Kant and the ontological argument


Immanuel Kant (1724-1804) is generally regarded as the definitive critic of the “ontological or Cartesian argument”  as he termed it. He is also one of the most important thinkers in the Western tradition, and one of the most difficult. Authors of books on Kant usually take refuge in this, and claim that they are only trying to say what they think Kant said or meant! Be that as it may, Kant’s view of the ontological argument can be stated quite briefly.

Taking Descartes’ contention that, “… because I cannot conceive God without existence it follows that existence is inseparable from him, and hence that he truly exists …”, Kant agrees with the first part of this claim, but denies that which Descartes thinks follows from it. Kant accepts that if we want to talk or think about the God that Christians appear to believe in, then we are correct in claiming that we must think of that God as necessarily or self-existent. As Descartes says, we “… cannot conceive God without existence”, and “existence is inseparable from him”. The concept of the Divine entails a necessary link between essence and existence.


But this is all true at the level of definitions. The truth, Kant says, is “analytic”. The claim that God actually exists, the assertion that there is a God, does not follow necessarily from the definition of God as necessarily existent. The assertion of faith in God is possible, Kant thinks, but it is made on other grounds and in a different way. How does he reach these views?

Kant’s life

In view of the fact that Kant’s thought is as influential as it is, it is perhaps surprising that his life was quite mundane. Born in Konigsberg, then in East Prussia, now Kaliningrad (and in the Russian Federation), Kant was brought up in a family committed to Pietism, a strictly ethical Protestant sect with an emphasis on moral duty as opposed to dogmatic or doctrinal interests. He attended the University of Konigsberg and then worked as a tutor to two aristocratic families (1746-1754). In 1755 he became a lecturer at the university, and he spent the next 15 years teaching a variety of subjects and writing a number of books on cosmology, aesthetics, logic and philosophy. None of these works were thought especially significant at the time, although some of his scientific writings were ahead of their time. But in due course, in 1770, Kant became Professor of Logic and Metaphysics, still at Konigsberg. Up until this time Kant’s outlook was in keeping with the conventions of the day. He worked according to the rationalist tradition arising from Descartes and Leibnitz and with particular reference to Christian Wolff, in whose thought it was assumed that all things, whether open to experience or not, were rationally comprehensible and thus legitimate subjects for analysis and investigation. This dogmatic rationalism was shattered in 1781 when the 57-year-old Kant published the first of a series of critical works, Critique of Pure Reason, (Hereafter CPR). Kant continued to teach at Konigsberg, whilst writing the other books for which he is now famous, until retiring in 1798/9. He died in 1804.

But what brought about the change, the revolution, in Kant’s thinking?

The influence of Hume

Kant makes it plain in his own writing that he was woken from his “dogmatic slumber” through his reading of the work of the Scottish philosopher and historian David Hume (1711-1776). Hume had propounded a pure and sceptical style of empiricism, claiming that so far as we could ever tell, reality was no more and no less than an accumulation of observable facts and events with no overall or intrinsic structure or determinant that we can know in the accepted sense.


Hume’s case was based on his reliance on two fundamental criteria for truth or knowledge: first, there was “abstract reasoning” – later termed “analytic” truths by Kant.  Here Hume had in mind matters of mathematics, logical certainty, self-evident truths and the like, within and for a given system of symbols or relations. Secondly, there was “experimental reasoning” on matters “of fact and experience” – “synthetic” truths for Kant – matters that were arrived at and verified through practical experience, experiment and testing.


Even with these criteria Hume seems to have held that in fact only a limited degree of truth was actually obtained. Reason, he stated, “is, and ought only to be the slave of the passions, and can never pretend to any other office than to serve and obey them”. “Relations of number” and logical truths were of value only within a limited, formal and perhaps somewhat abstract context, and truths of experience were immediate and fleeting and thus good only in relation to the situation or circumstance in question. Hume did not believe that our empirical knowledge was ever more than provisional in character and he was a stern critic of any who claimed to have advanced upon this position. He was, for example, sceptical about the concept of causal associations, preferring to think that the instances of cause and effect that we think we perceive are only as real as they seem to be to us. They have psychological power and associations in the mind in terms of habit and expectation are mistaken as evidence of patterns of actual causality.  He did not think that we had means to justify the generation of laws or principles of cause and effect, in which it was claimed that there were necessary connections between events. Hume’s was an affirmation of emotivism and utility in ethics, and as such a humane and liberal philosophy. But it generated a severe critique of religion, or of theology, as usually practised. In a famous passage from his Enquiry Concerning Human Understanding, (1748), he writes that,


“If we take in our hand any volume: of divinity or school metaphysics, for instance; let us ask, Does it contain any abstract reasoning concerning quantity or number? No. Does it contain any experimental reasoning concerning matter of fact and experience? No. Commit it then to the flames: for it can contain nothing but sophistry and illusion.”1


All of this made an impact on Kant. He took seriously the criticism of metaphysics and of science and the agnostic naturalism that Hume had presented, and he developed views along the following lines: human reason had a definite but limited function. It is tied to and limited by the finite world of experience. It is only in relation to phenomenal reality that reason operates, and it operates because of the inbuilt apriori categories of time, space, and cause. Reason, according to Kant, breaks down irretrievably if it attempts to transcend the realm of finite experience, and so the traditional metaphysics and theology cannot be counted as significant or as legitimate.


Kant was not, however, in the least dismissive of the religious concern. This point is examined briefly in a later section of this paper, but here we can note that Kant’s interest is indicated by this opening to CPR:


“Human reason has this peculiar fate that in one species of its knowledge it is burdened by questions which, as prescribed by the very nature of reason itself, it is not able to ignore, but which, as transcending all its powers, it is also not able to answer.” (CPR p7)2


Kant’s meaning is that the religious/metaphysical concern is one that arises inescapably and which cannot really be neglected by reason. Yet reason cannot answer these matters. This doesn’t worry Kant, however, because he believes that there is a deeper level of insight, reason, or intuition by means of which it is possible to make a transcendental, a truly religious, analysis of reality – and this comes (as all good pietists know), through ethical experience and action. However, this is important in relation to a consideration of Kant’s approach to the ontological argument: Kant criticises the argument because it is a part of a style of thinking that he considers presumptuous and mistaken. But from the religious point of view he was, he thought, “denying knowledge in order to make room for faith”, and here we may note an affinity between Kant and the scholar whose reasoning he, by implication, so severely questions, Anselm.


Kant’s discussion of the ontological argument

Kant, unfairly perhaps, seems to regard the ontological argument as an attempt to prove formally that God really exists via a statement and explanation of the concept of God. This does not do justice to what seems to have been Anselm’s purpose in writing, viz, within and for the community of faith. But Kant seems to have known the argument from the presentation it received in the work of Descartes, and it has to be said that Descartes’ exposition is rather less devotional in character than Anselm’s.

Descartes takes the view that God may be defined as the supremely perfect being, and necessarily existent, since such existence is a perfection to be included in that of the Divine. Descartes uses an analogy with the triangle3 to suggest that it would be as absurd to deny the existence of God as it would be to deny that there could be a triangle without three internal angles.


Kant begins his discussion by asking whether this is a fair analogy.


He can agree that in a pure, abstract sense existence, or rather necessary existence, ties in with the concept of God, this in a way analogous to the association between a triangle and three angles. In both cases there is a clear, necessary, analytic link between subject and predicate. In this sense, as we noted in the preface above, Descartes is right to maintain that we are not free to “conceive of God without existence”, any more than we are to think “triangle” without three internal angles.


However, this is all true at the abstract level of definitions. It only means that if we think of the concept “God”, and if we are intending to think about the God of traditional theism, then we include in that thought the idea that this God possesses necessary existence, aseity. In just the same way, if we think of a triangle we are bound to think of its three angles.  Arguments of this sort are theoretical elaborations of what a given concept implies, includes, or entails.


What Kant questions is the way that Descartes presses this sort of argument to the end of claiming that it provides a demonstration that this “God” actually exits and is real. Kant would allow that the argument suggests that if there were a God then he would possess necessary existence, but he thinks that that is all that the argument can do and that no contradiction is involved in the denial of the actual existence of God.


Kant’s point is that we can only fall into contradiction if we try to separate subject and predicate in a given concept:


“If, in an identical proposition, I reject the predicate while retaining the subject, contradiction results; and I therefore say that the former belongs necessarily to the latter. But if we reject subject and predicate alike, there is no contradiction; for nothing is then left that can be contradicted. To posit a triangle, and yet to reject its three angles, is self-contradictory; but there is no contradiction in rejecting the triangle together with its three angles.” (CPR p502)


Kant thinks the same holds true over the idea of God. Given what the idea of God generally entails, it would be a contradiction to try to maintain the idea whilst denying the aseity of the Divine. But there is no contradiction in rejecting both subject and predicate together in or as the judgement “There is no God”. When we make this judgement we “reject the thing in itself with all its predicates and no question of contradiction can then arise”. (ibid)


Kant thinks that this is all as clear as clear could be, and he thinks that if those who have propounded the ontological argument, and of course he means Descartes primarily, had not misunderstood the relations of subjects to predicates, there would never have been the difficulty that there had been over this matter.


To emphasise the point Kant considers the subject-concept “God” in relation to the predicate-concept “omnipotence”: the latter has a necessary relation to the former, “omnipotence” is an analytic predicate of “God”, and it is included in the subject-concept. There would be an obvious contradiction then, in a) positing “God”, ie, the concept of God, whilst b) denying his omnipotence. But “… if we say, ‘There is no God’, neither the omnipotence nor any other of its predicates is given; they are one and all rejected together with the subject, and there is therefore not the least contradiction in such a judgement”. (ibid)


The “there is no God” reference recalls the debate between Anselm and Gaunilo, who had written on behalf of “the fool who says in his heart that there is no God”. The recollection is fair enough, since Kant has re-worked the sort of objection that Gaunilo had in mind to the logic of the “ontological reasoning”, to the transfer that seemed to be made from the realm of definitions to the world of reality.  Kant is aware of the fact that the defender of the ontological argument will want to press the point that Anselm made in his reply to Gaunilo and that Descartes notes in both the Third and the Fifth of the Meditations, ie, that the concept of God is unique, and therefore not subject to the criticism that Kant has made. It might be said, Kant suggests, that the concept of God is of that which possesses all reality, that such a concept is not self-contradictory and so it is fair to allow that we are justified in assuming that such a being is possible, and since “all reality” includes “existence”,


“… existence is therefore contained in the concept of a thing that is possible. If, then, this thing is rejected, the internal possibility of the thing is rejected – which is self-contradictory.” (CPR p503)


Kant denies that he has fallen into contradiction. He takes the view that it is those who have presented the ontological argument who have erred, for they imported, “… the concept of existence … into the concept of a thing which we profess to be thinking solely in reference to its possibility”. Kant’s point is that the concept of existence as used by Descartes, in respect of the concept of God, is misleading when it is also used in the attempt to support the claim that God actually exists.


We may note here that Kant’s suggestion that Descartes (and Anselm by implication) were thinking about the “possibility” of the concept of God, is not an accurate explanation of Anselm’s “fides quaerens intellectum” method.


Kant is thus led into a further examination of the analytic/synthetic distinction. He holds, as we have seen, that analytic propositions or judgements are ones in which, for example, subject-concepts entail necessarily certain predicate-concepts, which it would be contradictory to deny.  Synthetic propositions are ones reflecting judgements or opinions arising from experience. And so Kant asks, “Is the proposition that this or that thing … exists, an analytic or synthetic proposition?”


With reference to the statement that “the perfect being exists”, the Kantian view is that if this is taken as an analytic proposition within the context of an elaboration of the concept of God, then it is true within that context and it would be a self-contradiction to deny it. This is clear from the points noted earlier. If we consider the proposition in an open context as analytic, then Kant thinks that it is again true. But that is so only because it is “a miserable tautology”, it simply means that the perfect being is a being. As John Kemp comments,


“… if there is a perfect being, it is of course tautologous to say that it is a being, and even if there is not, it is still tautologous to say that if there were a perfect being it would be a being.”4


Kant’s point is that we are not able to make any claim about the actual existence or reality of the Divine in consequence of matters that are established at the abstract level of analytic definition. Analytic truths can obviously reflect and defend concepts and understandings, but as we have seen, those concepts, subjects and predicates, can be rejected at the practical level. This confirms that the statement “… the perfect being exists …” is a synthetic judgement, an existential or experiential proposition. As such it is a reflection of practical experience and it is open to debate and to scrutiny such that a denial of its verity is not contradictory in the self-contradictory sense!

Existence as a predicate

Like a terrier in pursuit of a rat, Kant proceeds to what is now his best-known criticism of the ontological argument. The great illusion of those who defend the argument is the belief that “existence” was a real as well as a logical predicate. As we have seen, Kant is happy to accept that existence can function as a logical, formal, hypothetical predicate, at the abstract level of definitions. But he will not allow that existence can ever be a real predicate. He claims that no concept is actually effected by the predication of existence, since to claim that a concept has existence is not to say anything in addition to the concept, it is to assert that that concept has a referent, that there is an object, state, relation or whatever, to which the concept in question applies. Not only is this not an addition to the concept proper – though we may say that it extends our understanding of how the concept might be understood, it isn’t and couldn’t be a proof or demonstration of the real existence of the concept in question. It is simply the case, says Kant, that “being” and/or “existence” aren’t real concepts of something that could be added to the concept of a thing, “It is merely the positing of a thing, or of certain determinations of things, as existing in themselves.” (CPR p504)


As an example Kant again takes the statement “God is omnipotent”, in which the predicate-concept “omnipotence” is related to the subject-concept “God”. In the proposition “God is omnipotent”, the word “is” serves,


“… to posit the predicate in its relation to the subject. If, now, we take the subject (God) with all its predicates (among them omnipotence), and say … ‘There is a God’, we attach no new predicate to the concept of God, but only posit the subject itself with all its predicates, and indeed posit it as being an object that stands in relation to my concept.” (ibid)


The famous illustration of the 100 real versus the 100 imaginary thalers (or pounds or dollars!) follows. To show that the assertion of existence adds nothing to the concept of the thing in question, that it simply expresses the belief that there is an actual referent for that concept, Kant argues that whatever the practical difference between 100 real and 100 imaginary dollars, the concept of “100 dollars” is the same in both cases: “… the real contains no more than the merely possible.” And Kant concludes thus:


“By whatever and by however many predicates we may think a thing-even if we completely determine it-we do not make the least addition to the thing when we further declare that this thing is. Otherwise, it would not be exactly the same thing that exists, but something more than we had thought in the concept; and we could not, therefore, say that the exact object of my concept exists. If we think in a thing every feature of reality except one, the missing reality is not added by my saying that this defective thing exists. On the contrary, it exists with the same defect with which I have thought it, since otherwise what exists would be something different from what I thought.” (ibid)


It is Kant’s belief that the term “existence” should be restricted in its use as a predicate because whatever its logical status, from the practical point of view it cannot function in the role that it has been assigned in the ontological argument.

Positive elements in Kant’s analysis

In the preface to the second edition of CPR, (1787), Kant discussed the extent to which the critical philosophy had a negative character for the scope of human reason, human possibility, and religion. From the treatment of the ontological argument, it might seem that there can be little of positive worth for religious philosophy in the Kantian analysis, but such a judgement would be to misunderstand Kant’s purpose. We have seen in our brief review of his general philosophy that Kant believes that we cannot, via reason and experience, transcend the finite domain which is the context for that experience and for what we generally term “knowledge”. In that context we cannot obtain knowledge of “things in themselves”; we can only manage phenomenal knowledge. “Noumenal” experience, experience of the “thing-itself”, is however possible in one sphere – the moral or ethical – or as Kant often terms it, the practical. This is not the place to examine Kant’s ethical theory, nor the theological implications that it has. But it is with this in mind, plus the acceptance of the validity of the concept of the Divine at the level of definitions, and against the background of his view that the traditional natural theology was illegitimate – he says at one point that the history of thought shows that as a rule “men began where we should incline to end”, (CPR p666), ie, with the knowledge/question of God, that he claims in defence of the positive character of his approach that he has “found it necessary to deny knowledge, in order to make room for faith”. (CPR p29) It is, perhaps, debatable as to whether the critical point that Kant makes is wholly fair, but again it is worth noting a degree of affinity between Kant and Anselm on this matter.  If we recall the real extent to which Anselm was writing within and for the community of faith, and with a creative and expressive purpose that has the sort of analytic and synthetic validity that Kant should approve of.  Thus the logic of Anselm’s argument is fine as an analytic statement, and as he accepts that the argument presupposes and refers to the state of faith he should be happy to accept that the claim implied by the argument that “that than which a greater cannot be thought” must and cannot not exist is a synthetic judgement. This is akin to Kant who makes a denial of “knowledge”, but promotes an affirmation of faith via reference to a state of being – ie the life of good will in relation to the categorical imperative, on basis of which we find justification for faith in God.


The following explanation by Kant may be a help:


“To know an object I must be able to prove its possibility, either from its actuality as attested by experience, or a priori by means of reason. But I can think whatever I please, provided only that I do not contradict myself, that is provided my concept is a possible thought. This suffices for the possibility of the concept, even though I may not be able to answer for there being, in the sum of all possibilities, an object corresponding to it. But something more is required before I can ascribe to such a concept objective validity, that is, real possibility; the former possibility is merely logical. This something more need not, however, be sought in the theoretical sources of knowledge; it may lie in those that are practical.” (CPR p27n)


Can “existence” be a real “defining” predicate?

Finally, in this section, we want to reconsider one of the points made by Kant in his critique of the ontological reasoning of Descartes and, by implication of Anselm.  The points made here derive from original reflections on the point in question.


In the ontological argument as developed by Anselm and Descartes the essence of God is said to entail and imply the necessary and actual existence of the divine. God is “that than which a greater cannot be thought”, (Anselm) who “must and cannot not exist” in order to truly be “that than which a greater cannot be thought”. God is the “sum total of all possible perfections”, (Descartes) and necessary existence is a perfection God must thus possess and so it is not possible to think of God as not existing, thus he exists.


In the argument the concept of God entails the necessary existence of God, his aseity, as a defining predicate or quality, and it is said to follow from this that God exists actually.


“Necessary existence” is, as we have seen, a legitimate, indeed a necessary, defining predicate of the subject-concept “God”. The arguments of Anselm and Descartes are sound in this respect. But as valid arguments legitimacy is confined to the realm of definitions, to the level of understanding. Anselm and Descartes then press the point to the view that the subject-concept “God”, which “must and cannot not exist”, actually exists …


This is where the problems noted by Kant come to the fore. The logic of the connection between subject and predicate in the case in point is correct. In this sense necessary existence is a defining predicate of the subject-concept “God”. But when we assert that “God” (as defined) actually exists, we are making a claim of another sort, a synthetic claim, and one open to another form of verification.


It is evident, surely, that the truth of the logic in the definition of “God” as presented by Anselm and Descartes is clear within the context of the general understanding of the concept of “God”. But the actual existence of “God” does not follow from the logic suggesting that he should!


However, the sense, or a sense, of the reasoning in the ontological argument emerges from Kant’s claim that the assertion of the real existence of a given concept does not add anything to the concept as such, ie to our understanding of it as such.


Commonsense suggests that this view can be qualified. It seems that there are special cases where the existence of a given x entails an alteration to the concept of x, ie, cases where there is an alteration, not just to the understanding of the relation of a concept to the world, but an alteration of the understanding of the concept as such, that takes place in consequence of the experienced existence of the x in question.


Consider the following – a line of thinking which offers an original slant on a key aspect of the debate:


  1. Predication entails the assertion or affirmation of the qualities, attributes and features of a given subject or concept. The predication of the attributes of a given subject-concept thus gives a definition of that concept – the definition will represent what the subject concept is. Such definitions can have an analytic character, in so far as they have clarity and coherence for their subject which they unpack and lay bare.  Such definitions can be subjected to logical analysis.
  2. The assertion that a given concept “exists” involves “instantiation”. When we say that “So-and-so exists” we make a “that” statement, a synthetic judgement or claim that there are referents for such concepts in the world of experience. The purpose of such a statement is to express a view of the relation of the concept to the world, not to say anything about the subject as such. Such claims are therefore open to experimental investigation.
  3. Existence is not a defining predicate. A given concept doesn’t alter and gain a further predicate when we assert that it “exists”. The alteration that takes place is in our understanding of how the concept relates to the world – or of how the concept is believed to relate to the world …
  4. But all of this breaks down in the case of the concept x if x is the concept “mother” held by y with the view that the referent of the concept x is dead.


Let us imagine that this is so in the sense that x is believed by y to have died in childbirth. Thus y, has never had any direct conscious experience of x and has a concept of x based on other sources.


The breakdown comes when it transpires that y’s mother, the referent of the concept x, is alive and well and living in Blackpool. She did not die and is now en route to see her long-lost y!


In such a case the real existence of x alters the concept of x held by y …


  1. The alteration of the concept of x is not just an alteration of the understanding of the relation of the concept to the world. That has altered, but so has the basic concept of x as such as held by y.  This is an alteration, however, not a new concept.  The original concept develops organically via the encounter of y and x.


  1. Interpersonal encounter is the crucial element here. And the analogy with the state of faith in a God who is “never less than personal” (Tillich) is easy to see. In cases of encounter, encounters that have the character of personal exchange, the real existence of the “thou”, the other, is crucial to the definition/understanding of the concept of the self as well as of the “thou” in question.


  1. None of this vindicates the ontological reasoning as a persuasive proof of the existence of a necessarily existent God. But it assists in explaining how it is that those with “faith in search of understanding” come to the view that there is meaning and spiritual sense in the logic of the ontological argument. In the case of the actual, real existence of one’s mother altering the subject-concept “mother”, it is of course the actual, verifiable existence of one’s mother that allows the legitimate alteration of the concept as well as of one’s understanding of the relation of the concept to the world. And in the case of the person who has, as had Anselm, a faith that he could doubt but not not-have, it is that condition of faith that points to and is expressive of that which is and cannot not be.


1 Hume’s Enquiry, Selby-Brigge Edition p165.

2 All quotations from CPR are from the Kemp Smith edition.

3 See Descartes Op. cit p148.

4 J Kemp,  The Philosophy of Kant, OUP 1968, p50.







Past Questions

Disclaimer. Inducit Learning Ltd. is not responsible for any content outside of the pushmepress.com domain. If you are a rights holder and you think we have breached your copright, please email the editor and we will remove it.