Did Descartes argue in a circle?

Arnauld, within his objections to ‘Meditations on First Philosophy’, highlights what would come to be considered one of the most fundamental flaws in Cartesian reasoning; namely the evident circularity of reasoning from ‘Clear and Distinct perception’ to the existence of God, and vice versa. The problem has come to be christened ‘The Cartesian Circle’ and has lead to an abundance of philosophical discourse, both critical and defensive, on the subject. In this essay, I will seek to sustain the line of argument that Descartes is initially guilty of the charge of circularity levelled against him which he fails to escape within his reply. Though many have since proposed methods of overcoming the circularity issue which in themselves might seem cogent, ultimately they can only be achieved by the alteration of Descartes’ original argument or a forced interpretation of the text thus failing to demonstrate that Descartes did not, in fact, argue in a circle.

Before engaging in an analysis of the circularity problem it is perhaps worth briefly summarizing the steps which Descartes’ takes to arrive at his conclusion. The meditator having begun his contemplation doubting all that he knows and perceives, he eventually recognises the indubitableness of his own existence; Descartes’ famous ‘cogito’ reasoning states that he has to exist because he is thinking, the fact that he is thinking is evident from his doubting. Hatfield formally expresses how the argument subsequently unfolds: ‘1) I know with certainty that I am a thinking thing. 2) This knowledge is based solely on a clear and distinct perception of its truth. 3) Clear and distinct perception would not be sufficient to yield such knowledge if it were in any way fallible. 4) Therefore, clear and distinct perception provides a sufficient ground for knowledge; whatever I so perceive is true. ‘ (Hatfield, G, Descartes and the Meditations, chapter 5, p.144) Though Descartes has now established the basic reliability of his perceptions, they are still potentially open to doubt until they can be guaranteed by a non-deceptive God. The existence of this non-deceptive God can only be concluded by appeal to the proofs of the intellect which his existence supposedly validates. Cottingham summarizes the problem succinctly: ‘I need to trust my intellect in order to prove God’s existence, yet without prior knowledge of God’s existence I have in principle no reason to trust my intellect.’ (Cottingham, J. Descartes, pp. 66-70) Herein lies the problem of the Cartesian Circle. The method with which Descartes argues means that he never actually removes the doubt from any of his claims; we can doubt our clear and distinct perceptions (albeit only in a ‘slight’ and metaphysical’ way) which means we have to doubt the existence of God since we cannot trust the intellectual method we used to arrive at his existence. Yet, paradoxically, his existence is the only thing which could remove the doubt. ‘A particular method of ascertaining the truth (clear and distinct perception) is vindicated by proving that God exists and is no deceiver, but this proof relies on that very method.’ (Hatfield, G, Descartes and the Meditations, chapter 5, p.169) As a reader would naturally interpret it in the Meditations, therefore, the argument is almost certainly circular. I will now move on to consider how Descartes himself endeavours to escape the circle. In response to Arnauld’s criticism, Descartes appeals to the difference between what we clearly and distinctly perceive in the present and what we remember having perceived on a prior occasion. He argues that ‘…we are sure that God exists because we attend to the arguments which prove this; but subsequently it is enough for us to remember that we perceived something clearly in order for us to be certain that it is true. This would not be sufficient if we did not know that God exists and is not a deceiver.’ (Hatfield, G, Descartes and the Meditations, chapter 5, p.169) I think it is worth questioning whether this is actually a reply to Arnauld’s circularity worries; what Descartes seems to be implying here is that circularity doesn’t actually matter because all the existence of God does is allows the meditator to confidently rely on clear and distinct perceptions he is no longer having i.e. perceptions he only recollects. One can rely on current clear and distinct perceptions. Firstly, if, indeed, Descartes is saying what he seems to be saying, why exactly was there any need to doubt in the first place? As Hatfield notes, ‘this reply makes it seem as if the reliability of clear and distinct perception was never itself really placed in doubt ‘(Hatfield, G, Descartes and the Meditations, chapter 5, p.170), which, of course, it was. Descartes seems to have changed his tune, so to speak. Consequently, the reply doesn’t appear to tackle the circularity which Arnauld initially pointed out because Descartes has moved the goal posts. However, granting that he hasn’t and that this is the way he wishes, and has always wished, to argue, there still seems to be an issue; we still have to rely on God’s existence at some stage in order to trust our recollected perceptions and we still cannot prove his existence without presupposing the reliability of our clear and distinct perceptions. It seems that in order for the reply to work, we have to be able to prove God’s existence without clear and distinct perception. I don’t think that this is possible within Descartes’ framework as I now hope to demonstrate.

Descartes postulates two arguments towards the existence of God which, by virtue of their being arguments, already presume to trust the human intellect. However, aside from this fact, the arguments seem weak and thus unable to prove the existence of the God necessary to secure our certainty of recollected perceptions. Within the third meditation, Descartes explores the nature of ideas; he maintains that ‘there must be at least as much reality in the efficient and total cause as in the effect of that cause.’ (The Philosophical Writings of Descartes, trans. by Cottingham et alii. Meditation 3, P.28)The idea of a perfect, infinite God could not exist in us because we are finite; yet, the idea does exist in us. There must be a cause of the objective reality found in the idea of him; that cause would have to be God. However, in order for this argument to hold, the idea of God would have to be innate which does not account for the multitude of people who do not find this idea of God within them at all, or even the idea of a perfect being. In addition, there is nothing to suggest we could not create this idea of a perfect being from our own limitations i.e. imagine a limitless being. Descartes might argue that we cannot derive the idea of perfection from imperfection, presumably because we could not know limitations without the idea of perfection. One can easily conceive of us creating an idea of perfection, however, just by comparison with other, better humans and extending this to an extreme. Moreover, we might be unable to recognise a perfect being; what we think is a perfect being could be imperfect but simply more perfect than us. It seems, then, that this is a rather weak argument for God’s existence lessening our ability to use God to validate our recollected clear and distinct perceptions.

Descartes provides a second proof for the existence of God in the form of his version of the Ontological argument. The central idea is that necessary existence is part of the definition of a perfect being, the idea of which the meditator clearly and distinctly perceives. Therefore, God must exist. The most evident flaw in the proof rests in the fact that existence is not a predicate; if it were, anything could be brought into existence. Anything we could conceive of as perfect in our minds would necessarily exist and this is definitely not the case.

It seems that even if Descartes’ proofs for the existence of God were not as weak as they are, his reply against Arnauld’s circularity criticism still wouldn’t hold. Both arguments still appear to be relying on the clear and distinct perception of the idea of God which invites the return of the circle. Otherwise, how are we perceiving this idea? It could be argued that the idea of God comes from the ‘natural light’ (‘the intrinsic cognitive power found in all human minds'(Hatfield, G, Descartes and the Meditations, chapter 5, p.156)) which is not open to any doubt. However, if we could perceive things by this ‘natural light’, is there any need for the deceiving-God hypothesis? Hatfield also draws our attention to the fact that the natural light and clear and distinct perception are two different phrases representing the same idea: ‘…the cogito reasoning is seen by the natural light. Earlier, the same conclusion was attributed to clear and distinct perception. It therefore seems that the natural light and clear and distinct perception are the same thing described in to different ways.’ (ibid. p.157) One couldn’t be used in place of the other in this case. Descartes’ reply is also very limited if we can only clearly and distinctly perceive things in the present without God; as soon as we stop concentrating on the clear and distinct perception, it becomes a perception we have to recollect and we need God again. Hatfield is keen to mention, on behalf of Arnauld, that Descartes’ reply also does not seem to address the issue of the truth behind our clear and distinct perceptions: ‘whether or not we can doubt them, they might still be false. In that case, we should want a proof of their validity that does not rely on clear and distinct perception. It is that proof that Arnauld rightly says Descartes has not supplied.’ (Hatfield, G, Descartes and the Meditations, chapter 5, p.171)

To reiterate, then, Descartes’ reply to the circularity challenge states that we can clearly and distinctly perceive things without God but God is needed to remove our doubts about recollected perceptions. Most of our clear and distinct perceptions will be recollected since we cannot be clearly and distinctly perceiving things at all times. Therefore, the arguments for God’s existence have to be sound and cannot appeal to clear and distinct perception without falling back into circularity. The arguments which Descartes presents for God’s existence are either too weak to be convincing or appeal to clear and distinct perception so his reply is a weak one.

Since the publication of Arnauld’s circularity concern, scholars have attempted to interpret the meditations in such a way as to relieve Descartes of the charge of circularity or, at least, to interpret it in such a way that the challenge doesn’t matter. For example, some have suggested that by altering the way in which we view the intention behind the meditations, the circle can be avoided. If Descartes only intended to achieve maximum certainty as opposed to truth then his reply seems more generally adequate; we are certain of our clear and distinct perceptions until we stop having them at which point the doubt of the deceiving-God hypothesis can creep in. If we are certain/convinced of the proofs for a non-deceiving God’s existence then his deception is no longer a worry and we can be maximally certain of our perceptions. ‘We have not shown that clear and distinct perceptions are true and so have not shown that the proofs of God are true. But we have shown that they are maximally certain, thereby reaching our goal of unshakeable belief.’ (Hatfield, G, Descartes and the Meditations, chapter 5, p.171) It does seem, however, that Descartes aims towards truth on numerous occasions: ‘if I were unaware of God; and I should thus never have true and certain knowledge…’ represents just one example. He talks of knowledge and of truth. Loeb suggests that ‘knowledge, in the strict sense of scientific knowledge, is identified with unshakeable belief…’ (Loeb, L. ‘The Cartesian circle”, in The Cambridge Companion to Descartes, p.203) which would suggest that the idea of aiming for certainty is perhaps not such an unnatural interpretation of Descartes. Arguably, however, all this can do is lessen what the meditations seek to achieved; it doesn’t solve the problem, it simply dilutes Descartes’ intention until the problem no longer exists.

Another potential way of removing the circularity problem is removing the doubt in the first place. Hatfield offers a description of what this would entail i.e. arguing that God cannot possibly be a deceiver because the notion of God being a perfect being and the notion of him being deceptive are logically incompatible; ‘the supposed reason for doubt is removed and the circle is (allegedly) avoided.’ (Hatfield, G, Descartes and the Meditations, chapter 5, p.174) However, the circle is only avoided because this is not Descartes’ argument. The fact that there is a logical contradiction within his conception of a deceiving God is a separate problem but within the argument which we call circular, Descartes entertains the idea of a deceiving God. Therefore, though it may be able to remove the circularity problem, it doesn’t disprove the fact that Descartes argued in a circle. The ‘remove the doubt’ proposition also contains within it potential issues. For example, the idea of a perfect God could, in some way, be compatible with a deceiving God. (ibid.)

It suffices to say, by way of conclusion, that Descartes argues in a circle. He is guilty of the initial challenge postulated by Arnauld and then subsequently fails to provide an adequate response to the criticism. Though it is worth considering scholarly attempts to remove the circularity problem, these are only arguably useful when attempting to examine whether one could make the Cartesian reasoning work as opposed to whether Descartes made it work. Furthermore, it does seem that these propositions often rely on either an interpretation which doesn’t seem entirely as intended or a distortion of the argument into something which doesn’t necessarily resemble the original.