Archimedes used to demand just one firm and immovable point in order to shift the entire earth; so I too can hope for great things if I manage to find just one thing, however slight, that is certain and unshakeable. M2 p16
The best known form of Descartes’ famous argument occurs in the later Principles of Philosophy:
... this piece of knowledge—I am thinking, therefore I exist—is the first and most certain of all to occur to anyone who philosophizes in an orderly way. Principles I.7
After this Latin formulation (‘cogito ergo sum’) – and its French equivalent in the Discourse on the Method (‘je pense donc je suis’) – the argument is usually referred to in shorthand as the cogito. It is worth noting that this orthodox form of the argument does not appear in the Meditations, only an abbreviated version:
I must finally conclude that this proposition, I am, I exist, is necessarily true whenever it is put forward by me or conceived in my mind. M2 p17
Before evaluating the cogito it is important to establish exactly what sort of argument, or claim, it actually is. The very brevity of its phrasing seems to allow for different interpretations; indeed, such questions were raised by Descartes’ contemporaries and have been discussed extensively ever since.
The use of the term ‘therefore’ implies that the conclusion (I exist) is an inference from a premise (I am thinking). This type of inference suggests a syllogism, a basic logical argument of the form:
In that case the cogito is missing a premise, and the full argument should look something like this:
This syllogism may well be sound (it is certainly valid) but unfortunately Descartes is not yet in a position to make a general assertion about ‘all things’, and cannot therefore supply the major premise (1). If this is the hidden structure of the cogito, then Descartes has assumed too much. There also appears to be a problem with the use of the term ‘I’ in the minor premise (2): since the argument is seeking to prove the existence of just that ‘I’, it cannot already appear earlier in the argument without being circular.
Perhaps unsurprisingly, Descartes himself did not consider the cogito a syllogistic inference. He states this explicitly in the Second Replies:
When someone says ‘I am thinking, therefore I am, or I exist’, he does not deduce existence from thought by means of a syllogism, but recognizes it as something self-evident by a simple intuition of the mind. OR2 p68
Despite this clear statement, Descartes elsewhere suggests that we must recognise a general claim along the lines of ‘in order to think it is necessary to exist’ (Principles I.10), but insists that this is not what guarantees the truth of the cogito.
The fact that Descartes describes the cogito as ‘necessarily true’ (p17) may suggest a second approach. Necessary truths are usually expressed as analytic statements – that is, statements that are true by definition. The model would be a proposition like the following:
Obviously the statement is true, and necessarily so because the notion of being unmarried is already present in the term ‘bachelor’. It is as if I said: I am unmarried therefore I am unmarried. This brings out another feature of analytic statements – they are generally empty of significance: true, but trivially so. This is because the second half of the statement (I am unmarried) adds nothing to the first half (I am a bachelor).
If we approach the cogito as an analytic statement, we would expect a number of things: that the proposition is necessarily true, that the second half is already present in the first half, and that it is therefore empty. Of course, Descartes would vehemently deny that it is empty of significance: after all, this is the foundation of all future knowledge. The question of whether the second half is present in the first half, however, is an interesting one.
For this to be analytic, the concept of existence would need to be present in the concept of thinking (as ‘unmarried’ is implied in ‘bachelor’). But that is not the case: Hamlet thinks as much as anyone, but he does not exist. Existence generally is not that sort of property: you can’t predicate existence as you can predicate qualities like ‘moody’, ‘Danish’, or ‘prince’. (This very question will become central when Descartes formulates the Ontological Argument for the existence of God in Meditation 5.) We can therefore conclude that the cogito is not intended as an analytic statement.
The truth of the cogito is, Descartes insists, ‘self-evident... a simple intuition of the mind’ (p68). There is something unique about thinking in this context: Descartes is quick to reject Gassendi’s claim that any action could be used to prove the existence of the agent:
I may not, for example, make the inference ‘I am walking, therefore I exist’, except in so far as the awareness of walking is a thought. OR5 p68
In the case of thinking – more precisely, doubting – the belief and the act coincide: to doubt that one is thinking is self-contradictory, in the same way that ‘I am not talking’ cannot be said truly. Descartes can therefore be certain that he is thinking (although not certain that he is walking, breathing, smiling and so on). The step from thinking to existing seems to be the ‘simple intuition’: for as long as Descartes is thinking, he is unable to doubt that he exists.
I am, I exist—that is certain. But for how long? For as long as I am thinking. M2 p18
There seems to be a performative element involved in this intuition: his existence is guaranteed by the very act of thinking. In the same way, saying ‘I do not exist’ is made false by the act of saying it. The proof is in the utterance.
We should note that Descartes’ claim at this point is modest: with sceptical caution, he recognises that he can be certain that he exists only while he is thinking. But as long as he is, his existence seems equally certain.
Descartes presents the cogito as a simple and irresistible intuition, but other philosophers have found fault with Descartes’ reasoning. It is worth exploring three of the most effective objections.
It appears that even to produce the cogito, Descartes must take for granted a degree of prior knowledge: as Cottingham puts it, ‘the information does not, as it were, come out of a complete vacuum’ (p41). He must already understand the nature of thought and doubt, existence, even logical entailment (the implication of ‘therefore’). This apparatus must be in place before he can articulate the fact of his existence, but Descartes claims not to be taking anything for granted. Rather, his claim in Meditation Two seems to be that his existence is the only notion that resists his sceptical doubt. Elsewhere Descartes admits as much:
I did not in saying that [I am thinking, therefore I exist] deny that one must first know what thought, existence and certainty are, and that it is impossible that that which thinks should not exist, and so forth. Principles I.10
Perhaps more crucially (as Immanuel Kant noted) Descartes makes a number of metaphysical assumptions about substance and attributes. The cogito itself seems to imply the need for a substance, a thing, of which thinking is an attribute. Descartes has inherited a number of metaphysical notions that underlie his thinking here. As he puts it elsewhere:
... if we perceive the presence of some attribute, we can infer that there must also be present an existing thing or substance to which it may be attributed. Principles I.52
In other words, if there is thought there must be a mind. Perhaps this is the case, and perhaps Descartes has no choice but to retain some conceptual apparatus that allows him to meditate in the first place; but such considerations undermine the claim that the cogito is an independent first principle of knowledge. They also give us reasons to doubt the very foundational model of knowledge to which Descartes subscribes – it may be that there can be no first principles and that all of our knowledge is interdependent in some more complex fashion.
In some ways the cogito seems like a small step: after all, the existence of one thinker is a long way from a comprehensive body of scientific knowledge. Nevertheless, some philosophers have insisted that Descartes is not in a position even to intuit this much. Georg Lichtenberg famously argued that Descartes is only entitled to claim ‘there is thinking’, and his further intuition (‘I exist’) is therefore groundless. By analogy with phrases like ‘there is lightning’ or ‘it is raining’, Lichtenberg asserts that Descartes has no right to infer a thinker from the existence of thought.
This objection can be countered by accepting various assumptions about the nature of thinking (or lightning or rain), namely that there must be a substance to which thinking is attributed (or a medium enabling lightning or rain to occur). But as we saw above, such assumptions may not be justifiable at this stage.
Bertrand Russell made a similar but more searching claim to the effect that Descartes was inferring too much in the cogito:
When I look at my table and see a certain brown colour, what is quite certain at once is not ‘I am seeing a brown colour’, but rather, ‘a brown colour is being seen’. This of course involves something (or somebody) which (or who) sees the brown colour; but it does not of itself involve that more or less permanent person whom we call ‘I’. So far as immediate certainty goes, it might be that the something which sees the brown colour is quite momentary, and not the same as the something which has some different experience the next moment. The Problems of Philosophy p8
Although Descartes will later make more substantial claims about the nature of the mind, at this point he is only committed to the existence of the self for as long as it doubts or thinks (‘for as long as I am thinking’ p18). But Russell’s argument should make us cautious: it is easy to slip from the existence of a momentary consciousness to an enduring self, and Descartes is not yet in a position to claim that.
Later philosophers would come to very different conclusions regarding the existence of the self. Even Gassendi, in his Objections, suspected that Descartes was leaving too much unsaid:
Who doubts that you are thinking? What we are unclear about, what we are looking for, is that inner substance of yours whose property is to think. OR5 p71
Descartes’ claim here, as with the piece of wax, is that we can infer or intuit the existence of the self from its attribute or property – in this case, thought. Because we have direct evidence of thinking, we can assume the existence of a substance doing the thinking. But as we have seen, that is perhaps too much to claim. David Hume would famously make the point that we have no direct access to this self, this ego:
For my part, when I enter most intimately into what I call myself, I always stumble on some particular perception or other, of heat or cold, light or shade, love or hatred, pain or pleasure. I never can catch myself at any time without a perception, and never can observe any thing but the perception. Treatise I.iv.6
This is a direct challenge to the cogito: Hume insists that thinking is no evidence of a mental substance, an independent self, whatever Descartes says:
He may, perhaps, perceive something simple and continu’d, which he calls himself; tho’ I am certain there is no such principle in me. ibid
On this point, however, Hume is being a strict empiricist: he doubts the existence of the self because he has no direct perception of such a thing. As a rationalist, Descartes is not restricted only to empirical knowledge: his claim is that the intellect can legitimately intuit the existence of a self without any direct perception, in the same way that it understands the true nature of the wax beneath its changing properties. Whether such an intuition is legitimate is of course crucial to the success of the cogito.
A similar objection was made by Wittgenstein:
Where in the world is a metaphysical subject to be found?... nothing in the visual field allows you to infer that it is seen by an eye. Tractatus 5.633
Just as the eye sees what is in the visual field although it is not a part of it, so the (metaphysical) subject may become aware of the world directly, but not of itself. The self is not a thing in the world, Wittgenstein suggests, but a point of view on the world. And, of course, if Wittgenstein or Hume are correct, then Descartes is mistaken in claiming that such a self exists, and the cogito fails.
With so much intervening history and influence, it is perhaps difficult to see the cogito for the simple but radical argument it was. Thomas Nagel recognised this:
... the cogito is a philosophical Rorschach test, in which everyone sees his own obsessions. The Last Word p19fn
The cogito has been dismissed, defended, criticised and analysed in equal measure, and in the end it is not easy to evaluate its success as a philosophical argument. There is no doubting its influence, however. The model implied by the cogito was revolutionary: the focus on the individual seeker of knowledge; the undermining of authority and established opinions; the subjective, psychological criterion of certainty; the power of reason to determine the nature of things; the painstaking examination of ideas and how they appear to the conscious mind – the effect of these ideas would ripple outward from the cogito for the next few hundred years and change the nature of philosophy.