thinks must exist’ and therefore that he logically concludes that
he exists. This view, it seems to me, is wrong. It should be
stated on no occasion, in the Meditations, does Descartes write
‘I am thinking, therefore I am’, nor anything directly
equivalent. Rather, he says:
“Doubtless, then, that I exist and, let him deceive me as
he may, he can never bring it about that I am nothing, so long as
I shall be conscious that I am something. So that it must, in
fine, be maintained, all things being maturely and carefully
considered, that this proposition I am, I exist, is necessarily
true each time it is expressed by me or conceived in my mind.”
(p. 80).
The point here is that it is impossible to doubt the truth of the
proposition ‘I exist’ when one utters it. It is an indubitable
proposition, and one that will necessarily be presupposed in
every attack of the sceptic. Descartes is not yet entitled to use
syllogisms as the possibility of the malign demon is still very
much alive. As an aside, Descartes himself denies that the cogito
is a syllogism, although it should be mentioned that in some of
the Replies to Objections he seems to assert that it is in fact
a syllogism. Finally, in the Regulae ad directionem ingenii,
Descartes denies the usefulness of syllogisms as a means to
knowledge.
I believe that, given Descartes’ project, it is fair to grant him
that the cogito deserves the status he bestows upon it. For can
there be anything more certain than something that is so forceful
and so powerful that every time it is presented to our mind we
are forced to assent to it?
What Descartes did here was to jiggle about the way philosophy
normally approaches the construction of knowledge structures. By
starting with self-knowledge, he elevates the subjective above
the objective and forces his epistemology to rest upon the
knowledge he has of his own self (and inadvertently sets the tone
for the next 300 years of philosophy). This leaves him with a
problem. He can know his own existence, that he is a thinking
thing and the contents of his consciousness, but how can any of
this ever lead to any knowledge of anything outside of himself?
The answer is that, by itself, it can’t. Descartes, in the third
Meditation, attempts to prove the existence of God, defined as a
being with all perfections. This proof is to be derived from his
idea of a God, defined as a being with all perfections. So far,
so good – Descartes examines the contents of his consciousness
and discovers within it this idea, and we can allow him this. At
this point, however, he introduces a whole series of scholastic
principles concerning different modes of causation and reality
without proper justification:
“For, without doubt, those [ideas considered as images, as
opposed to modes of consciousness] that represent substances are
something more, and contain in themselves, so to speak, more
objective reality, that is, participate by representation in
higher degrees of being or perfection than those that represent
only modes or accidents; and again the idea by which I conceive a
God has certainly in it more objective reality than those ideas
by which finite substances are represented.
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
its effect; for whence can the effect draw its reality if not
from its cause? And how could the cause communicate to it this
reality unless it possessed it in itself?”
Whence do these principles draw their indubitability? Even if we
grant that it is contrary to natural reason that an effect can
have greater ‘reality’ than its cause, that the concepts of modes
and substances are coherent with Descartes’ method, let alone
possess the properties that he ascribes to them, then surely we
can still bring the malign demon into play? Is it not possible
that this all- powerful demon could bring it about that Descartes
has a notion of a being with all possible perfections that he
calls God? No, says Descartes, because the notion (representing
something perfect) would then have more objective reality than
the demon (as something evil and thus imperfect) has formal
reality, and ‘it is manifest by the natural light’ that this is
not possible. But why not? Maybe the demon has just made it seem
impossible, and it seems that Descartes has no answer to this.
Further problems remain. Cosmological arguments for God invoking
the notion of causation have always had to contend with the
problem of the cause of God. For if all events (or ideas) are
caused ultimately by God, then what about God Himself? Why should
He be exempt from this rule? The standard response to this is to
claim that God, being omnipotent, causes Himself. One of the
chief perfections that Descartes attributes to God is that of
’self-existence’, that is, that His existence depends on nothing
else but itself. But if we examine this idea, it seems a little
confused. If God is the efficient cause of God then we are forced
to ask how something that does not yet exist can cause anything.
If God is the formal cause of God, i.e. it is part of the
intrinsic nature of God that he exists – which seems more likely
- then it seems that we have merely a reformulation of the
ontological argument for God’s existence from Meditation 5.
It seems that Descartes may have anticipated the wealth of
criticism that the causal proof of God would inspire, and so,
after explaining how human error and a benevolent, non-deceiving
God are compatible in Meditation Four, he produced in Meditation
Five a version of the mediaeval ontological argument for God’s
existence. Unlike the causal argument, the ontological argument
doesn’t involve the covert import of any new principles. It
simply purports to show that, from an analysis of his own idea of
God, Descartes can show that He necessarily exists. The reasoning
goes like this:
I have ideas of things which have true and immutable natures. If
I perceive clearly and distinctly that a property belongs to an
idea’s true and immutable nature, then it does actually belong to
that nature. I perceive clearly and distinctly that God’s true
and mmutable nature is that of a being with all perfections.
Further, I perceive clearly and distinctly that existence is a
perfection and non-existence a non- perfection. Thus existence
belongs to God’s true and immutable nature. God exists.
One of the interesting things about this argument is that, at
first sight, it does not seem to depend in any way upon anything
that has been proved hitherto. It is an application of pure
logic, an analysis of what we mean when we say ‘God’ and a
inference from that analysis. Descartes explicitly says that an
idea’s true and immutable nature does not in any way depend upon
his thinking it, and thus upon his existence. Once he has
perceived clearly and distinctly that an idea’s true and
immutable nature consists in such-and-such, that is the case
whether or not he thinks it is, or even if he exists or not.
Descartes in fact recognises the primacy of the ontological
argument: “although all the conclusions of the preceding
Meditations were false, the existence of God would pass
with me for a truth at least as certain as I ever judged any
truth of mathematics to be.” If this is true, which it seems to
be, then this argument is only as trustworthy as the faculties
which enabled us to construct it, which are the same faculties
that enable us to know mathematical truths, and so it seems
worthwhile to ask how, under Descartes’ theory, we come to know
mathematical truths. Descartes claims we perceive them clearly
and distinctly. How do we know that what we perceive clearly and
distinctly is true? Because God, being perfect, is no deceiver,
and would not let it be the case that we could ever perceive
something clearly and distinctly without it being the case. It
seems then, that this proof of God, relying on the veracity of
clear and distinct ideas, relies on the certain knowledge that a
non-deceiving God exists. We have another proof of God, the
causal proof as described in Meditation three. But apart from the
patent futility of using one proof of p to construct another
proof of p, on examining the causal proof of God further, we find
that it, too, relies upon a methodology that can only be relied
upon if the divine guarantee is present, for if this guarantee is
not present, then, as I mentioned above, how can we be sure that
the all-powerful demon is not exercising his malignant influence?
This, of course, is the infamous Cartesian circle, first
identified by Arnauld in the Fourth Objections and discussed ever
since. Many philosophers have tried to get Descartes off the hook
in various ways, some by denying that there is a circle and some
by admitting the circularity but denying its significance. I will
here briefly evaluate a few of their arguments.
Some commentators have taken a passage from Descartes’ reply to
the Second set of Objections (Mersenne’s) to indicate that
Descartes is only actually interested in the psychological
significance of fundamental truths. The passage is as follows:
“If a conviction is so firm that that it is impossible for us
ever to have any reason for doubting what we are convinced of,
then there are no further questions for us to ask; we have
everything we could reasonably want.”
Under my interpretation, this is what it is about the cogito
that makes it so important for Descartes, so we cannot have any
argument with the principle expressed by him in the above
passage. But can it help break the circle? When we clearly and
distinctly perceive something, Descartes says, fairly I think,
that this perception compels our assent, that we cannot but
believe it. God’s r_le in the system, to these commentators, is
as a guarantor of our memory regarding clarity and distinctness.
In other words, once we have proved God’s existence, we can
happily know that any memory we have of a clear and distinct idea
regarding x is true i.e. that we really did have a clear and
distinct idea of x. But this does not seem satisfactory, as we
still do not have a divine guarantee for the reasoning that leads
us from the clear and distinct notions we originally have about
God to the proof of His existence. We can give assent to the
clear and distinct notions we have originally; in fact, we are
compelled to give this assent when the notions are presented to
our mind, but the logical steps we take from these ideas to the
final proof is still subject to the evil demon because God is not
yet proven. Furthermore, because these steps are needed, the
memory of the original clear and distinct ideas are themselves
subject to doubt because God is not yet proven. It seems that the
only way either of the proofs of God could be accepted would be
if we had an original clear and distinct perception of God
directly presented to our mind (qualitatively similar to the
cogito). But this in itself would make any future proofs
redundant. Interestingly, this sounds quite similar to a divine
revelation.
Harry Frankfurt, in his book ‘Demons, Dreamers and Madmen’, has
argued that what Descartes is actually looking for is a coherent,
indubitable set of beliefs about the universe. Whether they are
‘true’ or not is irrelevant. Perfect certainty is totally
compatible with absolute falsity. Our certainty may not coincide
precisely with ‘God’s’ truth, but should this matter?:
“Reason can give us certainty. It can serve to establish beliefs
in which there is no risk of betrayal. This certainty is all we
need and all we demand. Perhaps our certainties do not coincide
with God’s truth But this divine or absolute truth, since it is
outside the range of our faculties and cannot undermine our
certainties, need be of no concern to us.” (Frankfurt, p 184)
This is almost a Kantian approach to knowledge, where we as
humans only concern ourselves with the phenomena of objects as
they present themselves to us, not with the objects in
themselves. Can we ascribe this view to Descartes? It’s tempting,
given what we have said above regarding the prime importance of
indubitability, but it would seem that a God presenting ideas to
us in a form which doesn’t correspond to reality, and then giving
us a strong disposition to believe that they do correspond to
reality would be a deceiving God and contrary to Descartes’
notion of Him. Thus the belief set would not be coherent.
Perhaps, as we do not have clear and distinct ideas of the bodies
we perceive, and as the divine guarantee only extends as far as
clear and distinct ideas, we are being too hasty in judging that
reality is how it appears to be and if we stopped to meditate
further we would see that reality is actually like something
else. But aside from the fact that this seems unlikely,
Descartes never seemed to envisage the possibility.
So much for the Cartesian circle. Where does this leave the
ontological argument, which we had only just begun to discuss?
Aside from the methodological difficulties, there do seem to two
further problems with it. The first has been noted by almost
every student of Descartes over the years – that of the
description of existence as a property. Put briefly, this
objection states that existence is not a property like ‘red’ or
‘hairy’ or ‘three-sided’ that can be applied to a subject, and
thus it makes no sense to say that existence is part of
something’s essence. If we assert that x is y, we are already
asserting the existence of x as soon as we mention it, prior to
any application of a predicate. from the beginning. In
other words, to say ‘x exists’ is to utter a tautology and to say
that ‘x doesn’t exist’ is to contradict oneself. So how can
sentences of the form ‘x doesn’t exist’ make sense? one may well
ask. It is because these sentences are shorthand for ‘the idea I
have of x has no corresponding reality’ and it was to solve
problems like this that Bertrand Russell constructed his theory
of descriptions. To add existence to an idea doesn’t just make it
an idea with a new property, it changes it from an idea into an
existent entity.
Finally, if Descartes is right, there seems no reason why we
cannot construct any other idea whose essence includes
existence. For instance, if I conjure up the idea of an existent
purple building that resembles the Taj Mahal’, then it is the
true and immutable nature of this idea that it is a building,
that this building resembles the Taj Mahal, that the building is
purple, and that it exists. But no such building does exist, as
far as I am aware, and if it did exist, its existence would not
be necessary, but contingent. This in itself is enough, I think,
to show that the ontological argument is false.
Once we have destroyed Descartes’ proofs of the existence of God,
the edifice of knowledge necessarily comes tumbling down with
them, as we find that almost everything Descartes believes in is
dependent on God’s nature as a non-deceiver:
“I remark that the certitude of all other truths is so absolutely
dependent on it, that without this knowledge it is impossible
ever to know anything perfectly.” (p.115)
The only possible exceptions are those assent-compelling beliefs
such as the cogito. Even these, however, are doubtful when we
are not thinking about them, and the above passage does give
weight to Edwin Curley’s argument that:
“Descartes would hold that the proposition “I exist” is fully
certain only if the rest of the argument of the Meditations goes
through. We must buy all or nothing.”
This is not the end of the story, though. As far as Descartes is
concerned, by the end of Meditation Five, he has produced two
powerful proofs of God, has a clear and distinct notion of his
own self, has a criterion for truth, knows how to avoid error and
is beginning to form ideas regarding our knowledge of corporeal
bodies.. And so it remains only to explain why we are fully
justified in believing in corporeal bodies, and also to draw the
ideas of Meditation Two regarding self-knowledge to their full