Russelm or Anselm?

G. E. M. Anscombe

The Philosophical Quarterly, Vol. 43, No. 173, Special Issue: Philosophers and Philosophies.
(Oct., 1993), pp. 500-504.

Stable URL:
http://links.jstor.org/sici?sici=0031-8094%28199310%2943%3A173%3C500%3AROA%3E2.0.CO%3B2-I

The Philosophical Quarterly is currently published by The Philosophical Quarterly.

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at
http://www.jstor.org/about/terms.html. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained
prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in
the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at
http://www.jstor.org/journals/philquar.html.

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.

JSTOR is an independent not-for-profit organization dedicated to and preserving a digital archive of scholarly journals. For
more information regarding JSTOR, please contact support@jstor.org.

http://www.jstor.org
Fri May 18 08:17:16 2007
 The Philosophical Quarterly Vol. 43, No. 173

ISSN 0031-8094

RUSSELM O R ANSELM?

BY G. E. M. ANSCOMBE

It is pleasing to have notice taken of my article 'Why Anselm's Proof in the

Proslogion is not an Ontological Argument'. I thank Professor Williams. I n reply, I
have to make some objections.

1. I did not argue that Anselm's argument could be 'saved by deletion ofa comma',
only that it could so be saved from the stupidity of an Ontological Argument. Note
that commas in editing Anselm are merely editorial judgement, there being no
commas in the mediaeval MSS.

2. I wanted to show what Anselm's argument was. I did not claim that it was a valid
proof. I could not determine whether it was a valid argument.

3. I showed that ifit was a valid argument ofthe 'Ontological' class, it has a missing,
i.e., unstated, premise: 'What exists in reality is greater than if it exists only in the
mind'. (This missing premise is supplied by Aquinas, Summa Theologiae, Part I,
Q. I1 art. 1 ii, who discusses an argument, obviously current in his time and deriving
from Anselm's, though he does not attribute it to Anselm. All the same, he thinks it
wrong.)

4. Williams does not notice the need of this extra premise, saying directly (or
perhaps taking me as saying) that the premise, which (with the editorial comma
put into the text before 'quod maius est') says that existing in reality is greater than
existing only in a mind, 'seems to involve the fallacy of treating being as a real
predicate'. I t does not: the extra premise is requisite.

5. Williams does not notice that Proslogion 3 is an argument assuming the conclusion
reached in Proslogion 2. The latter has supposedly proved the existence of that 'quo
maius cogitarinequit' ('than which nothing greater can be thought of'). Henceforth I
shall abbreviate this as qmcn, and 'quo maius cogitaripotest', 'than which a greater can
be thought of', as qmcp.
Given that qmcn does exist in re (in reality), Anselm argues further that it cannot
not exist. This impossibility must be an impossibility of its ceasing to exist. Hence
Williains' accusation that I equate 'It can be the case that that than which nothing
greater can be conceived is not-existent' with 'That than which nothing greater
can be conceived can be non-existent' is wrong. Given that idqmcn has been proved
0T h e editor3 or The Philorophirnl Qwrlerly, 1993
 RUSSELM OR ANSELM? 50 1
to exist, it can further be shown that it cannot be not-existent, i.e., cannot become
non-existent, i.e., cease to exist. The argument from non-conceivability applies:
namely, that qmcn, if existent, cannot be thought of as ceasing to exist. For one can
conceive something - call it id qmcn - to be capable of ceasing to exist. But then it is
not qmcn, for one would be thinking of something greater if one thought of
something qmcn which was incapable of ceasing to exist.
It is a mistake to treat Proslogion 3 as a new and independent argument for there
being such a thing as something qmcn, i.e., for 'id qmcn' being a non-empty concept.
This shows that Williams' arguments against Proslogion 3 are misconceived. He says
that Anselm attempts a reductio ad absurdum proof of the proposition 'That than
which nothing greater can be conceived cannot be conceived not to exist'. But, to
repeat, Anselm's reasoning in Proslogion 3 assumes that that qmcn (that than which
nothing greater can be conceived) has already been proved to exist, and he is
arguing something further. Here he does rely on what Williams calls a premise for a
further proof of the existence of id qmcn, namely that if you conceive of something as
existing and capable of not existing, your conception is a conception of something
inferior to what you conceive if you conceive of it as incapable of not existing.

6. It is a mistake to say that 'That than which nothing greater can be conceived' is a
'definite description' in the sense that Russell gave to that term. Anselm's first and
repeated expression in Proslogion 2 is 'aliquid qmcn', i.e., 'something qmcn'. I n the
same - short - chapter he later speaks of id qmcn, which we can render 'that than
which . . .'. There is no reason to think he has switched to a Russellian definite
description, which would have in strictness to be rendered 'that which alone is qmcn'.
It is worth observing that Russell in 'On Denoting' has a footnote: 'The
argument [namely "the most perfect Being has all perfections; existence is a
perfection; therefore the most perfect Being exists"] can be made to prove validly
that all members of the class of most perfect Beings exist. . . '. I do not complete this
interesting footnote, but quote its beginning to show that Williams' assumption
that 'id qmcn' is a definite description d la Russell wants grounds. Reading Anselm's
argument, I might say (to Anselm): 'You've proved the existence ofsome qmcn, but
so far as that goes it leaves open the question "How many things fall under that
description?" '. In Proslogion 2 Anselm speaks of 'such a nature' ('talis natura'). It
may be that there is only one 'such nature', but more than one thing that has it.

7. Here I will give briefly what ought to be the received version of Anselm's
argument in Proslogion 2, accepting the editorial comma before 'quod maius est'. ( I
put in square brackets the extra premise which that version needs, and which is not
in Anselm.) '

If qmcn exists only in a mind, still it can be conceived to exist in reality as well.
Now, for any object, to exist in reality as well as in a mind is greater than for it to
exist only in the mind. [And therefore it too is greater than if it exists only in the
mind.] Hence conceiving it to exist in reality as well as in a mind is conceiving it as
greater than one is conceiving it as if one thinks it exists only in a mind. Therefore
one cannot conceive qmcn to exist only in a mind. For if qmcn is only in a mind, this
0The editors af T h e Philorophirnl Quarlerb, 1993.
 502 G . E. M. ANSCOMBE

qmcn is also qmcp - but that is a contradiction. But it certainly does exist in a mind -
namely in that of the fool, who says there is no such thing in reality. Therefore qmcn
exists in reality as well as in the mind.
This argument can be blamed like the Ontological Argument, since it derives a
thing's being greater from its existence in reality's being greater than existence only in
the mind.

8. Mrilliams, like many others, writes as if 'greater than' in Anselm's argument is

simply a binary relation between two things. Such a relation involves comparing,
e.g., two things that both exist in reality. Some have indeed noticed that Anselm is
not invoking such a binary relation, and it is sometimes suggested that one cannot
compare what is 'in reality' with what is 'only in the mind'. This is a mistake: we
often make such comparisons, e.g., 'The explosion was not as loud as I expected it
to be'. Does this mean that the explosion in my imagination (expectation) was a
louder explosion than 'the' explosion when it happened in reality? The suggestion
makes clear that we cannot be operating as if we were comparing two explosions,
one existing in the mind and the other actually happening. Williams' notation does
not provide us with a tool for handling such matters. I t is clear that in Anselm's
argument the comparatives are what is called 'intentional'. Id qmcn cannot exist
only in the mind, for if it does, what is greater can be thought to exist in reality as
well. Here I am disregarding that editorial comma and taking 'quod maius est' to be
the subject of 'can be thought to exist in reality as well' (potest cogitari esse et in re quod
maius est). Anselm makes a very quick step in his argument here - his aim is to reach
a contradiction between id qmcn existing in the mind of the fool who says there is no
such thing in reality, and a greater thing, which can be thought to exist in reality as
well. The fact that it is something that can be thought to exist in reality and not
only 'in the mind' is essential to the contradiction, precisely because the fool's
thought is of that than which nothing greater can be thought of - but here is
something greater than what the fool's thought is of, if that exists only in the mind.
Not: the existence in reality is the greater-making thing; but: the possible thought is
of something that is greater. And so something greater can be thought of than what
in the fool's mind was that than which nothing greater can be thought of: That produces a
contradiction: the thought which Anselm describes is a thought of something greater
than what, in the fool's mind, is: something than which nothing greater can be
thought of.
A sort ofparallel: I can think of a pain more intense than can be survived. That is
just what my thought is, and that is all there is to it: it is 'only in my mind'. But now
someone says that he had witnessed a pain more intense than could be survived: for
the sufferer died of it. So he, the witness, has a thought which is of an actual pain,
more intense than could be survived. Was not the pain that he can now think of
(and believe to be possible) more intense than the pain that is 'only in the mind' of
the man who conjures up the thought of 'a pain more intense than can be
survived'? One might say: the former's description is real, the latter's is just
imaginary.
This, as I said, is only a sort of parallel. It does not quite hit the mark hit by
Anselm's argument. For one brings in the pain that the witness witnessed, which
0The editors or Tht Philo,ophrml g u n r l r r ! ~ .1993.
 RUSSELM OR ANSELM? 503
did kill the sufferer. But in the Anselmian argument there is not an appeal to the
object that is such that nothing greater can be conceived. We remain in the realm of
thought; if the fool's object is 'only in the mind', then something greater can be
thought to exist in reality as well. When Anselm says this, the questions
immediately present themselves: 'What, that is greater? And what is greater about
it? Is it "that it exists in reality as well"?' Ifso, then Anselm is indeed guilty in a way
of the fallacy alleged of the Ontological Argument.
But that is not the answer. T o find the answer, we have to read Anselm's reply to
Gaunilo, though we may get some help soon from the rest of the Proslogion. I t is a
greater thing that we are thinking of, if we are thinking, not merely that it exists in
reality as well as in the mind, but that it can have neither a beginning nor a n end of
existence. O r rather, we may have as the content of our thoughts not only
something qmcn, and that it exists in reality, or can be thought to, but also that if
there was something else, comparable to it, existing in reality also, but capable of
ceasing to exist and having had a beginning, our thought of this latter thing would
be a thought of a less great thing than was the thought of something qmcn, existing
in reality and by inference having had no beginning and having no capability of
ceasing to exist.
These and other arguments in the Reply to Gaunilo give body to the phrase 'what is
greater' ('quod maius est') in Proslogion 2. Proslogion 3, which relies on the conclusion
of Proslogion 2, and infers that non-existence - i.e., ceasing to exist - is not a
possibility for id qmcn, might be said to give some body to the phrase 'quod maius est'
('what is greater'); but the existence in re of id qmcn is given once one accepts the
essential argument of Proslogion 2. Namely, that if id qmcn is 'only in the mind',
something greater can be thought (believed) to exist in reality as well.
A final parallel: the largest prime number exists in the mind of the mathematics
student, even one who knows - perhaps can prove - that there is no such thing. But
suppose one who is at a more elementary stage, though being something of a
thinker, and he says 'I've got the idea of the biggest prime number in my head, but
I bet there's really no such thing. However big a prime number you might show
me, I'd feel it wasn't big enough to be the biggest - there'd have to be a bigger one.'
(Cf. Anselm's reply to Gaunilo's taking him to be arguing about what is greater
(and better) than everything else: 'You could show me something that was greater
than everything else, but I could say: "But I can think of something greater still".')
Perhaps our elementary mathematics student could formulate an explanation of
why 'there'd have to be a bigger one'. He might even invent the famous proof, and
learn quite a bit in doing so. At his stage of mere disbelief, we might compare him to
Anselm thinking: 'If that were only in the mind, it would still be possible to think of
something greater that would be there in reality.'
That is the impression one might form from that short crucial sentence in
Proslogion 2. All that we are given is that it would be possible to think that
something greater existed in reality as well as in the mind. However, that
impression, which I have imagined, may be wrong, because that powerful mind
produced that sentence and, we are told, very much wanted to produce a short
knock-down argument. In thinking about it, we are led into the intricacies of
thought about thought and of intentional comparatives. But the fact that Anselm
The editors or The Philorophirnl Qunrler!~, 1993.
504 G . E. M. ANSCOMBE

directed that publication of his Proslogion should thereafter always include

Gaunilo's attack and his reply (of which I have only mentioned a tiny bit, a single
argument among several) - this fact suggests that he realized he should have been
more explanatory of his Proslogion 2 sentence: 'For if it is only in the mind, what is
greater can be thought to exist in reality as well'.

T o end, I should bow to Professor Williams' consciously ironical title 'Russelm'.

\Ye know he is addicted to Russell on existence, but is there not in that title a hint of
acknowledgement that he thinks he may be wrong about Anselm?

New Hall, Cambridge

0The editors of T h r Philorophirnl Qunr/er!v, 1993.