Descartes: Meditation Two

1. In Meditation One, Descartes considered scenarios in which both (i) and (ii) come out true:
(i) I believe that X is the case;
(ii) X is not the case.
in other words, cases in which my beliefs are wrong. The Evil Genius Hypothesis raises the
possibility that all of my beliefs are wrong, even my belief that I have a body. But not even
the evil genius seems able to deceive me about my own existence, so long as I think that I
exist. There is no way in which (i) and (ii) both come out true when my own existence is
involved:
(i) I believe that I exist;
(ii) I do not exist.
If (i) is true (ii) must be false, and vice versa.

2. Thus Descartes proposes the truth of the claim “I exist”, whenever I think or utter the claim,
as the very first item of knowledge that he can be absolutely certain of. The reasoning by
which Descartes arrives at this indubitable knowledge of his own existence is commonly
referred to as “the Cogito” (short for cogito ergo sum, which is a Latin sentence meaning “I
think, therefore I exist”).

3. The Cogito does seem intuitively certain, but how exactly does the argument work? Perhaps
it is a modus ponens argument, as follows:
(1) If I think, then I exist.
(2) I think.
(3) Therefore, I exist.
There are two problems with understanding the argument as a modus ponens:
(a) If “I exist” is seen as the conclusion of a modus ponens argument, we must be
absolutely certain of the premisses first, before we can be certain of the
conclusion: which means that claims (1) and (2) must count as first items of
knowledge, rather than (3). But according to Descartes, (3) is the first item of
knowledge, not (1) or (2). So the modus ponens must be the wrong way of
interpreting Descartes’s reasoning.
(b) Keep in mind that we are assuming that there is an evil genius deceiving us of
even the simplest and most self-evident truths of arithmetic and geometry, such as
“2 + 2 = 4”. If he can deceive us about such truths, then he can deceive us about
the validity of modus ponens arguments as well. So the modus ponens argument
cannot guarantee my own existence from the evil genius. This does not show that
Descartes is wrong; what is shows is that the modus ponens argument is not a
charitable way of interpreting Descartes’s reasoning.

4. To better understand Descartes’s reasoning, let’s recall what he says about the Cogito: “this
pronouncement ‘I am, I exist’ is necessarily true every time I utter it or conceive it in my
mind” (Meditations, p.18). So understood, the Cogito has three important features.

1
 a. The first person formulation is essential to the certainty of the Cogito. For instance,
my saying or thinking of someone else, “Colin exists”, is less certain than “I exist”,
and can be mistaken if the Evil Genius Hypothesis is true.
b. The present tense is essential. For instance, my saying “I existed 10 minutes ago” can
turn out false if the universe (along with my memories that seem to extend 10 minutes
back into the past) was created just 5 minutes ago. Likewise, my prediction “I will
exist 10 minutes from now” can turn out false if the evil genius annihilates me.
c. Thinking is essential. As Descartes puts it, “I am; I exist—this is certain. But for
how long? For as long as I am thinking; for perhaps it could also come to pass that if
I were to cease all thinking I would then utterly cease to exist.” (Meditations, p.19)
Note that some non-mental activity, such as walking, will not work: “I exist so long
as I walk”. This is because it can be false that I am walking, for instance if I am
dreaming or if the evil genius is indeed deceiving me. But so long as I am engaged in
conscious activities like dreaming or falsely believing, I exist as a conscious thing.

5. The second and better way of reading Descartes’s Cogito is to understand it as a performance
rather than an inference (e.g., as a modus ponens).1 Logic is interested in statements insofar
as they are true or false, but we can also use statements to do various things. For example:
“I do (take this woman to be my lawful wedded wife).”
“I apologize.”
“I name this ship the ‘Queen Elizabeth’.”
“I bet you five dollars it will rain tomorrow.”
“I congratulate you.”
“I promise that....”
Notice that we perform actions by uttering these statements. In saying “I do”, I marry my
betrothed. In saying “I apologize”, I apologize. In saying “I bet...” I make a bet, etc. These
statements are known as performative utterances, and the actions I perform in making these
statements are called speech acts. Notice also that these performative utterances must be in
the first person and in the present tense, just as Descartes’s “I exist” is. This makes it
plausible to read Descartes’s “I exist” as a performative utterance, and his thinking or
uttering this statement as a speech act.

6. It is not logically inconsistent to think or utter “I do not exist”. This is because the statement
“I do not exist” is not a contradiction like “I exist and I do not exist” (i.e., having the form: p
and not-p). But there is some sort of non-logical inconsistency or absurdity involved in
thinking or saying, “I do not exist”. Moreover, my saying or thinking “I exist”—i.e., the
negation of “I do not exist”—makes the statement true. To fix terms, let’s say that thinking
“I do not exist” is existentially inconsistent, and thinking “I exist” is self-verifying. Below is
a table of various forms of speech act inconsistency and self-verifying speech acts, to help
you understand how there can be non-logical inconsistencies or guarantees of truth.2

1
This interpretation was proposed by Jaako Hintikka, “Cogito, Ergo Sum: Inference or Performance” in the
Philosophical Review, 1962.
2
I owe these examples to my own philosophy professor Alfred Mackay who taught me Descartes.

2
 NON-LOGICAL FORMS OF INCONSISTENCY SELF-VERIFYING SPEECH ACTS
Existential: Thinking, “I don’t exist” Thinking, “I exist”
Inscriptional: Writing, “I am not writing” Writing, “I am writing”
Vocalic (spoken): Speaking, “I am not speaking” Speaking, “I am speaking”
Vocalic (musical): Singing, “I am not singing” Singing, “I am singing”
Decibelic: Shouting, “I am not shouting” Shouting, “I am shouting”

Interpreting Descartes’s Cogito (thinking, “I exist”) as a performance rather than an inference

seems to make best sense of Descartes’s remark that “this pronouncement ‘I am, I exist’ is
necessarily true every time I utter it or conceive it in my mind” (p.18).

Some Criticisms of Descartes’s Cogito

7. There are some philosophers (e.g., Bertrand Russell in The Problems of Philosophy) who
argue that Descartes is not warranted in claiming that he exists, given the fact that he thinks.
He is only warranted in concluding that there is thinking going on, there is doubting going
on, there is being deceived going on, etc.
A variation on this theme is that Descartes’s Cogito rests on an accidental feature of
sentences in Indo-European languages, which always must have a subject-predicate structure
(but this isn’t true in East Asian languages, where the subject can drop out entirely). As the
19th century German philosopher Nietzsche put it, “When there is thinking, something must
be there which thinks—that is merely a formulation of our grammatical habit, which posits a
doer for what is done…” (Notes, 1887).
David Hume brings up another variation of the same theme: he claimed that when he
examined his own mind, he found no unitary or enduring self that does the thinking:
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. [Hume, A Treatise of Human Nature, Bk.I, §VI]
These objections seem right to me. However, I don’t think it affects Descartes’s proof of
God’s existence and subsequent arguments. For instance, if we find Descartes saying, “I
have an idea of a perfect being”, we can change it to “There is thinking going on about a
perfect being”, and the proof would still work, although it would have to be completely
rephrased. So, in my judgment, these objections do not undermine most of Descartes’s later
arguments, but perhaps it may give you some clues about possible weaknesses in Descartes’s
chain of reasoning.
It’s also worth noting that the exact nature of the self is not addressed by Descartes at this
early stage in the Meditations, for he wants to avoid such abstruse questions in his search for
an indubitable first item of knowledge. This gives rise to some intriguing possibilities.
Perhaps Descartes’s thinking self is a mere figment of someone else’s dream (a possibility
raised by Lewis Carroll). Perhaps the evil genius is dreaming that he is Descartes thinking
that there’s an evil genius deceiving him. All these possibilities are compatible with
Descartes’s proof that the thinking self exists in Meditation Two. The exact nature of the self
as a mental (and non-physical) substance is discussed in Meditation Six.