ‘Schools’ approach neglected the all-important considerations of transcendental logic.

          With these preliminary reflections in mind, let us turn now to our main source of information about Kant’s meaning, the chapter on ‘The Architectonic of Pure Reason’ [Kt1: 860-79]. We have already seen that Kant regards each chapter in the Doctrine of Method as examining an aspect of the form of the same system whose content is examined in the Doctrine of Elements and that the key feature of that content is the discovery of the four logical functions of the understanding in any act of judgment, functions that give rise to the categories and eventually, when synthesized with intuitions, to the principles of pure understanding. The fact that Kant chooses to discuss exactly four topics in the Doctrine of Method suggests that here, as in most of his fourfold divisions, he is consciously attempting to follow the pattern set by the four basic logical functions. If so, the Discipline, Canon, Architectonic, and History of Pure Reason could be regarded as expressing, respectively, the quantitative, qualita­tive, relational, and modal aspects of transcendental philosophy’s form. At this point, of course, this is purely conjectural; but we should keep in mind the pos­sibility that Chapter III may correspond somehow to the category of relation.

          After the title of Chapter III, the remaining eight occurrences of ‘architec­tonic(ally)’ all appear in four paragraphs. The chapter’s opening paragraph contains the second and third occurrences: ‘By an architectonic I understand the art of [constructing] systems. As systematic unity is what first raises ordinary knowledge to the rank of science, that is, makes a system out of a mere aggregate of knowledge, architectonic is the doctrine of the scientific in our knowledge’.[14] The ambiguity in the first sentence [see note AIII.7] is probably best clarified by translating it as ‘... the art [exhibited by] systems’ or perhaps ‘... the art of systematizing.’ Kemp Smith’s ‘constructing’ has the disadvantage of leading English readers to believe Kant is explicitly appealing to a ‘building’ metaphor. Kant is certainly not denying the applicability of this or any other metaphor, but he is also not explicitly using it. All we can infer from his initial definition is that philosophers who want to be architectonic must be systematic. But the following sentence then clarifies that Kant is referring to the making of a system out of something inherently unsystematic (‘a mere aggregate of knowledge’). Kemp Smith’s ‘con­structing’ therefore does have a basis in the context, though as we shall see, ‘designing’ would be a better metaphor than ‘constructing’ to describe what happens when we impose order onto something previously unordered. The newest and most intriguing aspect suggested by this passage is that Kant calls architectonic an ‘art’, even though it is at the same time, somewhat paradoxical­ly, the formal factor that makes a body of knowledge scientific.

          The second paragraph explains what exhibiting or designing a system means in its architectonic sense (i.e., as one of the four formal aspects of transcendental philosophy). It means that rea­son prescribes laws that unify ‘the manifold modes of knowledge under one idea’ [Kt1:860]. This idea, Kant tells us, is ‘the concept ... of the form of a whole’ that determines both ‘the scope of [reason’s] manifold content’ and ‘the positions which the parts occupy rela­tively to one another.’ This nicely supports our foregoing speculation: the task of architectonic should be to determine the relation between the otherwise unre­lated parts of a transcendental system’s form. (Two sentences later Kant again emphasizes this relational aspect. Apparently, he did have an architectonic reason for placing this chapter third!) In the Doctrine of Elements, we learned that this function is fulfilled by the categories, applied in the schematized form of prin­ciples of pure understanding. So Kant appears to be alluding here to a neces­sary connec­tion between the formal structure of the categories and that of all architectonic reasoning.

          If this interpretation is accurate, then why did Kant not simply come out and state that architectonic uses the table of categories (or its predecessor) to impose systematic patterns onto our thought processes? The reason, I believe, is bound up with Kant’s strategy in dividing the Critiques into Doctrines of Elements and Method. Each time he does this, the two sections are meant to be independent of each other, in the sense that they work towards the same goal, but from opposite perspectives: content first, then form. None of the chapters in the Doctrine of Method appeal directly to the results of the Doctrine of Elements; rather, they each reveal in different ways reason’s need for just the sort of thing the foregoing Doctrine of Elements has provided. To connect architectonic too explicitly in Chapter III with the 4x3=12 pattern determined by the categories would have been to beg the question he was attempting to answer. To name the categories or even their numerical structure would be to focus on the content of architectonic; but as we have seen, Kant’s focus here is on its form.

          The second paragraph of Chapter III also states that the purpose of imposing on the aggregate of our knowledge an idea that clearly relates the parts to each other within a whole is to ‘further the essential ends of reason’ [Kt1:860]. Kant unfortunately does not explain what he means by this phrase. However, the remainder of the paragraph suggests he is thinking here of reason’s ultimate goal, the unification of all knowledge; for he claims this prescriptive function of reason (i.e., reason’s architectonic nature) ‘makes it possible for us to determine from our knowledge of the other parts whether any part be missing, and to prevent any arbitrary addition’, thus guaranteeing the completeness of the system being constructed [860-1]. Kant shows us no method of achieving such lofty aims in the Doctrine of Elements other than by patterning our systematic divisions on the formal structure established by the tables of categories and logical functions. To illustrate this point, he concludes the paragraph by comparing a rational system’s potential to ‘grow from within ..., but not by external addition’ to that of ‘an animal body’ [861]. This metaphor is easily understood as referring to Kant’s conviction that, when designing a categorial table in reference to any set of conceptual relations, we must resist any temptation to add a single new member (e.g., 4+1=5), for this destroys the unity of the conceptual relations under considera­tion. Instead, we must account for any new members by making further internal divisions, just as Kant does when he divides each category into three ‘moments’ (4x3=12).

          The third paragraph contains the next two references to ‘architectonic’. It begins by distinguishing between two ways in which a schema and an idea can be related. When viewed from the empirical perspective, the schema presents the manifold of knowledge to us independently of any unifying idea, whereas from reason’s a priori perspective, the schema ‘originates from an idea’ without ‘wait[ing] for [its ends] to be empirically given’ [Kt1:861]. The latter alone, Kant states, ‘serves as the basis of architectonic unity.’ One of the main differences between these two forms of relation is that when the schema is viewed ‘empiri­cally’, ‘the number [of its ends] cannot be foreseen’. But science requires certainty in its dis­tinctions and so must impose them a priori, ‘in architectonic fashion, in view of the affinity of its parts and of their derivation from a single supreme and inner end’ [862]. This is the best evidence yet that the a priori unity imposed on the aggregate by reason’s architectonic art has to do with the 4x3=12 pattern of the categories. For Kant’s point is precisely that reason’s architectonic form (as revealed in the categories) enables us to do what would be impossible using a merely empirical method: to determine the appropriate number of items composing any given set of concepts. Reason’s ability to discern the pattern in advance is the source of the ‘affinity of [the manifold’s] parts’ in an architectonic system.

          The fourth paragraph warns the reader that, although the founder of every new science bases it on a single idea, the initial attempts to schematize that idea are ‘very seldom adequate’, because ‘this idea lies hidden in reason’ [Kt1:862]. As a result, Kant encourages us to be willing to go beyond the descriptions given by the founders of any new science, whom we often find ‘are groping for an idea which they have never succeeded in making clear to themselves’; our focus should instead be on the idea and its grounding in reason. With this in mind, I believe my articulation of the logical structure of the architectonic form of Kant’s System, as given in KSP1:III.3-4, would meet with Kant’s approval. If Kant is to avoid being hypocritical, he would have to confess that he, too, like the founder of any new science, had only a vague grasp of the ‘idea of the whole’ that brought unity and completeness to his System of transcendental philosophy. My conscious goal in KSP1 was to apply this advice of Kant’s to the task of interpreting the architectonic structure of his own System.

          The next three occurrences of ‘architectonic(ally)’, coming in the fifth paragraph of Chapter III, do not tell us anything fundamentally new about Kant’s understanding of the term. The paragraph begins by lamenting that systems are typically put together initially as aggregates [862-3], and that only after ‘a long period ... does it first become possible for us to discern the idea in a clearer light, and to devise a whole architectonically in accordance with the ends of reason.’ (The fact that Kant makes substantially the same point in Kt1:105-7, with respect to Aristotle’s collection of categories, provides yet further evidence for my claim that the table of categories provides the most important clue to the formal structure of Kant’s architectonic.) After likening the development of systems to that of ‘lowly organisms’ [863], he claims that so much ‘human knowledge’ has now been gathered that ‘an architectonic of all human knowledge ... would not indeed be difficult.’ He then announces that the remainder of the chapter will merely ‘outline the architectonic of all knowledge arising from pure reason’.

          From this point much of Chapter III consists of a series of twofold divisions of reason and/or philosophy, intended to provide the reader with a bird’s eye view of the architectonic divisions in transcendental philosophy. For our purposes here we can skip over the details of Kant’s exposition, not only because the various divisions appear at times to be somewhat incompatible with each other, but also because they are advanced as examples of architectonic divisions, not as further explications of the meaning of the term as such.[15] Instead of recount­ing the details of each division, we can pass on to Kant’s final use of ‘architectonic’ in Kt1. Six paragraphs before the end of Chapter III, immediately after summarizing ‘the whole sys­tem of metaphysics’ in terms of ‘four main parts’ [Kt1:874], Kant reaffirms several aspects of the foregoing understanding we have gained of ‘architectonic’: ‘The originative idea of a philosophy of pure reason itself prescribes this division, which is therefore architectonic, in accordance with the essential ends of reason ... Accordingly the division is also unchangeable and of legislative authority.’ Once again we see that this term entails that reason has prescribed a division (in this case, 4=2÷2) ‘in accordance with the essential ends of reason’; because it conforms to those ends (i.e., the categories as applied in the principles), the division can be regarded as authoritative and ‘unchangeable’.

          We could now proceed to verify and deepen the foregoing analysis of the meaning of ‘architectonic’ in Kant’s writings in two ways. First, we could examine his usage elsewhere in Kt1 of other terms that we now know are closely related to his conception of architectonic. In particular, a thorough study of words such as ‘plan’, ‘science’, ‘unity’, and ‘complete(ness)’ —especially when they appear in close proximity—would provide added insight into the details of Kant’s conception of architectonic. For instance, we would find that some of the key terms used in Chapter III of the Doctrine of Method are also used in the second edition Preface, where Kant first introduces the analogy between his philosophical approach and the Copernican revolution. Perhaps most significant, however, is that, shortly after introducing the table of categories in Kt1, Kant says: ‘this table is extremely useful in the theoretical part of philosophy, and indeed is indispensable as supplying the complete plan of a whole science, so far as that science rests on a priori concepts, and as dividing it systematically according to determinate principles’ [109]. When the language of the Architectonic chapter of the Doctrine of Method is kept in mind, such statements can hardly be interpreted as anything other than a direct confirmation that the table of categories is the key to the formal structure of reason’s architectonic unity.

          The task of examining all instances of Kant’s use of such language would quickly get out of hand, of course, and is not strictly necessary here in order for me to fulfill my goal of setting out a thorough analysis of Kant’s conception of architectonic. A more manageable way to conclude the discussion will be to examine how Kant uses ‘architectonic’ in texts other than Kt1. Without attempting to exhaust all occurrences in Kant’s corpus, let us look briefly at the other occurrences in Kant’s main systematic works. Kant does not use the term ‘architectonic’ in Kt2, Kt3, Kt5, Kt6, Kt8, or Kt9. But he does use it once in the Preface to Kt4, four times in Kt7, and twice in Kt10. Let us therefore examine these passages in order.

          In Kt4:10 Kant reminds us that the Critical task ‘of determining the origin, contents, and limits of a particular faculty’ must by its very nature ‘begin with an exact and ... complete delineation of its parts.’ But in order for such a task to succeed, something more than just a collection of parts is needed, something ‘which is of a more philosophical and architectonic character’—namely, ‘to grasp correctly the idea of the whole, and then to see all those parts in their reciprocal interrelations, in the light of their derivation from the concept of the whole, and as united in a pure rational faculty.’ So here again we see that Kant’s architectonic has to do with the relation between the parts as they are united together in a complete system under a single organizing idea of reason.

        Although Kant does not use the term ‘architectonic’ in the first half of Kt7, he does provide some reflections on architecture (Baukunst) that are at least indirectly relevant to his metaphorical use of the former term. He points out that in architecture, as in all fine arts, ‘design is what is essential’ [225]. Later, he defines architecture as ‘the art of presenting concepts of things which are possible only through art, and the determining ground of whose form is not nature but an arbitrary end’ [322], adding that the ‘adaptation of the product to a particular use is the essential element in a work of architecture’ [322]. With these comments in mind, it seems highly unlikely that Kant would have dared refer to architectonic as an ‘art’ [Kt1:860] if he were actually attempting to combat the tendency to draw analogies between it and architecture. As it stands, his reference now seems to imply that, just as the architect imposes a conceptual design onto a physical structure so it can be used for a given purpose, so also the archi­tectonic philosopher imposes a logical design (the idea of the whole and the ordering of its parts) onto a conceptual structure so it can serve the purposes (ends) of reason more adequately.

          The second half of Kt7 refers to architectonic twice in the main text and twice in the Appendix. By highlighting the relevant terms in the first passage, we can see that it provides clear evidence that Kant has an architectural analogy in mind:

 

Every science is a system in its own right. It is not enough that in building [something] in the science we follow principles and so proceed technically; we must also set to work with the science architectonically, treating it as a whole and independent building, not as an annex or part of another building, though we may later construct, starting from either building, a passage connecting the one to the other. [381 (tr. Pluhar), t.b., e.a.]

 

Here Kant is referring to the relation between different natural sciences, but the same applies to Kt7’s role in the architectonic of his System as the ‘passage’ that connects systemt and systemp. A few pages later he mentions in passing that our ability to determine ‘actual natural purposes’ is based not on the mechanical principles of nature, but ‘on a wholly different kind of original causality, namely, an architectonic understanding’ [388-9]. The first Appendix passage merely affirms this identification of ‘an architectonic understanding’ with our ability ‘to assume a teleological principle for judging ... natural purposes’ [420]. Whereas these passages all emphasize the scientific aspect of architectonic, the second Appendix passage emphasizes its artistic aspect. Kant explains that a teleological conception of nature requires us ‘to subordinate the mechanism of nature to the architectonic of an intelligent author of the world’, that is, to God, regarded as the ‘supreme artist’ [438].

          Finally, the term ‘architectonic’ appears twice in Kt10. Near the end of a section of the Introduction entitled ‘Specific Logical Perfections of Cognition’, Kant describes ‘the architectonic of sciences’ [48-9(54)] as ‘a system according to ideas in which the sciences are considered in respect of their relationship and systematic connection in a whole of cognition that is of interest to mankind.’ The second passage further explains that a science’s ‘idea’ is ‘the general delineation or outline of it, thus the extension of all cognitions belonging to it’ and that ‘[s]uch an idea of the whole ... is architectonic’ [93(98)]. Here again we see architec­tonic being depicted as the unifying aspect of science, whereby an idea is used to establish a systematic relationship between the parts of a whole in order to fulfill the interest (i.e., ends) of humanity.

          Although I freely admit that I wrote KSP1 without having first carried out a thorough­going word-study of ‘architectonic’ in Kant, the foregoing correc­tion of that oversight has not necessitated any substantial change in the position I previous­ly attributed to Kant. On the contrary, this section has confirmed my interpreta­tion of Kant’s architectonic as a method of using logical divisions such as those embodied in the table of categories to organize otherwise haphazard collections of concepts in order to provide an ‘idea of the whole’ that will guarantee both completeness and systematic unity in any science. The new insights provided by this more in-depth study—e.g., that the function of architectonic in the Doctrine of Method corresponds to the category of relation, and that architectonic divisions must serve the ‘essential ends’ of reason (i.e., reason’s drive towards unity in all of its manifestations)—constitute refine­ments rather than alterations of my previous position.

 

4. Minor Corrections to KSP1

          The following are corrections of the minor errors I have come across in the text of KSP1. Any future printing(s)/edition(s) of that volume should incorporate these changes. They are presented here for the benefit of those who already own a copy of the 1993 edition:

 

p.60: in the indented quote, two words are partially missing, apparently due to a paste-up error; they should read ‘which’ and ‘forms’.

p.101 (Figure III.9): the ‘theoretical’ cross should be positioned beneath the ‘practical’ cross, rather than above it. In this way the structure will correspond more appropriately to the parallel diagram given above, in Figure AIII.2, as well as to the arrangement shown in Figure VI.1, below.

p.285 (Fig. VIII.1): ‘categorial’ in step eight should read ‘categorical’ (as shown on p.273).

p.339 (line 14): ‘]’ should read ‘)’.

p.342 (line 11): ‘I Cor.’ should read ‘1 Cor.’.

p.351 (line 12): ‘in tact’ should read ‘intact’.

p.434 (line 2 of footnote): ‘W4’ should read ‘W5’.

p.447 (listing for Pq15): ‘in press’ should read ‘pp.85-108’.

p.447 (listing for Pq17): ‘in press’ should read ‘pp.129-48’.

p.447 (listing for Pq19): ‘Theocracy’ should read ‘Biblical Theocracy‘ and ‘the publishing information should read ‘(Hong Kong: Philopsychy Press, 1993).’

p.467 (Index listing for ‘hope’): ‘37b’ should read ‘37n’.

  

  [1].  I would like to thank Man Sui On, perhaps the most insightful of all my former students, for devising this helpful method of using three intersecting circles to map the close relationship between a 1LSR (the three standpoints), a 2LAR (each set of four perspectives), and a 3LAR (the Perspective of the whole System). Note that in this way the three intersecting circles depict all three of the main ‘levels of perspectives’ discussed in KSP1:II.4 [s.e. Fig. II.1].

  [2].  Making this point was one of my main reasons for using a spiral in KSP1:IX.4 to map the inter­relations between the three systems [s.a. Fig. III.8, above]. The spiral, like the three intersecting circles here, starts with the unknowable thing in itself and passes through the System to rest in the fullness of immediate experience. The advantage of expressing this in terms of three circles is that it accurately reflects the fact that each system aims at such fullness, in its own way.

  [3].  This new way of depicting how the synthesizing function of Kt7 is related to the fact that it con­sists of two unusually distinct divisions was also suggested to me by Man Sui On [see note AIII.1] in a conversation we had in February of 1995. His acute understanding of and appreciation for the geometry of logic has been an enormous benefit and encouragement to me.

  [4].  Mc94:121. Thus, she criticizes my ‘theocentric reading of the architectonic’ as ‘requir[ing] a substantive defense that is not provided here.’ The reason for this, however, is that the three Chapters constituting Part Four of KSP1 are intended only as sketches of subsequent volumes, not as demonstrations in and of themselves. The missing substance is provided here in KSP2, one key purpose of which is to demonstrate the ‘theocentric’ thesis [see I.3].

  [5].  KSP1:II.3 applies this purpose to all of Kant’s perspectival equivalents, not only those that I claim in II.4 to be technical constituents of his System. Because Kant himself did not recognize the need for a set of technical terms to define his levels of perspectives, he sometimes used perspecti­val equivalents in very unsystematic ways. An unfortunate by-product of my decision to translate Kant’s perspectival equivalents as such is that some of his nontechnical uses (e.g., ‘human perspective’, epistemological perspective’, etc.) ended up appearing in KSP1. Although Krantz does not raise this issue, it was a major concern expressed by an anonymous reviewer of an early version of Chapter II, which I had hoped to present at a Kant Congress in the mid-1980s.

  [6].  Our initial debate was so engaging that Manchester and I decided to present a joint paper at the World Congress of Philosophy in Boston (August 1998), entitled ‘Kant’s Architectonic Turn as a Model for Philosophic Practice: The Philosopher as Architect, Teacher, or Friend?’. That paper raised many of the points made here and/or in AIII.3, but also went further in several respects. But after reading my strengthened position, Manchester asked me not to quote from her email messages and to avoid presuming to represent her current position. I am therefore unable to respond here to her specific criticisms of KSP1, several of them quite valid. What appears here should be regarded as my own inferences as to how Hu81 relates to KSP1, not as representing Manch­ester’s most up-to-date views on Kant, nor her best response to my treatment of Kant’s architec­tonic.

  [7].  This is Manchester’s preferred interpretation of the phrase Kemp Smith translates as ‘doctrine ...’ in the sentence ‘architectonic is the doctrine of the scientific in our knowledge’ [Kt1:860]. What Kant means by this phrase, I believe, is that architectonic determines—i.e., ‘teaches’ or ‘indoctri­nates’—science (i.e., knowledge) to be scientific; it does this by systematiz­ing the empirical data. Oddly enough, Manchester herself gives no explanation or defense of her unusual translation, nor does she adequately explain what the sentence in question actually means, despite using this phrase as the title of section 1.21. She merely repeats it throughout Hu81, as if this somehow justifies the assumption that architectonic is related to the art of teaching—an assumption that, as we shall see in the following section, is not supported in any of the other texts where Kant uses the term ‘architectonic’. Moreover, Manchester’s treatment of this aspect sometimes comes perilously close to the ‘aggregate’ approach to philosophizing that Kant explicitly regards as the antithesis of the architectonic approach. For instance, she says in Hu81:174: ‘Architectonic as a teaching of the scientific seems to be a kind of pedagogical exercise in which actually existing attempts at philosophy are given for critical examination.’ Maybe. But such ‘critical examination’ is not architectonic just because it involves teaching; it will be architectonic only if the teacher provides an ‘idea of the whole’, on the basis of which each failed philosophy can be explained and given its proper place in the historical realization of the essential ends of human reason.

  [8].  A similar possibility can be raised in response to Manchester’s treatment of Krug’s entry on ‘Architektonik’ in his 1832 dictionary. After quoting his definition, she observes: ‘Conspicuously absent is Kant’s conception of architectonic’ [155]. What she fails to note, however, is that the standard interpretation of Kant’s conception is present in Krug’s entry: he begins with an etymology that directly connects architectonic with architecture and architects, then proceeds immediately to call it ‘the art of building a scientific teaching structure, for which logic gives instruction, or a scientific basic outline’ [q.i. 155, e.a.]. So Manchester is begging the question. She falsely assumes that architecture and logic have nothing to do with Kant’s concep­tion of architectonic, then proceeds to declare anything relating to these two to be contrary to Kant’s conception, and finally dismisses as irrelevant any texts that seem to connect these factors to Kant’s conception. All along the outcome has been predetermined by her presupposition.

  [9].  McFarland voices the majority opinion on the issue of Kant’s architectonic when he says [Mc70: 39]: ‘If Kant were as much in the grip of his architectonic as his critics make out, he would not be the great philosopher that he is.’ He later [70n] provides a good sampling of references where English critics (Bennett, McFarland, Walsh, Weldon, and Wolff) adopt views of Kant’s ar­chitecton­ic virtually identical to Adickes and Kemp Smith.

[10].    I wrote a complete first draft of the book in 1985-86, as a background for defending my use of diagrams in my doctoral dissertation [see AI, above]—diagrams I ended up omitting anyway, in response to an ominous warning from my supervisor about the dire consequences of presuming to challenge the examiner’s initial impression of their incoherence. Though I have worked on revising that manuscript at various points over the past 14 years [see Pa99], the urgency of other projects (including this book) has prevented me from completing a publishable draft up to now.

[11].    Ignoring such facts, Manchester speaks loosely in Chapter III as if her hobby-horse comes directly from Kant [Hu81:106-7]: ‘As [Kant] points out in the Critique, when architecture becomes the model for the philosopher’s art, everyone goes off to erect a separate structure according to his own plan. But the deepest irony ... is the equation of leading and commanding with designing and construction.’ Yet Kant never ‘points out’ anything of the kind in Kt1 or any of his other writings! His criticisms, as I have explained, are not directed against the idea that philosophers should design and/or construct systems—this would make Kant into the biggest of all hypocrites!—but against the idea that one’s plans are improperly designed, so that the resulting building does not properly serve the ‘essential end’ of human reason, which is to attain unity.

[12].    A good example of Manchester’s antipathy towards Kant’s own chosen metaphors comes in Hu81:65, where she quotes from Kt18:317(37). The point of Kant’s complaint against philoso­phers who ‘keep changing or rejecting the ground plans according to the building materials pro­vided’ is not that philosophers have no business building or designing plans, but rather that in phi­losophy the plans ought to determine the nature of the building materials, not vice versa.

[13].    See the relevant entries in Pa87a. That Kant intends the categories to be the chief organizing structure in his architectonic plan becomes more and more obvious in some of his subsequent works. Perhaps the best example is Kt3, whose unity and completeness, as Ellington argues in El70, is explicitly based on the table of categories. That the categories can apply to the order of Kant’s exposition as much as to the order of reason is suggested by Kant’s remark in Kt9, that the division of categories, as applied to the moving forces, ‘constitute[s] the stages of the transition from the metaphysics of corporeal nature to physics’ [22.135(45), e.a.; s.a. 21.483(43)].

[14].    Kt1:860. Manchester rightly points out that Kant’s German has nothing equivalent to the word ‘constructing’ in this passage; I have placed it in brackets to show that Kemp Smith merely inserted it to clarify his own understanding of Kant’s meaning. (The Cambridge edition of Kt1 leaves the passage ambiguous by translating the phrase merely as ‘the art of systems.’) Manchester also retranslates Kemp Smith’s ‘doctrine of the scientific’ as ‘teaching of the scientific’—a change that is advantageous to her purposes, but not necessarily an improvement on Kant’s meaning.

[15].    Overlooking this fact, Manchester takes one of these examples, Kant’s division of all philosophy into the ‘school concept’ and ‘world concept’ types [Kt1:866-8], to be an essential part of Kant’s discussion of the conception of architectonic as such [see e.g., Hu81:87-100 and passim]. While she may be correct to claim that Kant takes the ‘Weltbegriff of philosophy’ to be the only concept that properly accommodates architectonic, so that an indirect relation holds between them, Manchester’s tendency to treat anything Kant says about this Weltbegriff as ipso facto a characteristic of architectonic is wholly unjustified by the text.

 

Back to the Table of Contents for this book.

Back to the listing of Steve Palmquist's published books.

Back to Steve Palmquist's home page

Send comments to the author: StevePq@hkbu.edu.hk