However by saying that his dictum is not the result of syllogism Descartes raises further confusion. Should it be the case that he is not deducing ‘sum’ from ‘cogito’, that is ‘sum’ is self evident, what is the purpose of mentioning ‘cogito’ in the dictum? For if the conclusion is contained within the premises there is no real need to mention ‘cogito’, in the same way that it is not necessary to say that a spinster is an unmarried woman. Many have tried to justify why Descartes makes the association between thinking and existence. Some maintain that the dictum is as a result of deduction, whilst others claim it to be a simple case of intuition. However there are some that take the middle way, saying that it is neither but instead a ‘performance’. Jaakko Hintikka looks at this idea is more depth.
Bernard Williams (Descartes: The Project of Pure Enquiry, ch. 3).
The special status of 'I think' and 'I am'
Descartes says that there is something special about his belief that he is thinking, and his belief that he exists. What exactly is special about these thoughts? Descartes says that it is impossible to doubt these beliefs. So what is it about them that makes them immune to doubt? Descartes says in the passage above that 'the proposition I am, I exist, is necessarily true whenever it is put forward by me or conceived in my mind'. If these propositions are necessarily true, then that might be why they cannot be doubted. But are they 'necessarily true', as Descartes says? No, or not strictly. There are possible worlds in which Descartes does not exist. Perhaps in those possible worlds, his parents never met. If Descartes is taken to be the referent for 'I', then in those worlds the proposition 'I, Descartes, exist' is false. The proposition that 'I, Descartes, think' is also false, in those worlds. And something similar will apply no matter who the thinker is. These propositions are not necessarily true, in the usual sense in which philosophers speak of necessary truth. These propositions are contingent. But Descartes is surely right about their special status.
One suggestion is that these propositions have the special character of being incorrigible, and self-verifying. This suggestion has been made by Bernard Williams (Descartes: The Project of Pure Enquiry, ch. 3). It is assumed in his definitions that we are dealing with contingent propositions: propositions that are not true in all possible worlds. With that assumption, he proposes that a proposition P is incorrigible when it satisfies this description: if I believe that P, then P. Here are some candidate examples of propositions that satisfy this description. 'I am in pain.' This is, arguably, incorrigible, since if I really believe I am in pain, I am in pain. I can't be wrong about it. 'That looks red to me.' Here again, if I really believe something looks red to me, then it does look red to me. I am the expert about how things look to me. I might be wrong, of course, on how they really are. But (arguably) I can't be wrong about how they look to me. Now consider the propositions from Descartes' argument. 'I am thinking.' Suppose I believe that I am thinking. It follows that I am thinking. 'I exist'. Suppose I believe that I exist. It follows that I do exist. The propositions 'I think' and 'I exist' both seem to be incorrigible, in Williams' sense.
The propositions are also self-verifying. This is a closely related concept, which concerns assertion rather than belief. A proposition P is self-verifying when it satisfies this description: if I assert that P, then P. Here are some candidate examples of propositions that satisfy this description. 'I am speaking'. 'I can speak at least a few words of English'. 'I promise to come to the party'. If I assert (out loud!) that I am speaking, then I am speaking. If I assert that I can speak at least a few words of English, then I can speak at least a few words of English. If I say that I promise to come to the party, then I do promise to come to the party. The latter is an example of what Austin called a performative speech act. Some philosophers who see a similarity between this example and the propositions of the cogito have developed the 'performative' interpretation of Descartes' argument, mentioned above (see Kenny). Is there a similarity? Yes, in so far as all are examples of self-verifying propositions. If I assert 'I am thinking', then I am thinking. If I assert 'I exist', then I exist. Contrast these examples with self-refuting statements. 'I am absent.' 'I cannot speak any English'. 'I cannot think'. 'I do not exist'. (Can you imagine situations where these propositions might be used in a way that is not self-refuting?)
If this is correct, then there is indeed something special about the status of the propositions of Descartes' argument. Although they are strictly speaking contingent propositions, not necessary ones, they have the special features of being incorrigible and self-verifying. That is why they cannot be doubted.
However, there is a puzzle now. The conclusion Descartes wants to reach is 'I exist'. If this proposition on its own has the vital properties of being incorrigible and self-verifying, then why does Descartes bother with his premise, 'I think', and trouble to present the argument as 'I think, therefore I am'? The answer is not obvious, but here are two suggestions.
The first is that there is one formulation of the argument presented in the Meditations which can be interpreted in just this way. Descartes says, 'the proposition...I exist, is necessarily true whenever it is...conceived in my mind.' This could be interpreted exactly in line with Williams' suggestion: 'if the proposition 'I exist' is believed by me (conceived in my mind), then it is true'. In Williams' terminology: the proposition 'I exist' is incorrigible. On this reading, the conclusion 'I exist' is inferred from a thought about one's existence. This still leaves all the other formulations of the argument, however, in which existence seems to be inferred from thoughts about something other than one's existence (25). 'If I convinced myself of something, then I certainly existed.' If there is a deceiver who is deceiving me, 'in that case too I undoubtedly exist'. If I 'thought anything at all, then I certainly existed' (French version). In all these cases the premise that is supposed to yield a conclusion about existence is not a thought about one's existence, but rather a thought about e.g. a deceiver.
The second suggestion is that there is indeed a special reason for the premise of the argument being 'cogito', 'I think'. The proposition 'I think' has a special feature that is lacking in the proposition 'I exist'. The proposition 'I think' is evident (to use Williams' label). A proposition P is evident to me if it satisfies the following description: if P, then I believe that P. Compare this definition to the definition for incorrigibility above. You can see that being evident is roughly the converse of being incorrigible. Incorrigibility says, if you believe it, it's true. Evidence says, if it's true, then you believe it. IF SOMETHING IS INCORRIGIBLE TO YOU, THEN YOU ARE AN EXPERT ABOUT IT, IN ONE WAY. IF YOU BELIEVE IT, IT'S TRUE. IF SOMETHING IS EVIDENT TO YOU, THEN YOU ARE AN EXPERT ABOUT IT IN A DIFFERENT WAY. IF IT'S TRUE, THEN YOU BELIEVE IT. IT DOESN'T ESCAPE YOUR ATTENTION.
IT MAY BE THAT DESCARTES THINKS THAT ALL PROPOSITIONS ABOUT THE MIND ARE INCORRIGIBLE AND EVIDENT. INCORRIGIBILITY SAYS: WHEN I BELIEVE SOMETHING ABOUT MY MIND, I GET IT RIGHT. IF I BELIEVE SOME PROPOSITION ABOUT MY MIND, THAT PROPOSITION IS TRUE. That, on its own, is compatible with there being all kinds of dark corners and alleys of the mind about which I know nothing. But then Evidence adds: I know all there is to know. If some proposition about my mind is true, then I believe it. The poet Gerard Manley Hopkins said, 'O the mind, mind has mountains; cliffs of fall / Frightful, sheer, no-man-fathomed.' Descartes would have disagreed; he seems to think there are no unfathomable depths to the mind. There are no hidden corners or dark alleys. The mind is transparent to itself. I can know about all the operations of my mind.
WHILE THE PROPOSITIONS 'I THINK' AND 'I EXIST' ARE BOTH INCORRIGIBLE, AND BOTH SELF-VERIFYING, THE PROPOSITION 'I THINK' IS EVIDENT, IN A WAY THAT THE PROPOSITION 'I EXIST' IS NOT. IF I THINK, THEN I BELIEVE THAT I THINK. IS IT TRUE THAT IF I EXIST, I BELIEVE THAT I EXIST? NO, IT SEEMS. WHILE I BELIEVE THAT I THINK, WHEN I AM THINKING, I DO NOT ALWAYS BELIEVE THAT I EXIST, WHEN I AM EXISTING. PERHAPS I CAN CONTINUE TO EXIST IN A DREAMLESS, THOUGHTLESS SLEEP, AND SURELY THAT IS WHAT COMMON SENSE SUPPOSES. IN THAT CASE I EXIST, BUT DO NOT BELIEVE I EXIST, SINCE I DO NOT BELIEVE ANYTHING. WHILE 'I EXIST' IS INCORRIGIBLE, IT IS NOT, ON THE FACE OF IT, EVIDENT. SO DESCARTES HAS A REASON FOR CHOOSING TO BEGIN HIS ARGUMENT WITH THE PREMISE 'I THINK'.
In the Fifth Set of Replies Descartes shows that the challenger fails to realise that he is using the same irrational line of thought that he accuses Descartes of employing. In this way Descartes saves himself from the ‘logical blunder’ he is accused of committing. To accuse Descartes of assuming that the ‘I’ is ‘none of these things’ (that is mind or body) is a mistake because he ‘did not know whether the body was identical with the mind or not, [he] did not make any assumptions on this matter, but considered only the mind, it was only afterwards… that [he] said there was a real distinction between the mind and the body’. This means that the challenger, to make his point, is assuming that ‘the mind is not distinct from the body’ which without having the rational basis for doing so seems self-contradictory. Here Descartes shows the irrationality of the criticism by using the same line of argument to falsify the critic’s standpoint.
Descartes, according to J. Cottingham, admits that to arrive at the conclusion ‘Cogito ergo sum’ there is a need for some relevant presuppositions to be made, despite the premises being a priori by nature. This is because to distinguish between thinking and not thinking the person meditating will need to know what thinking is. In the same way they will need to know what doubting is as well as presuming that doubting is a branch of thinking. In addition, it is important to note the interdependence of thinking and existing, for to think one also needs to exist, not merely think to exist. Here it is seen that Descartes’ analysis does not appear ‘out of a complete vacuum’. It is quite significant that Descartes does not choose to comment on this because it reveals that his line of thought is not as radical as they may appear at first glance. If the claim, ‘cogito ergo sum’ is accepted as valid, it provides the first stepping-stones for Descartes’ venture towards certainty. If not then it may seem like a desperate attempt to salvage some beliefs from the mechanism of doubt, in fear of being left with nothing.
Descartes, J. Cottingham (Blackwell, 1998)
'Cogito ergo sum: inference or performance?' (Philosophical Review 1962)
Descartes, J. Cottingham (Blackwell, 1998)