The Latin phrase cogito, ergo sum ("I think, therefore I am") is possibly the single best-known philosophical statement and is attributed to René Descartes. Cogito ergo sum is a translation of Descartes' original French statement, Je pense, donc, je suis.
Although the ideas expressed in cogito ergo sum are most commonly associated with Descartes, they were present in many of his antecedents, especially Saint Augustine in De Civitate Dei (books XI, 26) who makes this argument, and anticipates modern refutations of it. See Principia Philosophiae , §7: "Ac proinde haec cognitio, ego cogito, ergo sum, est omnium prima et certissima etc."
Having reached, at the end of the first meditation, what he considers to be the ultimate level of doubt—the evil demon —Descartes seeks one fact of which he can be certain. In the statement "cogito ergo sum", he finds two: that he is thinking and, following this, that he exists. This concept is stated at the start of the second meditation as follows:
- "I convinced myself that there is nothing at all in the world, no sky, no earth, no minds, no bodies; is it not therefore also true that I do not exist? However, I certainly did exist, if I convinced my self of something. There is some unidentified deceiver, however, all powerful and cunning, who is dedicated to deceiving me constantly. Therefore it is indubitable that I also exist, if he deceives me…he will still never bring it about that I am nothing as long as I think I am something."
At this point Descartes claims to know that he exists due to the occurrence of any cognitive process, including doubt (this argument will hereafter be referred to as the cogito). Thought entails existence, being as it is the final point of any assertion (this is true both in terms of statements such as "I am thinking about a rose", which is true as long as I am; and also such statements as "I breathe, therefore I exist", which itself is only true if it can be reduced to thought). Of course, the framing of the question 'How does Descartes try to prove his existence?' is unhelpful here, for the matter of Descartes proving his own existence as an historical figure is entirely different from his proof of the existence of some 'I'. The process by which he moves from the latter to the former is complex, interesting, and contentious; this article will be limited to the nature of 'I' only as a by-product of my discussion of the internal structure of the cogito. Three arguments will be presented against the cogito, all of which will be rejected, before an examination of Bernard Williams' rejection of the cogito on the grounds of its third-person comprehensibility.
Success of the cogito
There are countless different methods of doubting the validity of the cogito. The first of the three under scrutiny here concerns the nature of the step from "I think" to "I exist". The contention is that this is a syllogistic inference, for it appears to require the extra premise 'all thinking things exist' and that extra premise must surely have been rejected at an earlier stage of the doubt. Indeed, it appears to be untenable as a statement based solely on introspection. Descartes responds to this argument by conceding that there is indeed an extra premise needed, but denying that the cogito is a syllogism on the grounds that the extra premise—not 'all thinking things exist' but 'existence is necessary for thought'—is proved to be self evident by the cogito itself and is therefore receptive to analysis after the move from cogito to sum, but not present as a premise involved in the move. This response seems to be sound, but it anticipates another approach for criticising the cogito.
The second objection, that the notion that the cogito is not primary is intimated by the seeking of another premise before 'sum', and is again hinted at by Nietzsche in Beyond Good and Evil: "in the sentence 'I think'…[there are] a series of rash assertions which are difficult, perhaps impossible, to prove." Perhaps the ultimate claim in this direction is that the cogito is only actually comprehensible if a whole swathe of assumptions are left unquestioned by the doubt. These would include: an understanding of the syntax and semantics of the language used; a notion of what it is to exist; a notion what thought itself is. These could be dismissed by Descartes as trivial, due to the fact that it is nonsensical to undermine assumptions in these areas; if you could do so philosophical investigation itself would be meaningless.
Perhaps a more relevant contention is whether the 'I'; that Descartes assumes is justified. In Descartes, The Project of Pure Enquiry Williams provides a history and full evaluation of this issue—the third argument—. The main objection, given full voice by Georg Lichtenberg, is that rather than supposing an entity that is thinking, Descartes should have just said 'there is some thinking going on'. What is not at stake here is circularity. Whilst the statement 'I think therefore I am', as a sentence proving the existence of an entity is circular, the concept of 'thought proving existence' is not. This is because (ignoring the actual content of Lichtenberg's argument) any thinking supposes the entity—the proof is inherent rather than question begging. Returning to Lichtenberg, what exactly is the force of the statement that Descartes states too much in 'locating' the thought? It seems as though 'thought' is less substantial than 'I think', and so Descartes, given his reductionist manifesto, should be compelled towards the former. Whilst Williams is at pains to express why this is false, it seems that the best grounds for believing that it is false lie in its conceptual nature: "It is not at all clear that we really can grasp this supposed difference in the abstract." The point being that conceiving of thought in an impersonal manner is paradoxical. Perhaps in terms of parsimony Lichtenberg was right, but in combination with the second contention, that there are some unshiftable assumptions underneath the cogito, it seems that simply moving on is the only option, and the thought will have to be re-located in an entity: 'I'.
Whilst the preceding three arguments against the cogito (appear to) have failed, a very convincing argument has been made by Bernard Williams, in his exposition of why both Lichtenberg and Descartes were mistaken. He claims that what we are dealing with when we talk of 'thought' or when we say 'I think' is something conceivable from a third-person perspective; namely objective "thought-events" in the former case, and an objective 'thinker' in the latter. The obvious problem is that from introspection, or our experience of consciousness, we have no way of moving to conclude the existence of any 'third-personal' fact, verification of which would require a thought necessarily impossible being, as Descartes is, bound to the evidence of his own consciousness alone.
This appears to leave the slightly uncomfortable—possibly even amusing—position of introspective verification of existence, but without the possibility (on conceptual grounds) of objective knowledge of the self. Perhaps all that is left is to say 'I think, therefore I think', or, in the slightly more ominous tone of Franz Xaver von Baader "I think therefore I am thought".
- W.E.Abraham, Disentangling the Cogito, Mind, Vol. 83, no. 329 (1974)
- R.Descartes (translated by D.Clarke), Meditations on First Philosophy, in Meditations and Other Metaphysical Writings (Penguin, 1998)
- G.Hatfield, Routledge Philosophy Guidebook to Descartes and the Meditations (Routledge, 2003)
- F.Nietzsche, Beyond Good and Evil, translated by R.J.Hollingdale (Penguin, 1973)
- B.Williams, Descartes, The Project of Pure Enquiry (Penguin, 1978)
Last updated: 10-14-2005 18:36:59