Between the Indeterminate and the Determined-- from the Viewpoints of Dignaga and Kant.
Also an Attempt to Reconcile Epistemoligists and Metaphysicians
As Dan Arnold in his “Is Svasamvitti Transcendental?”[1] has captured, there seems to always exist a tension between epistemological approach and metaphysical approach towards the most urgent yet lasting question – what's going on with our very own existence and how we can better it. In the early middle age India there were Dignaga's epistemology and the following debate between Madhyamika opponents and Dignaga's advocates regarding the issue self-awareness (svasamvitti); in the eighteenth to the early nineteenth century Europe, there were Kant's epistemology and the following debate on the same issue between the anti-epistemic German-idealist reconstructors and the epistemologist-Kant's sympathisers. With presenting the parallel, as well as with entangling these two lines so that the both ends on the one line reflects upon their comrades on the other, Dan Arnold tries to demonstrate how a transcendental reconstruction of Dignaga's epistemology in Kant's fashion (Dan Arnold's interpretation of Santaraksita's comment on Dignaga) could help out the middle age Indian epistemologists from the attacks of Madhyamika thinkers and thus suggests a positive answer to the proposed question: Dignaga's svasamvitti as reconstructed by Santaraksita is indeed transcendental. In this article, as a supplement, I attempt to argue that with Dignaga's own teachings alone in the
Pramanasamuccaya, Dignaga's pramanavada agrees with Kant's transcendental idealism. And then, after the transcendental nature of both epistemologies get assured, I will focus on the relation between Dignaga's perception and inference and the relation between Kant's intuition and concept; on the one hand, I will try to clarify their systematical difference, and on the other, I will try to reconcile the two systems with a “schematic” reconstruction in terms of the swing between the indeterminate and the determined. Hopefully the effort here could also resolve the antinomy that epistemology such as Kant's and Dignaga's is a righteous course and that metaphysics such as Madhyamika's is a righteous course, too, regarding the most urgent and lasting question.
Kant distinguishes “transcendent” and “transcendental” as “not interchangeable terms” in the beginning of his “Transcendental Dialectic” in
Critique of Pure Reason, CPR, where he entitles the principles “whose application is confined entirely within the limits of possible experience, immanent,” i.e., transcendental, while the principles “which profess to pass beyond these limits, transcendent.” (CPR A 296/B 352) Also, the transcendental is distinguished from the empirical that the latter must be obtained via the employment of sensibility and thus must be
a posteriori, while the former must come
a priori, i.e., must be required so that the employment of sensibility is possible; the knowledge
a posteriori must be the result of our apperception, via which our consciousness arises, while the knowledge
a priori is the condition of our apperception and thus known to us in our consciousness yet as form or as transcendental ideas. Clearly we can see that the division between the transcendental and empirical is a division of the origin of knowledge and/or awareness and the effect of that origin; due to such a division, our knowledge and awareness are thus allowed to be characterized with two aspects: form and matter. In his
Critique of Judgment,
CJ, where the employment of our cognitive power is further characterized as an ability about “determination.” This development, I believe, is a further exploration of the primordial status of apperception and judgment, namely, his so-called “transcendental reflection” in CPR, the form of judgment prior to all judgments, one which makes possible the reference of a concept to an intuition. It is in this part of his transcendental task the notion of transcendental is put in a brighter light.
In Kant's theory of threefold synthesis in
CPR, the notion of transcendental is presented in a mechanical fashion: via the synthesis of apperception, the inner determination of our mind[2] is accomplished, that is, our mind becomes conscious of the object of the determination which is always accompanied by self-awareness, a thought[3], too. The synthesis is then taken as the condition of our consciousness and thus of our all possible experiences. It is said in the deductions (the first edition mainly) of
CPR that in order for experience to be possible at all, the apprehension of intuition, the reproduction in imagination and the recognition in concept must already be united in a schema; this is the condition of all possible experience. To put in a fashion more critical, we have our experience as such must be resulted from a system in coordination as such
a priori which causes, i.e., conditions and makes possible, the experience as such, or otherwise, the aspects of our experience (appearance, intuition, and concept) cannot co-occur in our experience as in unity.
In
CJ, judgment is divided into determined and reflective (indeterminate); in the former, the association of imagination is based on the rules of a concept which the object of the association is to be known about, while in the latter, the association of imagination gives a basis for understanding to apply various concepts to. To put in less Kantian technical vocabulary, judgment is an ability of locating a particular
as, i.e., in imagination, contained under a universal; when the universal is given and the judgment subsumes a particular under it, this is determined; when only the particular is given and the universal has to be found for it, the judgment is reflective. Kant describes our judgments as the interactions between the faculty of rules, understanding, and the faculty of association, imagination. In a determined judgment, understanding offers a set of rules which is implied by a concept while imagination associates in accordance with the rules. To characterize this with the model in
CPR, imagination relates the synthesized manifold intuition α as well as the appearance
X, i.e., manifold of sense through the synopsis of the manifold
a priori, in accordance with the conceptual unity 'α' of this synthesis through transcendental apperception. Once the condition is satisfied, the judgment that the particular
X/α is subsumed under the universal 'α' is made, from which arises the consciousness of an α which is known as a case of the governing concept 'α' as well. In an indeterminate judgment, imagination creates an object
X which understanding keeps trying to find a concept for. To characterize this with the model in
CPR again, the particular
X is given (in nature) or created (in art) first through imagination, and various universals such as 'α', 'β', 'γ', 'δ' etc. are tried to be found to grasp the manifold
X when imagination at the same time reproduces intuitions α, β, γ, δ etc. Here Kant has actually presented to us a further exploration of the transcendental.
First, we have experience without exception in such coordination, and hence we gain the reason to idealize a system coordinated as such. Second, the idealized system must be the origin and cause of our experience for we so idealize it. Third, the necessary coordination in our experience is that in every instance of consciousness there must be a relation between a particular and a universal, so we idealize a system with two faculties as the epistemic origins of the particulars and the universals respectively. Fourth, in order to approve the ontological commitment so that our daily life, our interactions with the world, the interpersonal activities and morality can make sense at all, we must take it for granted that the particulars have external cause. But to our experience, our consciousness must come after the employment of sense, imagination and apperception, and hence we have no idea at all about the external cause except for such a causal commitment. Fifth, the particulars in our experience is fuzzy that they are both manifold and singular. This is schemed in
CPR and explained in
CJ. In the former, the manifold of the particulars have their origin in the sense, perhaps with a relation to the conceptualization, that we are given the form of sense as manifold, and they can be reproduced in imagination and cognized in concept. In the latter, when imagination becomes an ability that not only produces and reproduces intuitions with the ways of association originated from understanding, but can create as well something indeterminable (the appreciation of nature or the artistic object in imagination are created), something as manifold as the manifold of sense through the synopsis of the manifold
a priori and yet welcome to a lot of, even all possible conceptions. We can see the critical role imagination plays in our idealization here; it links the three modes of synthesis in the idealized coordination system. The so-far so-called idealization distinguishes itself for its self-referential nature, i.e., on the one hand the idealized system is directly inferred from the necessity in experience while the necessity in experience is deduced from the idealization; on the other, it is so idealized that the idealized system is the cause of experience as its function and that because of our experience has such forms we can have good reason supporting such an idealization. This kind of idealism is thus distinguished from what he calls “empirical idealism” (including dogmatic idealism such as Berkeley's and sceptic idealism such as Descartes's) which in general is potential to lead to the conclusion of mistrust of the particulars; in dogmatic idealism the objects of sensation, i.e., the objects in space, is merely imagined, while in sceptic idealism the objects of sensation cannot be established, and only the inner objects, the objects in time, can be indubitable. And the ground for such a premature conclusion is that the idealization is made in experience so that the idealized are regarded as empirically reachable entities, either as something we are totally live in though it is only imaginary, or as something we can directly cognize and completely indubitable and veiling up the particular making it mediate. From here, the transcendental is shown as the nature of such a distinguished idealization that the idealized is put into the position of the systematical cause of the systematical function (experience) and as the condition which gives forms to the function, and via the conditioning the idealization can be reasonable at all. Consequentially, we do not give up any of the particular and the universal; moreover, we preserve the manifold and immediate nature of the particular and the determined and mediate nature of the universal in well explanation, as how we indeed experience. To concretely define the scope of the transcendental nature, the relation between the particular and the universal has to be regarded in the position of an idealized cause only whose result enables any of us to know things. If it is the particular alone or the universal alone that is idealized, the idealization is made solely in our knowledge, that is, the relation between the particular and the universal is linked empirically, either as an empirical induction from the direct particular to the indirect universal or as an empirical inference from the direct universal to the indirect particular.
Stick to the above characterization of Kant's transcendental idealism, we can find that Dignaga's epistemology meets the spirit of it in terms of the following two points. On the one hand, Dignaga holds that there's a causal relation, which yet makes no distinction, between means of cognition and cognition as result and there's correspondence between the forms of the result and the original means. On the other, he holds that the sharply divided two means of cognition are divided in the origin, rejecting valid reachable independent universals (such as Descartes's the indubitable), recognizing conceptualization to be part of valid cognition, i.e., rejecting the independent particular as mere imaginary. Both points can be conveniently shown in Masaaki Hattori's comments (Note 1.9.) in his translation of
Pramanasamuccaya[4]: “Dignaga's theory is unique on each of these four points: (1) He recognizes perception (
pratyaksa) and inference (
anumana) as the only two means of cognition, and does not admit verbal testimony (
sabda), identification (
upamana,), etc. as independent means of cognition... (2) He characterizes perception as “being free from conceptual construction” (
kalpanapodha), and does not recognize determinate perception (
savikalpala-pratyaksa) as a kind of perception... (3) He sharply distinguishes the particular (
svalaksana) and the universal (
samanya-laksana), which are respectively the objects of perception and inference. He denies the reality either of the universal as an independent entity or of the particular as qualified by the universal... (4) Rejecting the realist's distinction between the means and the result of cognition, he establishes the theory of non-distinction between the two” – Dignaga writes “we do not admit, as the realists do, that the resulting cognition (
pramanaphala) differs from the means of cognition (
pramana). The resulting cognition arises bearing in itself the forms of the cognized object and thus is understood to include the act of cognizing (
savyapara)” (k. 7cd-8ab.) and “it can be maintained that the self-cognition or the cognition cognizing itself (
svasamvitti) is here the result of the act of cognizing” (k. 9a.) … “because the determination of the object (
artha-niscaya) conforms with it, viz., with the self-cognition” (k. 9b.).
Before I mixed the two epistemologies from the viewpoint of determination versus indeterminacy, I have to point out the systematical discrepancy between the two and also clarify the issue of Kant's I-think as a thought while Dignaga's
svasamvitti, self-awareness, as an intuition, which is also noticed by Dan Arnold.
Although both rejects the reality of independent universal (mere conceptual construction without a proper particular for it), Kant and Dignaga holds different views of truth: Kant tempts to take as truth the proper reference between the particular and the universal, whereas Dignaga groups the particular as qualified by universal, i.e., determined perception, with the untrue cases, embracing the particular which can never be generalized or conceptualized as truth. Moreover, the distinction between intuition and concept in Kant and the distinction between perception and inference in Dignaga are not well-paralleled. To Kant it is not sharply divided between appearance (manifold of sense through the synopsis of the manifold
a priori) and intuition, both being the object and product of imagination, while to Dignaga the determined particular is not true perception, only the indeterminate perception is perception. Due to the discrepancy, there occurs the chance for the issue whether Dignaga and Kant disagree with each other on the nature of self-awareness being a perception or a thought.
The issue can be conveniently get passed by bringing Dignaga's words: “even conceptual construction, when it is brought to internal awareness, is admitted as a type of perception. However, with regard to the external object, the conceptual construction is not admissible as perception, because it conceptualizes the object.” (k. 7ab) To Kant, the unity of apperception must be a thought, I-think, that unites the manifold of sense and the manifold of intuition; the thought here is not an empirical thought, but a transcendental thought which conditions and makes possible the self-awareness in cognition. Actually we can observe here that Kant's self-awareness and Dignaga's self-awareness do not equate each other: first, Kant presents the notion in transcendental vocabulary, while Dignaga presents the notion not in such a strong distinction between
a priori and
a postiriori; moreover, Kant's talking about self-awareness is confined mostly within the scope of being the unity of apperception while Dignaga tends to include the inner feelings, which to Kant would also be intuition yet merely as empirical one and hence not included in the talking. Nonetheless we can explain as well that Dignaga takes the conceptualization unity as part of perception, too, with his own explanation that such a conceptualization is not one over an object, but a mental activity which is brought into internal awareness. But, there's much more to talk about in this issue. Because Kant emphasizes more on the reference between the particular and the universal, stowing away the fuzzy relation between indeterminate particular and the determined particular by assigning both tasks under imagination's shoulder, his major goal is naturally set at explaining how such a reference can be possible, as consequence of which the uniting side of the transcendental unity pops out as the core nature of self-awareness. However, in CJ, we can find Kant moves his attention on the united side of the transcendental unity --indeed, in CPR, Kant already holds that the intuition has to be ready for conceptualization; however, in CJ, he reaches the principle of purposivessness which works in the relation between imagination and understanding and in the relation of the object of imagination and the object of understanding, and he gets closer to the standpoint that the indeterminate judgement is prior to determined judgment. On the contrary, to Dignaga what is important is the manifoldness of perception, the independent particularity. Unlike Kant's holding the firm ground of how the particular is determined so that the indeterminacy becomes the pursue, Dignaga embraces the indeterminacy in the outset so that it becomes an issue needing explained that in a valid cognition, how the conceptual construction can be admitted as perception, i.e., self-awareness. But the significance of the above-mentioned meeting of the two directions (from uniting side to the united side and the reverse) lies at one common goal of both projects: there must be two transcendental origins/means so that its result, the cognition, is possible as such – as with two aspects: the particular and the universal; there must be a transcendental unity of the two means as the cause and origin of our cognition as such, so that we can experience as such – as with the reference between the particular and the universal; the transcendental unity must be the idealized origin of what we attribute with “I” and hence it must be a combination of the uniting and the united. It is at this very point that Kant and Dignaga departs away from each other: Kant grasps firmly the combination so that the empirical reality is committed, while Dignaga leaves the combination and embraces the united, or better, the to-be-united, in order to get close to which an epistemology as such (making clear with strict and careful critique and excluding step by step the uniting conceptions, then the particular as qualified by the uniting conceptions, helps reach the to-be-united) must be made.
Dignaga's refusing the determined particular to be the untruth corresponds to Madhyamika's distinction between the ultimate truth and the conventional truth (二諦), whereas his admitting perception and inference can be altogether transcendentally idealized may upset the Madhyamika thinkers for in consequence this admitting could blur the sharp distinction between the ultimate and the conventional. The worry can be eased by arguing that, not forgetting the distinction of perception and inference lies in idealization and hence as mere forms or aspects of reality so that we should not simply identify perception with the ultimate truth and identify inference with the conventional truth, the sharp distinction between ultimate truth and the conventional truth is safe because the distinction is actually a distinction between the transcendental and the empirical: before determined and cognized, perception is indeterminate, the apperceptive result of which,
viz.,
pramanaphala, however must be always determinedly known – both perception and inference being transcendental means makes no harm to the distinction of their being transcendental ideas and their being forms/aspects of cognition. What is ultimately true is the indeterminable perception with manifoldness, which is thus ineffable unless in symbolism, but in order to assure a workable and moreover valuable conventional reality, it is significant to insist the inference must be properly related with perception, inference here including at once the conventional languages and especially the symbolisms targeting the true perception, such as theologies, metaphysics etc. – significant as Dignaga's doing
pramanavada that expresses the two aspects of cognition are originated from two means of cognition and yet resulting in single cognition with two aspects, the particular and the universal, as well as Kant's idealizing those conceptual constructions for which no intuitions can be found and admitting the rest as physically true with the firm groundwork
a priori expressed in
CPR that the appearance (manifold sensible awareness
a priori), intuition (sensible awareness
a posteriori) and concept of an empirical object are transcendentally united. Only after assuring the valid inference and physical reality and its relation to the pre-cognitive state, our views of the conventional truth and the ultimate truth can both at once be well explained and well re-oriented (oriented in accordance with what is given to us) so that paralogical symbolisms targeting the ultimate which greatly enhance people's confusion can be avoided.
Kant's final effort in his critical philosophy sketches the swing between the determined and the indeterminable. What is cognized is what is determined; and determination is made through imagination's following understanding's rules in association producing intuition. But the pre-cognized intuition,
viz., appearance, which is part of the apperception in its origin but loses the indeterminacy in its result, gives to cognition nonetheless the characteristic of the manifoldness as openness to all possibilities of conception and readiness for all possible conceptualization. With such a basis
a priori we are able to obtain a united, determined awareness so that we know and know about (in analysis) something, and with the same basis we are able to produce various intuitions (particularly determined awareness) in relating to the intuition itself (pre-cognitive perception with a manifold synopsis) in “free-play” to “appreciate” something when various concepts to match the indeterminable intuition can keep occurring and along holding up various correspondent particular intuitive awareness of it; because of such purposiveness of the employment of understanding and the employment of intuition, that intuition and concept are necessarily (such a necessity is assured for the unity's being transcendental) referable, first, the indeterminable can be known determined, and second, with the determined the indeterminable is witnessed. The significance of the above is, as Hattori comments on Dignaga: “rejecting the realist's distinction between the means and the result of cognition, he establishes the theory of non-distinction between the two” (Note 1.9.), first, although the known is always determined, since the known must have the aspect of indeterminacy, for there is the conceptual-unity-based crossover between the manifold
a priori and the manifold
a posteriori, the pre-cognitive indeterminable, i.e., the manifold of sense through the synopsis of the manifold
a priori, must agree
in form with the indeterminate witnessed with the determined, i.e., the openness to all possible conceptions and the readiness for all possible conceptualizations; and then, the intuition which is accompanied with a proper reference to its counterpart concept, i.e., Dignaga's “determined perception” is identical with the indeterminable perception, for if we exhaust all possible ways of presenting the intuition, i.e., if we apply all possible concepts to the intuition so that the intuition's all possible aspects are shown in unity – in the intuition itself alone, i.e., if we appreciate freely the object, the known object is directly the true perception, the indeterminable
a priori. This is evidenced in the fact that we only experience one world, that we experience the sensible world and the intellectual world as one identical world, as well as in Dignaga's insisting on rejecting realists' distinction between the idealized means and the reality of cognition and in Kant's refuting empirical idealism. Both the realists Dignaga rejects and the idealists Kant refutes are empirically separating the world into reality and ideas. They both take the means of cognition as objects known, namely, empirical objects, though they entitle them with “ideas”, and take the transcendental idealization as cognizable causal relation; that is to say, they mistakenly realize the transcendental ideas. Consequentially, they think they should try to make themselves go beyond the phenomena and get true knowledge of the “transcendent” ideas, and thus beside of the world of cognition, another world of ideas is diverged. It is exactly this point that bothers Madhyamika thinkers. Candrakirti's criticism against Dignaga's epistemology is basically with the incentive to reject svasamvitti's being real, as if there were indeed svasamvitti in an empirically spiritual, ideal world.
Candrakirti argues, according to Dan Arnold, that Dignaga mistakenly creates something defining svalaksana, which conventionally means the property of the thing in itself and the reference its name refers to, depriving the names of their external, in-itself references on the one hand, and castrating svalaksana's self-defining nature on the other. Besides, the thing that Dignaga believes to define svalaksana must be itself svalaksana again and requires another created thing to define it, inviting the endless regression problem. The criticism above is made on the misunderstanding of the means of cognition as cognizable, empirical objects and on mistaking the causal relation between means of cognition and cognition to be cognizable, empirical causality, which leads to Dignaga's object of refutation: the empirical distinction between means of cognition and cognition. But more significantly, what bothers Candrakirti is that Dignaga seems to admit that “self” indeed exists as any empirical object, to which of course Dignaga directly expresses his dissent in his
Pramanasamuccaya. But Candrakirti is not satisfied with the simple dissent. The following argument might respond to this attack. Though there is svasamvitti and svasamvitti is perception in Dignaga, svasamvitti is not empirical, for he won't call anything empirical “perception” since true perception does not include determined perception. Hence, svasamvitti should be something which allows means of cognition to operate like this so that its result, cognition, is brought up to us as known (known object and known to us), but not itself one object of cognition. Since svasamvitti is true perception and not itself one cognizable object, it is impossible to equate Dignaga's svasamvitti with the conventional, empirical self. And based on this clarification, we can point out that the conventional meaning of svalaksana leads to something Candrakirti may dislike. Granted that svalaksana is independent property that defines itself and allows itself to be referred to by names, then it becomes an issue how it is known to self. The answer to the issue will be that either we have the imaginary of everything and that's all mere imaginary or we have intensions which correspond to but never equate the extensions. Both of the candidates needs an empirical I. I think the Madhyamika thinkers, Dignaga and Kant all won't be happy with it.
On the contrary, Madhyamika's antinomy that the ultimate truth and the conventional truth are sharply distinct and yet the ultimate truth and the conventional truth coexist so that we can live our conventional life sincerely and yet are possible to reach nirvana are in such an epistemological project explained and expressed with conventional words which are understandable to conventional people. Hence the tension between Madhyamika thinkers' as well as German Idealists' metaphysical talk about emptiness (the indeterminable)'s being the ultimate truth while encouraging conventional practices, and Dignaga's as well as Kant's epistemological project establishing the means of valid cognition, should be reconciled. (Dept. of Philosophy, NCCU)
Reference
Arnold, Dan. “Is Svasamvitti Transcendental? A Tentative Reconstruction Following Santaraksita” in
Asian Philosophy Vol. 15, No. 1, March 2005, pp. 77-111.
Hattori, Masaaki trans.
Dignaga, On Perception, being the Pratyaksapariccheda of Dignaga's Pramanasamuccaya from the Sanskrit fragments and the Tibetan versions. Cambridge & Massachusetts: Harvard University Press, 1968.
Kant, Immanuel.
Critique of Pure Reason. Trans. by Norman Kemp Smith. Bedford, St. Martin's, Boston & New York: Macmillan, 1929.
Kant, Immanuel.
Critique of Judgment. Trans. by Werner S. Pluhar. Indianapolis & Cambridge: Hackett Publishing Company, 1987.
Endnotes:
[1] Arnold, Dan. “Is Svasamvitti Transcendental? A Tentative Reconstruction Following Santaraksita” in
Asian Philosophy Vol. 15, No. 1, March 2005, pp. 77-111.
[2] Representation, our awareness of an object, is defined in
CPR as “inner determination of our mind in this or that relation of time” (CPR A 197/B 242).
[3] “In the transcendental synthesis of the manifold of representations in general, and therefore in the synthetic original unity of apperception, I am conscious of my self, not as I appear to myself, nor as I am in myself, but only that I am. This representation is a thought, not an intuition” (CPR B 157).
[4] Hattori, Masaaki trans. and comment.
Dignaga, On Perception, being the Pratyaksapariccheda of Dignaga's Premanasamuccaya from the Sanskrit fragments and the Tibetean versions. Cambridge & Massachusetts: Havrod University Press, 1968.
Edited 12 time(s). Last edit at 11/11/2009 09:06PM by gustav.
(
編輯記錄)