The existence of God is crucial to the project of the Meditations in a number of ways. It is one of two claims (along with dualism) celebrated in the book’s subtitle – in which are demonstrated the existence of God and the distinction between the human soul and the body (p12) – which indicates its central importance. More significantly, the existence of God provides a bridge from subjective certainty to objective truth: with a good God established, Descartes can trust that his clear and distinct perceptions portray the world truthfully.
Descartes provides two main arguments for the existence of God, both arising in the context of an examination of his ideas. In both cases, the peculiar fact that he possesses the idea of a perfect being drives the argument. The Trademark Argument is concerned with the very existence of that idea; the Ontological Argument involves an analysis of its contents.
It is worth remembering that in both arguments the grounds for inferring God’s existence are clear and distinct perceptions. But clear and distinct perceptions are only guaranteed to be true by God’s existence. This problem is known as the Cartesian Circle and must be addressed if the arguments are not to fail through circular reasoning.
The so-called Trademark Argument for the existence of God arises quite naturally as Descartes attempts to catalogue the various ideas in his mind. The argument itself is simple – his idea of God can only be explained by the existence of God – but it employs a set of assumptions and principles inherited from earlier philosophy that are far from obvious.
The first aspect of the argument is Descartes’ reference to degrees of reality. Of course, this sounds strange to modern ears: things are either real or not. But for Descartes there is a strict hierarchy: substances are more real than their modes (things are more real than their properties), and infinite substances are more real than finite ones. The ‘reality’ of an object is matched by the ‘reality’ of its idea: a substantial thing causes a substantial idea. The object possesses ‘formal’ reality, and the corresponding idea possesses ‘objective’ reality (an unfortunate and counterintuitive use of the term ‘objective’). The objective reality of an idea is its representational content, and clearly big ideas can be said to have big content. This correspondence between a thing and its idea operates also with all the properties of the thing. So if I have an idea, say of the Taj Mahal, the idea is ‘of’ something, it has content. And the properties of the Taj Mahal (majesty, symmetry etc) are said to be formally present in the Taj Mahal, and objectively present in the idea.
With that in mind, Descartes asserts the first premise of his argument, an apparently self-evident rule that is generally referred to as the causal adequacy principle:
Now it is manifest by the natural light that there must be at least as much reality in the efficient and total cause as in the effect of that cause. For where, I ask, could the effect get its reality from, if not from the cause? M3 p28
The relation here is between a cause and an effect, whatever specific things those might be, and the principle asserts that a thing cannot be caused by something less substantial than itself. As Descartes elaborates in the Second Replies:
... it is just the same as the common notion ‘Nothing comes from nothing’. For if we admit that there is something in the effect that was not previously present in the cause, we shall also have to admit that this something was produced by nothing. And the reason why nothing cannot be the cause of a thing is simply that such a cause would not contain the same features as are found in the effect. OR2 p84
In the context of the Trademark Argument, the cause is the thing itself, and the effect is the idea of that thing. Any idea which possesses objective reality, must have been caused by something with at least as much formal reality. Descartes several times gives the example of ‘a very perfect machine, the idea of which is in the mind of some engineer’ (Synopsis p10f, see also OR2 p84 and Principles I.17). If the engineer was not smart enough to design such a machine, we would look elsewhere for its cause (perhaps he consulted with others, or copied it from a genius). Insubstantial ideas may well be the product of Descartes’ imagination: after all, he is substantial himself and has enough formal reality to generate ideas containing as much or less objective reality – perhaps he invented his idea of grass, say, or of the piece of wax which is a confused and imperfect idea, as he has shown. But if there was an idea that had more objective reality than Descartes possesses formal reality – a conception too substantial for it to come from Descartes himself – then that would require the existence of something other than Descartes.
The idea of God is such an idea. Descartes’ (rather traditional) idea of God is of an infinite, eternal, perfect, omniscient, omnipotent creator of the universe. These properties are objectively present in his idea of God, and as a result this idea contains maximum objective reality. It is easy to see why: God is simply more than Descartes in every point of comparison: God is infinite and perfect, Descartes is finite and imperfect. Since ‘there is more reality in an infinite substance than in a finite one’ (p31), the idea cannot come from Descartes alone, and can only be explained by the existence of such a formally perfect being. Of course, Descartes may have picked up the idea from another person, just as the engineer may have copied the ‘very perfect machine’ from a better engineer: but that merely pushes the problem one step backwards. At some point, such an idea must have been originally caused by God.
Descartes has already determined that all of his ideas must be adventitious, invented or innate. Since he has no direct sensory experience of God, the idea cannot be adventitious; the causal adequacy principle means that neither Descartes nor anyone else could have invented it; therefore the idea must be innate:
And indeed it is no surprise that God, in creating me, should have placed this idea in me to be, as it were, the mark of the craftsman stamped on his work—not that the mark need be anything distinct from the work itself. M3 p35
This engaging metaphor is what gives the Trademark Argument its name.
We can summarise the argument as follows to identify the individual premises:
In such a form, the argument seems plausible. However, each of the particular steps in the argument can be questioned, thus undermining its conclusion.
Premise 1 is, of course, the causal adequacy principle. The notion of causal adequacy is not easy to evaluate. There are simple examples that seem to confirm the principle: for example, if faced with a broken window, we can assume that the cause was sufficient to shatter the glass – not a grain of sand or a fly, but a brick or a goat. However, there are suggestive counter-examples too: the ‘Butterfly Effect’ describes substantial effects arising from insubstantial causes; the theory of evolution explains how complex creatures can evolve from simple organisms; the wetness of water is not present in its chemical elements of hydrogen and oxygen (emergent properties generally seem to contradict causal adequacy).
However, the principle of causal adequacy does seem plausible when applied to ideas. Descartes’ own example of ‘a machine of very ingenious design’ (p84) is convincing: we would need an account of where such ingenuity came from. Bernard Williams develops the idea: if we found two drawings in an ancient cave in some archaeological dig, one of an elaborate machine and one of a pile of sticks, the former would raise questions that the latter would not. We need to account for the elaborate and sophisticated machine with a suitably sophisticated origin story: ‘complex representational content requires a complex cause’ (Cottingham p53).
So there may be some value to the principle of causal adequacy. However, Descartes does not argue for the truth of the principle: he assumes it is self-evident ‘by the natural light’ (p28). It is a weakness in the argument that he does so: even if the premise is true, there is some way to go before it is beyond doubt.
Premise 2 claims that the idea of god has more reality than Descartes himself possesses, meaning that he is not sufficient to be the origin of the idea. The notion of degrees of reality is itself difficult to understand – ‘this unintuitive and barely comprehensible principle’, Bernard Williams calls it (p135) – but the main issue here is whether Descartes’ idea of God has any real substance at all. Descartes himself describes it as the ‘most clear and distinct of all my ideas’, but surely this is not the case. Even a passing acquaintance with the Philosophy of Religion reveals that the idea of God is riddled with difficulties: the omnipotence paradox, the problem of evil, questions of divine foreknowledge – to say nothing of the ideas of eternity, self-existence and infinity. Pierre Gassendi picked up on this last property:
... if someone calls something ‘infinite’, he attributes to a thing which he does not grasp a label which he does not understand. OR5 p81
Descartes’ reply to this charge is informative: he distinguishes between human understanding and ‘a fully adequate conception of things’. He acknowledges that he can understand the idea of God only as far as the scale of human intellect allows: which suggests that perhaps the idea of God is not too ‘objectively’ real to be invented by Descartes himself.
The idea of God, therefore, is hardly clear and distinct; and hardly contains a degree of ‘objective’ reality that cannot be accounted for. Descartes has inherited many of these assumptions about degrees of reality, and the idea of God seems so substantial perhaps because of a seventeenth-century apparatus of preconceptions about the supernatural that we tend to reject nowadays. Descartes’ argument is in this respect, at least, an orphaned child of its time.
Premise 3 follows from the first two premises, but without them we are not compelled to accept the conclusion that God must be the cause of the idea. This is, to the modern mind, the weakest step in the argument. The idea of God is not so substantial as to demand a supernatural cause, and even if it did contain such a degree of ‘objective reality’, we do tend to generate such ideas by extrapolating from our own experience. We do, in fact, come to understand the infinite from our experience of the finite.
This is a common claim of later empiricists like John Locke (eg see his Essay II.17). We experience certain aspects of the world in degrees: power, knowledge, virtue and so forth. By meditating upon this spectrum of possibilities, we can extrapolate to an infinite degree and thus conceive the ideas of omnipotence, omniscience and perfect goodness.
Descartes, however, rejects this possibility and claims instead that he can understand his own finite nature only by comparison with the notion of infinity:
For how could I understand... that I was not wholly perfect, unless there were in me some idea of a more perfect being which enabled me to recognize my own defects by comparison? M3 p31
He insists that the ideas of infinity and perfection must be ‘in some way prior’ to the ideas of finiteness and imperfection, which can then be understood by way of comparison. But this is surely wrong. When we learn the meaning of paired terms like finite and infinite, we need both for either to be meaningful: but neither has priority. We grasp them simultaneously, each as the negation of the other.
Unfortunately the Trademark Argument does not succeed in proving the existence of God. Even if some principle of causal adequacy is invoked, it does not seem to be the case that the idea of God demands any special explanation.
Towards the end of the Third Meditation, Descartes seems to introduce a second argument for the existence of God. He asks, if God did not exist, could Descartes exist?
From whom, in that case, would I derive my existence? From myself presumably, or from my parents, or from some other beings less perfect than God; for nothing more perfect than God, or even as perfect, can be thought of or imagined. M3 p33
Descartes rejects the possibility that anything but God could be the cause of his existence. If he was the author of his own being, he would have made himself perfect – he would be God. He is not perfect, and therefore not responsible for his own being. He considers the idea that he might always have existed, and therefore needs no beginning: but he then would need also to preserve himself continually, as things cease to be without being preserved. Since he is not aware of such a power in himself, he must depend on some other being. In a physical sense, he derives his existence from his parents (or another imperfect intermediary); but that only shifts the question to their existence, and a potentially infinite regress. More importantly, the causal adequacy principle demands that the author of Descartes’ existence also possesses the perfections evident in his idea of God. No imperfect being meets that description, therefore Descartes can only derive his existence from God.
The various arguments used to discount the alternatives are hardly persuasive. But, in the end, the Contingency Argument depends upon the Trademark Argument. Descartes could indeed be the product of a being less than God, if that being had sufficient formal reality to create and preserve him: but that would not account for the objective perfection of his idea of God. And so the Trademark Argument is necessary to prove that this idea is what demands the existence of a perfect being. And the failure of the Trademark Argument also renders the Contingency Argument invalid.
The Ontological Argument appears in the Fifth Meditation. At this stage of his enquiry, Descartes is still investigating the ideas he possesses: since he has yet to establish the existence of the material world, he is restricted to conceptual analysis. Both the Trademark Argument and the Ontological Argument start from the idea of God: but whereas in the former case it is the existence of the idea that is crucial, here it is the content of the idea.
As Descartes demonstrated in the Second Meditation, he understands the true nature or essence of an idea to be inseparable from it. The essence of something is those properties the thing needs to be what it is; essential properties cannot be removed without the thing ceasing to be what it is. In the Second Meditation, he realised that he could not imagine himself not thinking, but could imagine himself without a body – thus establishing his true nature as res cogitans.
In this case, he ‘unpacks’ the idea of God and realises that he cannot imagine God not existing.
Certainly, the idea of God, or a supremely perfect being, is one that I find within me just as surely as the idea of any shape or number. And my understanding that it belongs to his nature that he always exists is no less clear and distinct than is the case when I prove of any shape or number that some property belongs to its nature. M5 p45
God by definition is a being who possesses all the perfections. Existence is a perfection (in that non-existence is less than existence). Therefore, God possesses existence. This can be known a priori and with absolute certainty.
It is quite evident that existence can no more be separated from the essence of God than the fact that its three angles equal two right angles can be separated from the essence of a triangle, or than the idea of a mountain can be separated from the idea of a valley. Hence it is just as much of a contradiction to think of God (that is, a supremely perfect being) lacking existence (that is, lacking a perfection), as it is to think of a mountain without a valley. M5 p46
To say that God does not exist is to say that the perfect thing is not perfect; or the thing which lacks nothing lacks existence. It is a contradiction. The argument therefore is entirely self-evident, as clear and distinct as a geometric truth about triangles.
... what is more self-evident than the fact that... God, to whose essence alone existence belongs, exists? M5 p47
Descartes’ version of this argument is not the first. The canonical presentation is St Anselm’s version of the argument, although the basic structure is the same. For Anselm, God is ‘a being than which nothing greater can be conceived’, and he goes on to demonstrate that such a quality is incompatible with non-existence. The same demonstration underlies Descartes’ claim that perfection entails existence.
That there is something wrong with the argument was evident even to Anselm’s contemporaries. One such contemporary, a monk named Gaunilo, described an island ‘more excellent’ than any other, and argued that since a real island is more excellent than a non-existent one, this island must by definition exist. Others have noted that the argument seems to define into existence anything with the property of perfection.
Descartes, however, rejects this line of argument and insists that only God can be described in this way: ‘to whose essence alone existence belongs’ (p47). To justify this, he distinguishes between necessary and contingent properties. Descartes expresses this distinction in terms of separability: some properties cannot be logically separated from a concept, like a valley from a mountain or the fact that a triangle’s three angles equal two right angles. Without these essential aspects, the mountain or the triangle would cease to be what they are. However, a made-up concept like a perfect island or a flying horse can be analysed into separate concepts: these properties are merely contingent. ‘It is the necessity of the thing itself’, as Descartes puts it, ‘which determines my thinking in this respect’ (p46).
So a perfect island is an imaginary collection of properties that can logically be detached without contradiction. But that raises the question of whether God is not also an equally fictional bundle of properties. Several philosophers have noted that despite Descartes’ argument, ‘God does not exist’ is simply not a contradiction. We seem to have no difficulty separating existence from our idea of God. Descartes would argue that to do so involves a failure to understand the true nature of God (see his discussion of this point on p46). At first glance, this appears to be merely a quibble over definitions: Descartes defines God as a real entity with such-and-such properties (omnipotence, omniscience, creator etc); a sceptic defines God as an imaginary entity with such-and-such properties. But we can agree on the definition of God, without agreeing whether such a being exists. And that is because existence is not, in fact, a property like omnipotence – necessary or otherwise.
This realisation provides the strongest traditional assault on the Ontological Argument: existence is not a predicate, and the argument contains therefore a fatal grammatical error. The objection is usually attributed to Immanuel Kant, but Gassendi made the same point in the published Objections:
It is quite all right for you to compare essence with essence, but instead of going on to compare existence with existence or a property with a property, you compare existence with a property. OR5 p95
Existence is not just a quality which is attached to some things and not others, like ‘red’ or ‘almighty’. When we conceive of something – say, a cat – we can then predicate various things of the concept: a large cat, a red cat, a cat in a hat. To then add ‘which exists’ does not alter the concept in any way. If something exists, it does not possess the property of existence: rather, the independent fact is true that there is something in the world to which the concept corresponds. This explains why ‘God does not exist’ is not a contradiction: because ‘existence’ is not contained within the subject of God in the same way ‘omnipotence’ is. To say ‘God does not exist’ is merely to claim that there is nothing in reality to which the concept of God corresponds. It is a fact independent of the meaning of the word ‘God’. In Descartes’ language, existence is not a perfection like omnipotence: it is an altogether different type of thing, grammatically and ontologically.
Perhaps unsurprisingly, the Ontological Argument fails to establish a priori the existence of God. Some knowledge can be derived purely from an analysis of a clear and distinct idea, as we can discover that the internal angles of a triangle always equal two right angles. But existence cannot be discovered that way. Whether something exists or not is a purely empirical question.