Pathways to PhilosophyKindle eBooks by G Klempner

on this page

Or send us an email

Application form

Pathways programs

Letters to my students

How-to-do-it guide

Essay archive

Ask a philosopher

Pathways e-journal

Features page

Downloads page

Pathways portal

Pathways to Philosophy

Geoffrey Klempner CV
G Klempner

International Society for Philosophers
ISFP site

Home   Lawrenz 1   Lawrenz 2   Lawrenz 3   Lawrenz 4

pathways (essays)

Jürgen Lawrenz

Kant’s Metaphysics:
End of the Road or Unfulfilled Future?


In regard to Metaphysics, Kant’s results were seemingly the opposite to what he strove to achieve, cf. the claim, in his Introduction, that “In this enquiry . . . I venture to assert that there is not a single metaphysical problem which has not been solved, or for the solution of which the key has not been supplied.” In the summing up of his Prolegomena, he records with evident pride in achievement: “Anyone who has read through and grasped the principles of the CPR . . . will look forward with delight to metaphysics, which is now indeed in his power.”

Yet the image of an “Alleszermalmer” persists, who dismantled the foundations of a philosophical edifice which had barely withstood the ravagement of Hume’s onslaught on its “occult fancies”! These discrepancies should make us wonder how one of the three greatest thinkers of all time could be so far deluded as to miss the outcome and import of his efforts! I propose to consider this problematic issue from a slightly different than ‘usual’ perspective.


To put the question of the suitability of metaphysics as a science is equivalent to asking, “what is metaphysical knowledge?”, hence “can metaphysical conclusions be verified?” In the CPR, this issue is encapsulated in the question, “How are synthetic a priori judgements possible?”

But before I address it, let me offer something as a curtain raiser:

The metaphysical possibility as well as the epistemology of newtonian absolute space remained a mystery until Kant solved it . . . [Guyer 10].

My intention here is to throw into focus the possibility of metaphysics having stood still in the interim; that no successor (as Kant indubitably expected) has taken up the cudgels and remoulded these fundamentals in the light of progress in the sciences. But I must hold this thought in abeyance until it’s time for my parting shots. Let me therefore begin at the beginning with a well-articulated statement of principle:

If one and the same faculty of reason is employed in empirical and metaphysical judgement, and the empirical employment of reason is legitimate, then so should be its metaphysical employment; and if metaphysics results in contradictions, then reason as a whole contradicts itself . . . Because the problem of metaphysics is ultimately a matter of reason’s relation to itself, the route to its solution, Kant argues, must also be reflexive. That is, reason must examine itself. [Gardner 21-2].

This identifies the claim by Kant to have wrought a ‘copernican revolution’ in philosophy. The whole perspective is rotated by 180û: not the world imposing its meaning, but meaning imposing on the world.


Knowledge and intuition

Kant’s first step, furnishing arguments in favour of the apriority of metaphysical cognition, is evidently mandatory. He shows initially that there is no inferring from veridical observations upon “the riddle of the universe”, while conclusions about what is are not necessarily divulged by causal entanglements between empirical phenomena. Kant in fact claims that the character of metaphysical knowledge is intrinsically a priori and synthetic along with mathematics, geometry and natural science. [B14-18].

In all these disciplines, recourse must be had to concepts. Kant points out that thinking is possible because we are able to employ concepts. No thinking in the absence of concepts! The obverse side of the coin is that concepts do not of themselves convey knowledge; a concept does not even vouchsafe existence to a thing that may be its referent. Therefore concepts need to be brought into conjunction with other concepts or predicates in a judgement, necessitating a clarification of the concept/predicate pairing.

Briefly, in an analytic a priori judgement, “the predicate is contained in the concept”, whereas in a synthetic judgement the predicate is not contained in, but added to, the concept. Now it is fairly plain that a concept+predicate conjunction is tautologous; accordingly Kant maintains that only synthetic judgements add to our knowledge:

There is a distinction in judgements, as to their content, by virtue of which they are either merely explicative, adding nothing to the content of knowledge, or ampliative, increasing knowledge: the former can be called analytic, the latter synthetic judgements. [Proleg §2].

The difficulty now is, how can synthetic judgements leave home base and transgress into the analytic part of the field?

The world, Kant asserts, does not present itself to us with any of its specification visibly labelled, therefore whichever predicates we choose can never be contained in the concept but must be submitted to a synthetic judgement. An exemplary proposition is, “every equilateral triangle is equiangular”. In that assertion we ascribe to triangles a property only found in some of them, whereas triangularity as such is contained in the concept. Hence we are dealing in the former case with an ampliative, in the latter an explicative, judgement.

The site where these skirmishes are being fought out is the Transcendental Logic. Now logic, which determines the conditions of valid inference, operates of its own accord on an a priori basis. Pure logic has no empirical principles, Kant writes [B78]. Should it prove possible on the same basis to acquire knowledge, then a means of verifying metaphysical concepts might offer itself.

Thinking alone, however, has no more prospects of disclosing knowledge than concepts alone. In contrast, sensory perception delivers indubitable knowledge; thus a burning topic moves to the fore, namely, whether the conditions under which knowledge is extracted from observations and experiences comply with compulsory features of perception. For those might then conduce to an a priori synthetic judgement being verifiable.

Now the conditions pertaining to the reception of ‘external’ data must be such that our senses and perceptive faculty are attuned in a special way to their interpretation. More specifically, perceptions must reflect reality; and this cannot be a simple one-to-one relation. The mind is a faculty, whereas percepts are objects of sentition with, to some extent, inscrutable features. Accordingly the impress of data must necessarily undergo transformation, to become representations in the mind. Kant’s pivotal insight here induced him to dismiss the notion of a reality mapped either (as in the rationalist’s universe) to a sort of ‘inner projection’ of perceivable rational order or (as in the empiricist’s conception) to a kind of tactility equivalent to causal impingement. If either of these notions held, then too much would remain unexplained, e.g. the plain insufficiency of our senses as well as their blatant selectivity.

Kant’s answer: it’s an ‘inside job’. His analysis of space and time serves to ground the all-important insight that spatiotemporality is a non-negotiable, structural feature of cognisability. It enables what could not even be thought of in its absence, namely the ‘where’ and ‘when’ of any object or event including myself. Basic to this insight is a notion of identity, this object, that event, my vis-à-vis to them, possible only because my mind is capable of apprehending a co-ordinate uniquely spatial and/or uniquely temporal in my relation to it. It is constitutive for perception as well as experience.

The “third factor”

This puts us en route to the subsidiary question which must be answered before the a priori synthetic hoves into sight, namely how objects-in-general are possible for us as relata in the perception/cognition manifold.

Thinking is not veridical in itself. Nor can concepts disclose knowledge by themselves. To achieve this it is necessary for the mind to combine concepts, for knowledge can only arise from comparison, evaluation etc. But even this is not enough — a “third factor” is required by Kant as the complement of concept and judgement, namely access to verifiable observation. [B194, 271-2]. For though it might seem inviting to rest content with the (interim) conclusion that “knowledge can only be obtained by judgements”, Kant rightly rejects this — it pertains only to knowledge, he says, whereas metaphysics pertains to judgements on objects-in-general or on all things taken distributively, e.g. causality or the principle of sufficient reason. Knowledge in the kantian metaphysic is not therefore primarily concerned with itself, but is an avenue towards the understanding of things, existence and being. [B18].

We already know that analytic judgements are barren of new knowledge; yet metaphysics is essentially about enlarging the horizon of our comprehension of what is. The relay station at which we are momentarily poised indicates that a solution must be sought by co-ordination of the faculties thus far passed in review:

(a) the necessary conditions under which our senses operate to deliver intuitions fit for conceptualisation and

(b) the necessary conditions under which our thinking results in knowledge.

To scale this last elevation, Kant discriminates between three kinds of knowledge, viz. through concepts, judgements and inferences. [B92-4]. Only by the second of these, he maintains, can knowledge be got. For as neither concepts nor intuitions suffice, his famous bon mot is powerfully cogent, “Thoughts without concepts are empty, concepts without intuitions are blind”. [B75].

So the core idea to emerge is facilitation of judgements. The Table of Categories is a device in that behest, its 12 subdivisions indicating the nexial entailments available for linking concepts together in a judgement. And the most considerable point to be made in respect of the connective concepts of the Table is that they are neither sensibilia nor empirical derivatives; and this leaves them unmistakably in the a priori basket. [B106].

Yet Kant insists that nonetheless they represent object features to which we are sensibly susceptible. [B145]. But this being the case, they are observables and thus qualify under both heads — perceptive and cognitive — for metaphysical service. The unity achieved by such concept linkage Kant terms the synthetic unity of the manifold. [B130].

Knowledge cannot be achieved in any other way. Kant employs the special term ‘apperception’ to denote this consciousness-bound activity and names the outcome a “synthetic unity of apperception”:

It must be possible for the “I think” to accompany all my representations. [B131 §16].

This is the central insight to which the whole analysis points. It is the raison d’être for his Table of Categories, which ultimately serves as a guarantee that the concept connected to a predicate is through the latter connected to a reality capable of vouchsafing its possibility, actuality or necessity.


Criticism has been levelled at the Table of Categories from the beginning. Hegel swept it away as an archaism, while Schopenhauer mocked its childish delight in symmetry and collapsed the dozen into three. Altogether the consensus has been that Kant seriously failed to convince us that his Tables omit nothing, and contain only, what is essential. Assuming this assessment to be correct, how damaging is it to Kant’s central concern?

Let me answer this by focusing on the philosophical issue. Mind and mind concerns are not, and cannot be, scientific problems — no exact, observational or reductive methodology has access to metaphysical problems. Hence Kant’s categories are not conclusive as principles of a scientific metaphysic. But in the larger context this means nothing more than a failure to ground a single set of principles as the only vehicle towards a metaphysical science. The defect of the Categories is not, therefore, fatal: it is mistaken only in its belief that is must be the ship which takes us to the land of metaphysical truth by cresting the waves of the scientific swell. Instead it showed that we can point its prow in another direction and still make landfall at our sought-for destination, where the principles stood revealed as concepts of necessity from concepts of possibility and vice versa.


That ship is the Transcendental Deduction, whose purpose is to unravel how a priori consciousness of observables can hold for any possible object of knowledge. It relies on our power to retain representations in memory and to combine these not only with fresh observations but also with other remembered representations. The faculty disposing of this power is the imagination.

Now it goes without saying that imagination is responsible also for fictions and illusions. But this is precisely where the categories still render indispensable service. For example, a judgement is true if observations on which it is based conform to the categorial criteria. Accordingly every object of knowledge must pertain to something other than itself, whether another object, a consequence, a limitation, or some qualitative or quantitative condition.

I pass over Kant’s geometrical illustrations as too far-reaching for this essay, except to note that their outcome is confirmation that metaphysics is indeed empowered to disclose such properties of things as must pertain to them in order for us to absorb conceptual knowledge through the agency of the categories. But any such knowledge can be attained only by judgements which are both a priori and synthetic; while recourse to the categories enables those metaphysical principles which pertain to the judgements to be made explicit.

This line of argument culminates in the claim that synthetic a priori judgements are verifiable (true) in one of two ways: (a) either by pertaining to the necessary conditions under which sensory perception operates or (b) the necessary conditions which hold as far as our ability to think is concerned. Inevitably, however, the logical concomitant to verification is ultimate recourse to empirical knowledge; and this must raise our concern if this is not a debilitating restriction.


Kant’s discussion of time and space serve inter alia the purpose establishing that they are not empirical concepts, but an a priori intuitions. So here is knowledge of the most indubitable kind that is impossible to derive empirically. But they facilitates the collection of empirical knowledge. [B40, 64-5]. Now these considerations allow, for all their exemplariness, of generalisation, for the point is, throughout, to make a case for the possibility of a priori synthetic judgements and thereby fulfil the brief of the whole Critique’s basic premise:

If we are to judge synthetically with a concept, we must go beyond this concept and appeal to the intuition in which it is given. For should we confine ourselves to what is given in the concept, the judgements would be merely analytic . . . But I can pass from the concept to the pure or empirical intuition corresponding to it, in order to consider it in that intuition in concreto, and so to know . . . what are the properties of the object of the concept. [B749]

From whence the conclusion (a) that metaphysics must rely on a priori synthetic judgements to acquire knowledge of objects-in-general; (b) these judgements cannot be verified except by recourse to pure intuition; and (c) intuition by itself is unfit to represent objects-in-general. From here to the issue of proof:

As to the mode of proof . . . what now is left us? The possibility of experience as a knowledge in which all objects must in the end be capable of being given us, if the representation of them is to have objective reality for us. In the third factor . . . we have found rules of synthetic unity a priori . . . [B264]. No synthetic proposition can be made from the categories alone . . . We are able to prove it . . . only as a principle of the possibility of experience, and therefore of the knowledge of an object given in empirical intuition. [B289]. Through intellectual concepts pure reason does, indeed, establish secure principles, not however directly from concepts alone, but always only indirectly through relation of these concepts to something altogether contingent, namely, possible experience. [B764].

The one criterion of surpassing importance in this is that Point (b) above contains the clue to verifiability: that from pure intuition the synthetic judgement receives the warrant that a condition pertaining to empirical truth prevails. So that Kant’s various assertions boil down to this:

The synthesis of representations rests on imagination; and the synthetic unity of it, which is required for judgement, on the unity of apperception. In these three, therefore, we must look for the possibility of synthetic judgements; and since all three contain sources of a priori representations, they must also account for the possibility of pure synthetic judgements. [B194].

The synthesis of all variety is the work of imagination. This is the ladder on which Kant at length mounts the platform of his general solution. For imagination is the gateway into the possibility of experience [A120], where synthetic judgements do present themselves for verification:
What the schematism of the intellect effects by means of the transcendental synthesis of imagination is simply the unity of all the various matters of intuition in the inner sense . . . The schemata of pure intellectual concepts are thus the true and sole conditions for furnishing them some reference to objects and therefore some meaning. [B185].

Imagination may strike Kant’s readers as a highly precarious membrane to hold up such weighty freight as metaphysical judgements. But his point is that any empirical object of given shape, dimension and other attributes of whatever kind constrains the imagination to the same representation as is held by intuition. Accordingly the judgement we might perform on that object or its attributes can be subjected to test by the actual representation of the object; and our belief in its verifiability is enhanced by our a priori knowledge of the necessary conditions by which imagination is constrained in the delivery of its confirmation. Moreover, in the imagination we possess a faculty capable of ascertaining these features by projection into intuition independently of confrontation with the empirical object. “The possibility of experience,” writes Kant, “is what gives objective reality to all our a priori knowledge.” [B195].

But this conclusion — that metaphysics deals securely only with empirical data and objects — essentially spells out that for all its differences of aims and methods, it has no mandate to reach further than, and beyond, science. And this seems precisely the critical issue — what need for metaphysics in those circumstances?


I’ll stick my neck out to indicate by a few strokes what I believe to be salient clues.

1. Kant was steeped in science, lectured and wrote on physics and evidently had complete faith in scientific truth and its revelatory power. Surely his rehabilitation of metaphysics is grounded in the conviction that the ways of metaphysics harbour a greater potential than science.

2. Since Kant’s day, however, scientific progress has proceeded at such tumultuous pace that philosophers increasingly lost touch with it, even to the extent of remaining incognisant of fundamental shifts with immense potential to a rejuvenation of the metaphysical impetus.

3. Conversely, scientists, of necessity mentally cocooned in their respective methodologies, tend to cling to various forms of naïve realism or positivism or verificationism, none of which are philosophically watertight and an all-too-easy prey to attacks by sceptics.

But if we look at the state of science today, it is not difficult to fathom the sort of challenge Kant might have relished. Concepts like the continuum hypothesis, Feigenbaum number, Mandelbrot set, string theory, Brusselator, residual electric potential etc. all have in common that they are scientific objects. But as a philosopher, you might do more than wonder. You might question what exactly scientists are doing with these concepts. And when you look closely, you’ll find them to be speculating and theorising about their meaning — an endeavour which science hijacked from philosophy, even though it is singularly ill equipped to handle them, for it brings into these pursuits the fallacy of looking to objects for the source of their meaning and to constitute their own being. The subject of cognisance is left out in the cold.

Our cognitive experience as well our perceptual horizons have been immensely enlarged since Kant’s day. Perhaps I could induce you to read my ‘curtain raiser’ again, before carrying on. For the point is, that the above concepts clearly demand to be treated philosophically, being related to fundamental issues in understanding and intelligibility, raising perplexing problems concerning essence and substance, time and space, epistemic and ontological status, and casting a new light on the whole question of being. Their instrumental facets are nothing more than thin slithers of accidental conjugations of ‘facts’ arising out of algorithms designed with other goals in mind. In short, what stands revealed with these concepts are mappings into products of enquiry which are as evanescent as the famous smile of the Cheshire cat — species of noumena unsuspected before they emerged from the nether realm of I-know-not-what and whose essence cannot be grasped by the resource of concepts of reality which science projects into them.

In a word, they are all and sundry metaphysical problems. As such, they could not have been foreseen by Kant or anyone else; but when I surmised above on Kant’s belief that the ways of metaphysics harbour a greater potential than science, it was with these developments in mind.

It is easier, of course, to write down such a claim than to substatiate it. Allow me therefore to conclude with some hints on those possibilities:

1. The concept of Infinity needs desperately to be updated from its inadequate foundation in scholastic thinking.

2. The concept of Complexity has a purely metaphysical aspect not thus far taken into the fold of philosophy.

3. Determinism in its manifold guises is becoming an irrelevancy, but this has not seeped through to full philosophical acceptance.

4. The concept of Anentropy is a capital metaphysical issue [cf. my paper on “Death”, Pathways ref]

5. The concept of Simultaneity has not (despite vocal advocacy by Hans Reichenbach) become the metaphysical issue it ought to be.

A mere smattering, but in my view a horizon beckoning for metaphysics by which to orient itself.


Kant, Immanuel: Kritik der reinen Vernunft & Prolegomena. Cited according to standard method.

Secondary texts

Adorno, Theodor W.: Kant and the Critique of Pure Reason. Polity Press, Oxford 2001.
Ameriks, Karl: Kant and the Fate of Autonomy. Cambridge UP 2000.
Dryer, P. D.: Kant’s Solution for Verification in Metaphysics. Allen & Unwin, London 1966.
Gardner, Sebastian: Kant and the Critique of Pure Reason. Routledge, London 1999.
Guyer, Paul (ed.): The Cambridge Companion to Kant. Cambridge UP 1992.
Heidegger, Martin: Kant und das Problem der Metaphysik. Vittorio Klostermann, Frankfurt 1973.
Janaway, Christopher: Self and World in Schopenhauer’s Philosophy. Clarendon, Oxford 1989.
Pinkard, Terry: German Philosophy 1760-1860. Cambridge UP 2002.