04 December 2012

Kant and Aquinas on the Beginning of Time

[The following was written as a term paper for one of my courses at the Dominican House of Studies this semester.]

Immanuel Kant and Thomas Aquinas both take up the question of the knowability through reason of the world's beginning in time, and both conclude, contrary to a large number of their contemporaries, that the human intellect in its present state is incapable of determining whether the world has an infinite past or terminates historically in a moment of origin.  The two differ on many other points, and so their consonance on this issue is striking.  It invites a comparison: how does each thinker arrive at this conclusion?  What role does it play in their treatment of the relationship between faith and reason?  In the following we will attempt to sketch out an answer to these two questions.

We begin with Kant.  Kant's discussion of the finitude of the past is located in the second half of his classic Critique of Pure Reason.  In the first half, the “Transcendental Analytic”, Kant outlines the formal preconditions (sensible and intellectual) for the possibility of human cognition.  These are twofold: first, the pure forms of intuition (space and time), second, the pure forms of understanding (the twelve categories).  Kant's primary objective in the Transcendental Analytic is to show, contra Hume, that basic metaphysical categories like substance and causation can be legitimately applied to the contents of experience.  To this end, he reverses the traditional ordo rerum, according to which the reality of metaphysical concepts is grounded in the being of things, and makes it accord instead with the ordo inventionis.  In other words, for Kant the mind's action in discovering being is also the action by which being is conferred upon the objects of thought and experience.  (He calls this inversion the "Copernican Revolution in Metaphysics".)  This reversal is, from the perspective of common sense, bizarre, and it proves ultimately devastating to the very project of metaphysics, but it serves Kant's purposes very well.  By inverting the order of being and identifying it with the order of discovery, Kant can offer a strong response to the Humean skeptic.  Metaphysical concepts have the right to apply to experience, because they are the fundamental means by which experience itself is constituted.  

He argues for this position by deducing it from a representational theory of cognition, pointing out that what we know of objects is always conditioned by the receptivity and structure of the mind: Every mental state is a representation.  Representations are in the mind, not in things.  Therefore they are conditioned primarily by the parameters of the mind, having been conformed to and limited by its potential.  Since we cannot know except by representations, and since all representations must, as representations, be possibly present in the mind as some mental state (conscious or not), it follows that the determination of the essential nature of the objects present in experience should flow from the investigation into the structures of the mind by which they are possible.  Thus epistemology undercuts metaphysics as first philosophy.  "Being" cannot responsibly be treated by the philosopher except as constrained by the nature of the mind, and therefore all the abstract questions about Being which philosophers traditionally discuss must be reined in by a critical appraisal of the legitimate extension of metaphysical concepts.

The second half of the Critique of Pure Reason, the "Transcendental Dialectic", is exactly such a critical appraisal, applied to what Kant considered the three core subjects of traditional metaphysics: rational psychology, cosmology, and theology.  Kant’s treatment of these subjects is preceded by a discourse on what he refers to as “transcendental illusion” (B249ff.), the habitual tendency of the mind, equipped with certain apparatus for the synthetic construction and interpretation of the objects of experience, to direct this apparatus toward absolute or “transcendental” objects.  The particular power of the mind which is directed toward transcendentalizing the ordinary functions of understanding and taking them to an absolute limit beyond the bounds of possible experience is called “Reason” in Kant.  Thus he distinguishes three powers of the intellect, parallel to the three operations traditionally discussed in scholastic epistemology: first apprehension, second judgment, third reason.  

Reason draws out new judgments from old intuitions by investigating and exploiting the prior synthesis of our representations performed by the intellectual and sensible faculties.   It moves from effects to causes, from accidents to underlying substrates, from wholes to parts, from extensive magnitudes to bounded totalities, etc.  The operations of reason are all applications of the internal relations, logical and metaphysical, among the Categories.  In its merely logical use, reason is totally legitimate, but once it takes on a transcendental function it quickly generates errors.

For example, in applying the notion of causation to some empirical object, Reason follows the chain of effects backwards to uncover their causes.  This is legitimate.  But in its transcendental use, Reason bypasses this local investigation and asks the big question: Does the chain of effects terminate in some ultimate cause?  Here trouble arises, because each answer to the question seems capable of raising fatal objections to the other side:  if the series of causes terminates, then there must be an uncaused cause, which is unintelligible and violates the ordinary notion of causation.  However, if the series of causes is interminable, then there is no sufficient explanation which grounds the action of causes in the present, again seemingly invalidating the ordinary notion of causation.

In his “Antinomy of Pure Reason”, Kant presents his readers with a series of similar aporias, each consisting of a pair of apparently sound arguments demonstrating the falsity of each other’s conclusion by reductio ad absurdum.  The first of these (B454ff.) concerns the limits of space and time.  The arguments on both sides make reference to the successive synthesis of an infinite series of events.  There must be a beginning, because an infinite synthesis can never be completed and thus, without a beginning in time, the present would not exist.  There must not be a beginning, because a synthesis cannot coherently join the preceding void or “empty time” to the first moment of change.  

Kant’s next move is to solve the aporia by rejecting both positions and suggesting that the difficulty they provide is illusory and parasitic upon an illegitimate, transcendental use of reason.  The problem is this: in asking questions about the limits of time and space, Reason has erroneously hypostatized the world.  But, Kant says, the world itself is not a measured totality discovered by the mind as an actual whole, but rather the product of a progressive synthesis performed by the mind.  There is no determinate totality to point to as “the world”, nor can there be a “whole” of history or time, because the succession of empirical phenomena within space and time is a product of the ongoing combination of sensible manifolds under conceptual unities.  To ask about the beginning of time or the edge of space is a category error, but, interestingly, this category error is almost inevitable because the structure of the mind is founded on the use of the categories to construct and interpret the contents of experience.  When we turn our minds to ultimate things, we will almost necessarily slip into error, not realizing that our conceptual tools are not equipped for such high tasks.

Kant’s conclusion, then (B545-52), is to say that the synthesis produced by the mind in reaching backwards to prior events will always be finite and determinate, but will also always be open to further extension.  In any particular consideration of things, the mind reaches back to a first moment, not as absolutely first but as conditionally so, while treating this beginning as itself apt to investigation.#  The human mind is thus totally incapable of discovering the world’s beginning, because the world is a product of the mind itself, which is finite and conditioned by its own synthetic structure.

At the beginning of the Critique of Pure Reason, in his preface to the second edition, Kant explains that his goal in clearing away the delusions of metaphysics was to abolish knowledge in order to make room for faith (Bxxix).  In some cases, this view of his project is plausible.  He affirms that it is possible to believe in God, but also rejects the provability of his existence.  He defends the abstract possibility of free will on the transcendental plane, but says that empirically the notion makes no sense.  But for the question of the world’s beginning, Kant’s solution does not leave any room for faith.  Rather, because space and time are merely subjective forms of intuition imposed upon sense data by the mind, the lack of any subjective boundary for space and time implies absolutely that there can be neither a beginning in time nor an infinite past.  There is no transcendental analogate to a “beginning in time” for Kant, so this belief seems to be strictly ruled out.

Now we move to Aquinas.  St. Thomas’s treatment of the question of the world’s beginning in q. 46 aa.1-2 of the first part of the Summa Theologiae comes in two parts.  First, he asks whether the world always existed, and second whether it is an article of faith that the world began.  Before proceeding to the articles themselves, we should contrast St. Thomas’s underlying epistemological and metaphysical views with Kant’s.  

Thomas is operating in a traditional metaphysical schema.  The knowledge we have of things is consequent upon the being of things, and not vice versa.  Aquinas is not battling a phantom skeptic, and thus does not feel the need to construct an inverted metaphysics predicated on subjective unity of consciousness.  Instead, the task of the Summa is to revise the order of investigation for students new to theology so that the material is presented according to the intrinsic order of the subject matter, rather than being subordinated to an accidental process of discovery.  Aquinas begins with the subject of theological science in itself (God himself, ipsum esse per se subsistens) and then works outward to all of the consequences and effects which flow from that principle, deviating from the order of being only insofar as deviation is necessary in for the understanding of the student.  The student is led forward by questions motivated by possible objections to sound doctrine, so that at every point various arguments are being pitted against each other in order to further elucidate and deepen the student's appreciation for the truth.

Beyond his methodological outlook, Thomas differs from Kant in his understanding of what is grasped by the mind in the act of knowing itself.  Where for Kant we are apt only to receive representations of objects which are utterly detached from their un-cognizable transcendental objects, Aquinas is a realist about human knowledge.  He affirms that the forms of things present in the mind are actually the same as those present in the external objects: in knowing things, the object of our knowledge is not a representation or image in the mind, but the object itself.  The form is simply the means in the mind by which we know the thing exterior to us.  Given this extremely high view of the human capacity for knowledge, the fact that Aquinas places bounds upon that knowledge is all the more interesting.

So then, how do we know things, such that some things are unknowable?  According to St. Thomas, human knowledge is always mediated by the senses.  We cannot know without our knowledge being somehow traceable to experience.  Without experience, the mind is a blank slate.  This does not mean, however, that the mind can only know about things which are themselves available to our senses.  Rather, because we can know how things really are by means of our senses, i.e., because through abstraction from sense experience the real forms of things are imprinted on the intellect, we can also come to know the principles which govern things.  So, by observing corporeal things, we come to know principles of motion which govern them, and can thus deduce unobserved but necessary effects from observed causes, or vice versa.

What is essential in comparing Kant and Aquinas is the way the structure and principles of reality are related to things.  In Kant, nature is a kind of open field produced by prior mental syntheses, in which individual objects dwell with properties determined by that field.  For Aquinas, nature is primarily in things.  Time and space are not independent realities, nor are the laws which govern motion.  Instead, all the principles which we deduce from our observations of the world are ontologically rooted in the natures of individual subsistent beings.  Time and space are relative for both Kant and Aquinas, but the basis of their relativity is different.  For Kant, time is relative to the operations of the conscious subject.  For Aquinas, time is relative to the motion of objects.

St. Thomas adopts his notion of time from Aristotle, who defines it as the quantity of motion with respect to before and after.  Because Thomistic ontology is grounded in the actuality of individual things, and since only present things actually exist, we can properly say that only the present exists.  Time is a quantity, but not an extensive quantity like length.  When we speak of the quantity of motion, we do not have in mind the fourth-dimensional length or the duration of a successive synthesis of phenomena.  Rather we mean the extent of change undergone in an object between one actual state and another.  In a world like our own, filled with accurate, regular clocks, it is very easy for us to slip into a conception of time as some objective dimension in which we move (thus the popularity of fantasies about time travel, which are objectively absurd).  But we forget that our measurement of time is always simply the comparison of quantities of change between objects.  We think of clocks as absolute reference points, but this is fundamentally arbitrary, as shown by the regular observation that time is passing "slowly" or "quickly" depending on our own experience or observation of events.  

Given that time is not a static, objective framework, but a comparative measure of change among physical objects, the question of time's beginning takes on a different shape.  We are not looking for the bounds of a receptacle which determines and limits the possibility of real events.  Instead we are asking a question about the existence of mobile being itself and its nature.  Since our natural knowledge of the world is based on the received forms of material things, our knowledge of its beginning in time depends on the nature of material being.  Is it necessary that there was always something?  Is it necessary that things came to be?  

We do not have sufficient space to consider all of the objections and replies given in Ia q.46, though it is worth noting that Objection 8 in the first article is very close to Kant's argument for an infinite past, and the sixth objection of Article 2 is virtually identical to Kant's argument for a determinate beginning.  Aquinas's resolution to the apparent conflict of reason in this matter is, as with Kant, to remove it from the sphere of philosophical inquiry altogether.  But why does he think this question is irresolvable from natural reason?  Let's take up each side of the problem.

First, why can we not know that the world always existed?  On one hand, because the generation or un-generated-ness of things does not bear upon the possibility of their creation at some particular moment in time.  The fact that matter does not naturally arise from and collapse into non-being is not an actual demonstration that matter has always existed.  Rather, because any individual can come into being or cease to exist, by virtue of its contingency, we must conclude that it is very much possible for the world not to have been, prior to some point in time.  On the other hand, we cannot know that the world always existed, because the notion of a limit in time is (contrary to the objector) logically cogent.  Such a limit would not need to be determined on both sides (as Kant suggests), because time, as the measure of motion, would not exist before the existence of things.  As for the causal account of the instantaneous generation of things ex nihilo at the moment of creation, this can be attributed to the divine power.  God acts without motion, and thus the creation of the world in time is not on his part a temporally qualified act, since it was merely ordained eternally that the world should come to be in such a fashion.  He did not "precede" it except analogously or causally.  Thus the difficulty of moving from an empty time to the beginning of things is eliminated.

Second, why can we not know that the world began?  On one hand, because time, as a measure of change, is not an actual being, and thus is potentially infinite, just as any quantity is potentially infinite.  Time is the quantity of motion, and a measure of difference is always between two determinate points. But even without a beginning of time no two points will be infinitely removed from each other.  Thus we eliminate the apparent problem of non-traversable temporal separation.  

Again, there is nothing to prevent an object from having undergone an arbitrary amount of change, always capable of being increased, and likewise there is nothing to prevent the whole universe of objects from having undergone an arbitrary amount of change.  Because the past change undergone by an object is not actually present in its present being, the extent of this change or its duration has no direct bearing on the actuality of the present.  Thus the notion of a limitless past is not in itself incoherent.  

Nor, on the other hand, is a limitless past opposed to the intrinsic nature of things.  Generation is always from being to being, so that nothing about material being indicates the necessity of a beginning in time.  We can know that the world was created, but again, this creation is not a strictly temporal event and thus simply underlies the being of things as the constant reason for their existence.  It need not have been manifested temporally by a beginning, nor could the present state of things point back indubitably toward a first moment of being, since creation is totally outside the order of natural causation, and is not properly a temporal change, as we have already pointed out.  Thus, Aquinas concludes, nothing about our natural knowledge of material being can conclusively tell us whether the world began in time or not.

The comparison between Kant and Aquinas on this point is interesting largely because it is not widely recognized.  Everyone knows that Thomas Aquinas stands for natural theology and Immanuel Kant stands for the destruction thereof, and comparisons and responses on each side of the matter are quite common.  But it is delightful to see the two thinkers arrive at the question of the world’s beginning in such a similar way and, by observing their solutions, to come to know the characteristic methods and concerns of each in working out a proper understanding of the issue.