2011 The Author

Metaphilosophy 2011 Metaphilosophy LLC and Blackwell Publishing Ltd

Published by Blackwell Publishing Ltd, 9600 Garsington Road, Oxford OX4 2DQ, UK, and
350 Main Street, Malden, MA 02148, USA
METAPHILOSOPHY
Vol. 42, No. 5, October 2011
0026-1068

THE MEDIEVAL ORIGINS OF CONCEIVABILITY ARGUMENTS

STEPHEN BOULTER

Abstract: The central recommendation of this article is that philosophers trained in

the analytic tradition ought to add the sensibilities and skills of the historian to
their methodological toolkit. The value of an historical approach to strictly philo-
sophical matters is illustrated by a case study focussing on the medieval origin of
conceivability arguments and contemporary views of modality. It is shown that
common metaphilosophical views about the nature of the philosophical enterprise
as well as certain inference patterns found in thinkers from Descartes to Chalmers
have their origin in the theological concerns of the Scholastics. Since these assump-
tions and inference patterns are difficult to motivate when shorn of their original
theological context, the upshot is that much post-Cartesian philosophy is cast in
an altogether unfamiliar, and probably unwelcome, light. The methodological
point, however, is that this philosophical gain is born of acquaintance with the
history of ideas.

Keywords: Scholasticism, conceivability arguments, modality, philosophical

methodology, Cartesianism.

1
History has never been the strong suit of analytic philosophers. And not
without reason. Philosophy, whatever it is, is not history. But one can
accept this incontrovertible point without accepting the conclusion com-
monly drawn from it, namely, that philosophy has little or nothing to gain
by investigating its own history. One of the aims of this article is to
illustrate the methodological point that the close study of the history of
philosophy can be a source of philosophical insight not readily attainable
by any other means. Consequently, the analytic philosopher ought to avail
herself of the methods and sensibilities of the historian. To lend plausibil-
ity to these claims I offer some reflections on the Scholastic roots of
Cartesianism by way of a case study.
To assert the philosophical value of a study of history will strike some
as implausible; to advocate on behalf of medieval philosophy may seem
nothing less than reckless. But this attitude is mistaken, a product, in most
instances, of a profound ignorance of the period in which Western Europe
initiated the political, economic, and intellectual developments that even-
tually resulted in the emergence of modernity. The intellectual develop-

2011 The Author

Metaphilosophy 2011 Metaphilosophy LLC and Blackwell Publishing Ltd
618 STEPHEN BOULTER

ments in theology and the sciences are well known, and resulted in the
Reformation and the Scientific Revolution, respectively. What is rarely
recognised, however, is that the self-image of philosophy underwent a
similarly radical transformation in the period of transition from the
Middles Ages to Modernity. What is more, many elements of the new
self-image remain with us today. The result is that many deep and usually
unexamined assumptions of the current philosophical mainstream con-
cerning the nature of the philosophical enterprise, the rules according to
which its business should be conducted, and even the validity of certain
inference patterns, are the products of Scholasticism. What has hidden
this historical fact from view is that the ingredients of the metaphilosophy
of the philosophical mainstream are usually first encountered as part and
parcel of Cartesianism, that is, a philosophical system which allegedly
broke decisively with the medieval past. But contrary to popular opinion,
philosophy did not leave the Middle Ages behind with the advent of
Descartes. Rather, it is precisely through Descartes that Scholasticism
continues to shape our thinking.1
Making this case will not be easy, as it runs counter to the prevailing
view of Descartes as having swept away all previous philosophical tra-
ditions and introducing in its stead a radical new philosophya view
most imbibe in their philosophical infancy.2 And many will wonder
whether Descartes really does continue to shape our thinking in any
meaningful way, thereby casting doubt on the claim that an unrecognised
Scholastic influence needs to be brought to light. Both concerns have to be
confronted and addressed.
I begin in section 2 by drawing attention to a selection of arguments
and claims from leading philosophical figures ranging from Chalmers to
Hume to illustrate a number of recurring inference patterns and assump-
tions that have animated post-Cartesian philosophy down to the present
day. Section 3 then characterises the mind-set that lies behind these argu-
ments, and notes that it is part and parcel of the Cartesian framework.
Having seen in section 2 that this mind-set is still very much with us, I go
on in section 4 to show that its roots lie not in Descartes but in the
theological concerns of thirteenth- and fourteenth-century Scholasticism.
Finally, in section 5 I consider some objections to the conclusions I draw
on the basis of this history. The main conclusions are as follows: The
Cartesian mind-set has little to recommend it once its theological under-
pinnings have been exposed, so a re-evaluation of its characteristic infer-
ence patterns and assumptions is required. The broader methodological
1
This article is broadly sympathetic to the line taken by Friedman and Nielson that in
the higher or speculative sciences no sharp divide exists between the later Middle Ages, on the
one hand, and the Renaissance and early modern period, on the other (2010, 1).
2
These quotes are taken from the back cover of a recent Oxford Worlds Classics
translation of A Discourse on the Method, translated by Ian Maclean (Oxford: Oxford
University Press, 2006).

2011 The Author

Metaphilosophy 2011 Metaphilosophy LLC and Blackwell Publishing Ltd
 THE MEDIEVAL ORIGINS OF CONCEIVABILITY ARGUMENTS 619

point, however, is that a collection of mere historical factsfor that is

what this article is ultimately built uponleads to the recognition that this
mind-set, hitherto found entirely natural, perhaps even compulsory, must
be revisited with a critical eye. This use of history is anything but philo-
sophically sterile.

2
Consider the following sample of arguments and claims from leading
figures in the discipline.

Chalmers and the Zombie Objection to Physicalism

David Chalmers (1996) is perhaps the best-known proponent of the
zombie objection to various forms of physicalism in the philosophy of
mind. The objection has been around for some time now, but it
has recently returned to prominence.3 Following Braddon-Mitchell and
Jackson (2007, 12324) it can be presented as follows:

1. We can conceive of a world which is physically identical to ours but

which lacks features that ours has. In particular, we can imagine
people (zombies) who are physically identical to us but who lack
consciousness.
2. Conceivability is a good guide to logical possibility.
3. Zombies are logically possible (from 1 and 2).
4. We are not zombies, since we are conscious.
5. There is a logically possible world which is a physical duplicate of
the actual world while being mentally different (from 3 and 4).
6. Physicalism, that is, the view that a complete physical description of
the world exhausts all that it contains, is false (from 5).

Nagel on the Explanatory Gap Between the Brain and Qualia

Nagel (1974) claims that nothing we currently know about the brain, nor
anything we can imagine finding out about the brain, could explain qualia
and other aspects of consciousness. On the basis of this he has argued that
the mind/body problem is insolvable within our current conceptual frame-
work. His argument can be reconstructed as follows:

1. Causal explanations in the natural sciences have a kind of causal

necessity. (Causal explanations explain why a system has to be in the
state it is in.)
3
A very early version of this argument can be found as far back as Joseph Priestlys
Disquisitions Relating to Matter and Spirit. See Yolton 1983, 114.

2011 The Author

Metaphilosophy 2011 Metaphilosophy LLC and Blackwell Publishing Ltd
620 STEPHEN BOULTER

2. Necessity implies inconceivability of the opposite.

3. No account of neuronal activity can explain why, given a particular
neuronal state, one has to be, for example, in pain because one can
always conceive of a state of affairs in which a system in that par-
ticular neuronal state is not in pain (there is no logical contradiction
in such a state of affairs).
4. Thus there can be no adequate scientific explanation of the mental in
terms of the physical.
5. Thus we will not be able to solve the mind/body problem (at least
not within our current conceptual scheme).

Quine Against Essentialism

Quine (1953) argues against essentialism on the following grounds:

1. If E were essentially true of X (if E were of the essence of X), then X

is E would be a necessary proposition.
2. If X is E is necessary, then X is E is analytic.
3. But X is E is analytic is true or false only relative to our concep-
tions of X.
4. So, contra essentialism, essential properties are observer relative,
not features of things in and of themselves.

Wittgenstein on Necessity
Bearing in mind that the last lines of the Tractatus appear to retract much
that has been previously asserted, Wittgenstein (1986) states, but does not
argue for, the following claims:

1. Outside of logic all is accident. (6.3)

2. A necessity for one thing to happen because another has happened
does not exist. There is only logical necessity. (6.37)

Moores Refutation of Naturalism in Metaethics

In Principia Ethica (1993, sec. 13) Moore argues against all versions
of reductive naturalism in ethics using the Open Question Argument.
Although the proper formulation of the argument is a matter of some
debate, if one recalls that Moore is interested in the natures of things
referred to by means of language rather than in the meanings of words per
se, then the OQA can be reconstructed along these lines:

1. If the referent of good is identical to the referent of desirable (as

a naturalist might claim) then, since everything is necessarily iden-
tical to itself, it is necessarily the case that the referent of good is
identical to the referent of desirable.

2011 The Author

Metaphilosophy 2011 Metaphilosophy LLC and Blackwell Publishing Ltd
 THE MEDIEVAL ORIGINS OF CONCEIVABILITY ARGUMENTS 621

2. But since the only necessity is logical necessity, good ought to be

conceptually analysable as desirable. That is, If X is desirable,
then X is good ought to be an analytic truth because it would
amount to a barren tautology.
3. But If X is desirable, then X is good is not analytic, because it is
deemed to be an open question by all competent speakers of English,
so
4. The referents of good and desirable are not identical, contra the
naturalist.
5. Steps (1) through (4) can be reiterated for any other naturalistic
analysis of good, so
6. Ethical naturalism is false.

Some Programmatic Remarks from Kant

On the standard reading of Kant, his project in the Critique of Pure
Reason takes as one of its points of departure the assumption that Hume
correctly asserted that (i) there is no impression of necessity to be had in
our experience of the empirical world and (ii) that there is no conceptual
connection between causes and effects.4 But, unlike Hume, Kant was not
a sceptic about the possibility of scientific knowledge, which he took to be
knowledge of necessary propositions, because he believed that human
beings actually possess genuine a priori knowledge in the field of the
general science of nature as well as pure mathematics (1933, 128). In
order to account for this obvious discrepancy Kant argued for a possibil-
ity that never occurred to Hume, namely, that the understanding
might itself . . . be the author of the experience in which its objects are
found (127). Kant did not maintain that the only necessity is logical
necessitythere are, after all, on his account, synthetic a priori proposi-
tions. But because he maintains that necessity is not to be found in the
world in and of itself, and so cannot be discovered a posteriori, Kant is
forced to argue that the ground of this necessity lies in the mind-dependent
nature of the phenomenal world. It was this that never occurred to Hume;
but it would never have occurred to Kant if he had not accepted Humes
two claims.

Hume on the Connection Between Causes and Effects

One historically effective reading of Hume has Hume submitting the
common sense notion of causation to analysis and finding that one of
4
On the first point Hume says: Experience teaches us that a thing is so and so, but not
that it cannot be otherwise. First, then, if we have a proposition which in being thought is
thought as necessary, it is an a priori judgment (1933, 44). On the second we find the
following: We cannot determine from mere concepts how . . . because something is, some-
thing else must be, and how, therefore, a thing can be a cause (253).

2011 The Author

Metaphilosophy 2011 Metaphilosophy LLC and Blackwell Publishing Ltd
622 STEPHEN BOULTER

its key ingredients, namely, the necessity of the connection between

cause and effect, does not survive scrutiny. Hume then suggests that
this notion of causation ought to be replaced by the weaker notion
of constant conjunction. A cause is constantly conjoined with its effect,
but there is no necessary connection between the two as we are wont
to suppose, or at least we are not justified in believing that there is.
The essential claim for our purposes is as follows: Upon the whole,
necessity is something, that exists in the mind, not in objects; nor is it
possible for us ever to form the most distant idea of it, considerd as a
quality of bodies (1989, 16566). The claim that necessity is never
grounded in the nature of things themselves but only in the relation of
ideas is supported by two arguments. The first stresses that there is no
impression of necessity in the connection; the second stresses that there
could be no such necessity because cause and effect are two distinct
entities. It is the latter argument that is of interest here. It runs as
follows: The mind can never possibly find the effect in the supposed
cause by the most accurate scrutiny and examination. For the effect is
totally different from the cause, and consequently can never be found in it
. . . every effect is a distinct event from its cause. It could not, therefore,
be discovered in the cause, and the first invention of conceptions of it
. . . must be entirely arbitrary (1985, 2930, emphasis added). This is
important because: As all distinct ideas are separable from each other,
and as the ideas of cause and effect are evidently distinct, twill be easy
for us to conceive any object to be non-existent this moment, and exist-
ent the next, without conjoining to it the distinct idea of a cause or
productive principle. The separation, therefore, of the idea of a cause
from that of a beginning of existence, is plainly possible for the imagi-
nation; and consequently the actual separation of these objects is so far
possible, that it implies no logical contradiction or absurdity; and is
therefore incapable of being refuted by any reasoning from mere ideas
without which it is impossible to demonstrate the necessity of a cause
(1989, 7980, emphasis added).

3
I have isolated these particular arguments and claims because they are
familiar, have been of undoubted historical significance while remaining
the subject of current debate, and illustrate a particular philosophical
mind-set. What is interesting about these arguments and remarks is that
they all rest on one characteristic assumption, and, insofar as they are
arguments, they contain a set of closely related and highly characteristic
inference patterns.
The central assumption is that the only concept of necessity which has
positive application is logical necessity. In many cases what I will call the

2011 The Author

Metaphilosophy 2011 Metaphilosophy LLC and Blackwell Publishing Ltd
 THE MEDIEVAL ORIGINS OF CONCEIVABILITY ARGUMENTS 623

necessity claim is simply asserted.5 Logical necessity in this context

should be understood to embrace both strict logical necessity and what
might be called conceptual or semantic necessity (a strictly logically neces-
sary proposition being true or false in virtue of the laws of logic on their
own, while a conceptually or semantically necessary proposition is true
or false in virtue of the laws of logic combined with the definitions of
non-logical terms). The important point for present purposes, however, is
that the necessity claim implies the denial of the existence of physical,
causal, or nomic necessity as well as what one might call metaphysical
necessity.6
The necessity claim is almost invariably accompanied by a thesis con-
cerning how one is to trace the realm of the possible. In its most straight-
forward formulation, the thesis is that what one can conceive of is a good
guide to possibility and necessity. If something, or some state of affairs, or
some event is deemed conceivable, then one is warranted in asserting its
possibility; if not, then not. A recent expression of this line of thought is
found in Williamson: We assert A when our counterfactual develop-
ment of the supposition A robustly yields a contradiction; we deny A
when our counterfactual development of A does not robustly yield a
contradiction. . . . Similarly . . . we assert A when our counterfactual
development of the supposition A does not robustly yield a contradiction
. . . ; we deny A when our counterfactual development of A robustly
yields a contradiction. Thus our fallible imaginative evaluation of coun-
terfactuals has a conceivability test for possibility and an inconceivability
test for impossibility built in as fallible special cases (2007, 163). Of
course, just what conceivability amounts to is a contentious matter. Some-
times the conceivable is taken to be the imaginable, and the two terms are
often simply stylistic substitutes for one another. Chalmers (2002) has
tried to offer some help here by distinguishing various kinds of conceiv-
ability, but the core suggestion is that something is conceivable if a con-
ceiver, relying on background knowledge, runs a mental simulation of a
putative possibility and notices no contradiction arising in the course of the
simulation. While conceivability on this account is always relative to a
conceivers background information, the general thrust is that conceiv-
ability is grounded in logic. A thing, or a state of affairs, or an event is

5
This is obvious in some cases, and at least implicit in most of the others. But it is not true
of Chalmers. Chalmers does not deny that other forms of necessity need to be recognised.
For example, while he claims that zombies are logically possible, he does not maintain that
they are physically possible in this world due to current natural or physical laws. Nonethe-
less, on the basis of their logical possibility he does claim that physicalism is actually false.
6
I take logical, conceptual, or semantic possibility/necessity to concern entailment rela-
tions between propositions; physical, causal, or nomic possibility/necessity concerns rela-
tions between events in the physical realm; metaphysical possibility/necessity concerns states
of affairs in and of themselves, in particular the ontological relationships that obtain between
entities in the various categories.

2011 The Author

Metaphilosophy 2011 Metaphilosophy LLC and Blackwell Publishing Ltd
624 STEPHEN BOULTER

conceivable, and therefore possible, if no contradiction is perceived in the

positing of the thing, state of affairs, or event.
When the necessity claim is combined with the conceivability thesis it is
but a small step to an important corollary that can be called the Principle
of Separability.

PoS: If one can conceive of X apart from Y because neither is included

in the definition of the other, then X and Y are distinct entities
and can exist apart from each other no matter how closely con-
nected they might be. Conversely, if one cannot conceive of X
apart from Y because one is included in the definition of the
other, then X and Y are not distinct entities, and cannot exist
apart from each other.

This principle is of some significance because it warrants the following

inferences. Assuming the necessity claim, if one can conceive of X apart
from Y then one can conclude that:

i. Neither can be reduced to the other because X and Y are not

identical.
ii. The existence of the one cannot be inferred from the existence of the
other, because either can exist without the other.
iii. One cannot be explained in terms of the other, again because one
can exist without the other, so one cannot be the cause of the other.
iv. Neither can be part of the mind-independent essence of the other,
because neither is included in the definition of the other.

Moores argument against naturalism in metaethics and the zombie argu-

ment against physicalism in the philosophy of mind exemplify (i); Humes
views on the nature of causation exemplify (ii); Nagels explanatory gap
argument exemplifies (iii); and Quines argument against essentialism
exemplifies (iv). Wittgensteins remarks in the Tractatus are a clear expres-
sion of the necessity claim, in the absence of which Kant would never have
entertained the Copernican Turn.
Now the first philosopher in the orthodox Western canon to employ the
principle of separability was Descartes, and it lies behind his most char-
acteristic theses.7 His first, and most decisive, intellectual move was to
abandon Aristotelian physics with its particular view of material bodies.
Aristotelian physics assumed that material bodies have a number of prop-
erties that play an explanatory role in the account of physical phenomena.
These properties included heat, cold, moisture, and dryness. By
contrast, Descartes insists that his physics will include no reference to such
properties, because they are not part of the essence of material bodies and
7
This point is recognised by Gendler and Hawthorne (2002).

2011 The Author

Metaphilosophy 2011 Metaphilosophy LLC and Blackwell Publishing Ltd
 THE MEDIEVAL ORIGINS OF CONCEIVABILITY ARGUMENTS 625

so do not figure in the explanation of physical phenomena. His material

bodies have only the properties of motion, size, shape, and arrangement of
parts.8 His argument for this claim is that he can conceive of material
bodies without their having the properties of heat, cold, moisture, and
dryness, but he cannot conceive of a material body without its being
extended. The same point is made again later in Principles of Philosophy
(part 2, section 11):

Suppose we attend to the idea we have of some body, for example a stone, and
leave out everything we know to be non-essential to the nature of body: we will
first of all exclude hardness, since if the stone is melted or pulverised it will loose
its hardness without thereby ceasing to be a body; next we will exclude colour,
since we have seen stones so transparent as to lack colour; next we exclude
heaviness, since although fire is extremely light it is still thought of as being
corporeal; and finally we will exclude cold and heat and all other such qualities,
either because they are not thought of as being in the stone, or because if they
change, the stone is not on that account reckoned to have lost its bodily nature.
After all, we will see that nothing remains in the idea of the stone except that it
is extended in length, breadth and depth. (1990, 227, emphasis added)

Descartes repeats the point in rule twelve of Rules for the Direction of the
Mind. Concerning simples, that is, entities that cannot be analysed into
more basic elements, he writes: The conjunction between these simple
things is either necessary or contingent. The conjunction is necessary when
one of them is somehow implied (albeit confusedly) in the concept of the
other so that we cannot conceive either of them distinctly if we judge them
to be separate from each other. It is in this way that shape is conjoined
with extension, motion with duration or time, etc., because we cannot
conceive of a shape which is completely lacking in extension, or a motion
wholly lacking in duration (1990, 46)
The point of these passages with respect to Cartesian physics is twofold.
First, they exhibit the principle of separability at work. Second, Descartes
decisive break with Aristotelian physics relies on the implementation of
this principle.
It is not just in physics that Descartes employs the principle of separa-
bility. The arguments of the First Meditation depend on the idea that no
proposition can be known unless it is logically impossible for it to be false.
This is most clearly seen in the malin gnie scenario. The principle of
separability prevents one from arguing from ones perceptions to the
nature, or even existence, of an external world because there is no con-
ceptual or logical connection between effects and causes that would
warrant such an inference. It is, after all, logically possible (because
entirely conceivable) that our perceptions are not caused by anything at

8
See The World, in Descartes 1990, 89.

2011 The Author

Metaphilosophy 2011 Metaphilosophy LLC and Blackwell Publishing Ltd
626 STEPHEN BOULTER

all, or caused by a malin gnie manipulating our minds in some fashion

rather than ordinary material objects in the external world stimulating our
sensory receptors. Consider also Descartes argument for the real distinc-
tion between mind and body. Descartes argues that his mind cannot be
identified with any material body, because he can formulate a clear and
distinct perception of their distinctness. At bottom, what Descartes is
claiming is that our concept mind does not include, nor is included in, our
concept body, and so one can conceive of one without the other. Thus
Descartes opening moves in physics, epistemology, and philosophy of
mind all depend upon his use of the principle of separability.
It is worth stepping back from Descartes characteristic claims for a
moment, and pausing to get clear about the fundamental assumptions
underlying the Cartesian framework. The following are the chief charac-
teristics of the Cartesian mind-set, some or all of which are at work in the
arguments noted in section 2.

1. Philosophers should accept only what is certain, or what follows

from what is certain by accepted rules of inference.
2. What is certain is indubitable. Indubitable propositions are either (i)
reports of the operations of ones own mind (e.g., I am currently
experiencing a red patch in my visual field, I am now thinking,
and so on) or (ii) those that could not possibly be false (e.g., those
that are necessarily true).
3. The only necessity is logical, conceptual, or semantic necessity.
4. Conceivability is a good guide to logical necessity and possibility.
5. What one can and cannot conceive of or imagine is an important
source of information on a range of substantive metaphysical and
epistemological issues.

It is clear that while some elements of the Cartesian mind-set are no longer
widely accepted, the passages noted in section 2 suggest that points (3) to
(5) are alive and well.9

4
Why trace this mind-set back to Descartes? The point becomes clear when
we recognise that this mind-set has not always found favour with philoso-
phers of the highest repute. Aristotle, for one, would not sign up to
propositions (1) to (5). However, as we saw in section 2, many important
post-Cartesian philosophers from very different schools have found them

9
However, in defence of the lingering effects of (1) and (2), many epistemologists are still
at work on what has become known as the traditional epistemological projectthe
project, to use Carnaps terminology, of constructing the world of physics, other minds, and
cultural objects on the basis of ones autopsychological experiences.

2011 The Author

Metaphilosophy 2011 Metaphilosophy LLC and Blackwell Publishing Ltd
 THE MEDIEVAL ORIGINS OF CONCEIVABILITY ARGUMENTS 627

to be entirely plausible, perhaps even compulsory. A question of some

significance, then, is when and why this cluster of claimsthe necessity
claim, the conceivability claim, and the principle of separabilityfirst
became philosophical commonplaces.
Two possibilities immediately present themselves. The first is that the
Cartesian mind-set comes into being with Descartes himself. This is the
view implicit in the popular vision of Descartes as a radical reformer of
the discipline. The other possibility is that significant events occurred in
the intellectual history of the West in the period between Aristotle and
Descartes, events that shed light on precisely this shift in philosophical
sensibilities. If this latter possibility were true, one would expect to find:
(1) evidence of such an event or events; (2) elements of the Cartesian
mind-set in the work of previous generations; and (3) evidence that Des-
cartes, either directly or indirectly, was exposed to these ideas. And this
is precisely what the historical record shows. The core elements of the
Cartesian mind-set can indeed be found in generations of previous think-
ers, and historical studies have already established that Europe was awash
in works containing these views at the relevant period. But what is par-
ticularly important for present purposes is the precise origin of these ideas.
The story is a long and complex one, spanning two millennia of philo-
sophical activity. I can only sketch the broad outlines of that story here,
and am forced to simplify so sweepingly at times that some distortion is
inevitable. Nonetheless, the essential features of the plot are clear enough.
Let us begin with the principle of separability itself. When did it first
strike thinkers in the West as plausible? A passage from Descartes Prin-
ciples of Philosophy provides the first clue:

Real distinction between two or more substances . . . is discovered from the

mere fact that we can clearly and distinctly conceive one without the other. For
when we come to know God, we are certain that he can do whatever we
distinctly understand. . . . [So] even if we supposed that God has conjoined
some corporeal substance to such a conscious substance so closely that they
could not be more closely conjoined, and had thus compounded a unity out of
the two, yet even so they remain really distinct. For however closely he had
united them, he could not deprive himself of his original power to separate
them, or to keep one in being without the other; and things that can be
separated, or kept in being separately by God are really distinct. (1990, 213)

The attitudes expressed in this passage were present right from the begin-
ning of Descartes career. In The World Descartes says, It is certain that
he [God] can create anything we can imagine (1990, 92). What is particu-
larly striking about this passage is that the principle of separability is
justified by overtly theological considerations. This is a significant clue to
its provenance. A close look at the historical record reveals that the
principle of separability gains a foothold in the Western philosophical
tradition because of very specific theological concerns prevalent in the late

2011 The Author

Metaphilosophy 2011 Metaphilosophy LLC and Blackwell Publishing Ltd
628 STEPHEN BOULTER

thirteenth and fourteenth centuries. In particular, it was events that came

to a head in 1277, the infamous Condemnations of tienne Tempier, then
bishop of Paris, which brought about a decisive shift in assumptions
regarding the metaphysics of modalities, a shift that quite naturally leads
to propositions (1)(5) coming to appear compulsory.
To appreciate the marked nature of this shift one needs to recall certain
features of the mind-set of the ancient Greek phusikoi, or natural philoso-
phers. The overriding concern of the phusikoi was to provide explanations
of physical phenomena. But they differed from mythologists in that they
sought explanations that were systematic, comprehensive, and economical
(i.e., they sought to explain all physical phenomena from the origins of the
universe to weather patterns using the same methods while appealing to
the fewest explanatory principles possible). But most important of all, they
sought explanations that relied solely on entities and processes found
within the natural order. In particular, they assumed that the gods, if there
were any, did not interfere in the course of nature. Internal principles and
features of the natural world were to do the explanatory work, not the
external, arbitrary, capricious, and ultimately inscrutable will of the gods.
The key explanatory notion employed by the phusikoi was nature.
Each natural object has its own inner principle of growth and motion.
These principles were thought to be intrinsic or essential features of the
object, and it was by adverting to these that one would explain the prop-
erties and behaviour of the natural object. But a crucial ingredient of the
natural philosophers notion of an explanation was that these properties
are rendered intelligible only if one can show that they could not have
been otherwise, given the nature of the object. The idea that genuine
knowledge of the natural world is of what is necessary in a non-logical
sense was thus a background assumption of the phusikoi, along with the
general assumption that the natural world was, in Epicuruss phrase,
governed by inescapable and merciless necessity.
That strict scientific knowledge is knowledge of what is necessary is
repeatedly insisted upon by Aristotle throughout his corpus. In book 1,
chapter 2, of the Posterior Analytics, for example, he writes: We suppose
ourselves to possess unqualified scientific knowledge of a thing, as
opposed to knowing it in the accidental way in which the sophist knows,
when we think we know the cause on which the fact depends, as the cause
of that fact and of no other, and, further, that the fact could not be other
than it is (1941, 111). In book 1, chapter 4, he writes: Since the object of
pure scientific knowledge cannot be other than it is, the truth obtained by
demonstrative knowledge will be necessary (1941, 115). But Aristotle did
not believe that scientific knowledge was confined to logic or mathematics,
as the following passage from the Physics (bk. 2, chap. 8) makes clear:
We must explain then . . . about the necessary and its place in physical
problems, for all writers ascribe things to this cause, arguing that since the
hot and the cold, &c., are of such and such a kind, therefore certain things

2011 The Author

Metaphilosophy 2011 Metaphilosophy LLC and Blackwell Publishing Ltd
 THE MEDIEVAL ORIGINS OF CONCEIVABILITY ARGUMENTS 629

necessarily are and come to be (1941, 249). Indeed, it would appear that
in this respect Aristotle was right about his ancient compatriots. For
example, Heraclitus maintained that all things take place in accordance
with strife and necessity, while Democritus held that nothing occurs
at random, but everything occurs for a reason and by necessity (Kirk,
Raven, and Schofield 1991, 193 and 419). Talk of inescapable and mer-
ciless necessityEpicuruss phraseis also present in Platos Timaeus
(1987, 47e). Such attitudes found their way into Roman thinking as well.
Ciceros report in On Fate clearly implies that it was only in the sphere of
human action that it ever occurred to anyone to doubt that the natural
world was governed by necessity (Inwood and Green 1997, 186). But the
main point is that, for Aristotle, that which cannot be otherwise clearly
extends well beyond the bounds set by logic. Moreover, at no point does
Aristotle maintain that ones ability to conceive of certain states of affairs
arising has any bearing on the nature of things in themselves. Most
emphatically, Aristotle did not believe that the mere ability to conceive of
an entity apart from another allowed one to infer the separability of the
two. As though anticipating Descartes argument for the real distinction
of mind and body, Aristotle insists in De Anima (bk. 2, chap. 2, 413b
2729) that the other parts of the soul are not separable, as some assert
them to be, though it is obvious that they are conceptually distinct (1941,
558). The principle of separability simply has no place in the Aristotelian
framework, and so there is no room for the inference patterns one finds in
Descartes and beyond.
Matters are more complicated when one moves into the Middle Ages.
Aquinas, like the phusikoi, insists that the explanation of physical phe-
nomena is to be based on the natures or powers of physical things.10 He
also agrees that strict knowledge (scientia) of the natural world can only be
had of those features of it that could not have been otherwise. And he goes
to some pains to insist that the natural world, despite its having been
created voluntarily by an omnipotent God, has diverse modes of non-
logical necessity to be found within it.11 However, Aquinass writings
contain seemingly contradictory passages on the principle of separability.
On the one hand, Aquinass criticism of Platos theory of Forms and the
separate existence of mathematical objects turns on the rejection of the
principle of separability. A telling passage in Aquinass Commentary on
Aristotles Metaphysics (bk. 1, l. 10) reads:

Therefore, considering the nature of the intellect, which is other than the nature
of thing known, the mode of understanding, must be one kind of mode, and the

10
Aquinas assumes this in his general metaphysics and epistemology, but argues explic-
itly for this against both Averroes and Avicenna in Summa Contra Gentiles (bk. 3, chap. 69,
in Aquinas 1997).
11
See in particular Summa Contra Gentiles, bk. 2, chap. 30 (in Aquinas 1997).

2011 The Author

Metaphilosophy 2011 Metaphilosophy LLC and Blackwell Publishing Ltd
630 STEPHEN BOULTER

mode of being, by which things exists, must be another. For although the object
which the intellect understands must exist in reality, it does not exist there
according to the same mode which it has in the intellect. Hence, even though
the intellect understands mathematical entities without simultaneously under-
standing sensible substances, and understands universals without understand-
ing particulars, it is not therefore necessary that the objects of mathematics
should exist apart from sensible things, or that universals should exist apart
from particulars. (1995, 5960)12

Although Aquinas is only pointing out here that mathematical objects and
universals need not exist separately from sensible things, the implication is
that their separability in thought tells us nothing about what might obtain
in extra-mental reality.
However, if Aquinas had one characteristic metaphysical doctrine it
was the positing of a real distinction between an objects essence and its act
of being. This was argued for in his early treatise On Being and Essence
using a now familiar inference pattern. He writes: Every essence or
quiddity can be understood without anything being known of its existing.
I can know what a man or a phoenix is and still be ignorant whether it
exists in reality. From this it is clear that the act of existing is other than
essence or quiddity, unless, perhaps, there is a being whose quiddity is its
very act of being (1986b, 46). Aquinass point here is that there is no
conceptual connection between an objects essence and its existence, and
from this he infers that an objects essence or nature is really distinct from
its act of being. But this is precisely what one would not expect Aquinas to
agree to, given the line of thought employed in the critique of Plato.
Clearly there is some ambivalence in Aquinas in regard to the principle of
separability, and this is due, I suggest, to tensions at the heart of the
Scholastic project.13
The main task of the Scholastic theologian was to produce a coherent
body of doctrine based on the narratives of the Old and New Testaments.
The inconsistencies and tensions to be found within sacred scripture, as
well as their challenges to ordinary common sense, are the problems
falling to the theologian qua theologian. However, one of Augustines
historically significant contributions to this effort was to argue that theo-
logians should seek to remove the inconsistencies whenever possible, and
to avail themselves of whatever might be of use in the work of pagan
philosophers in the pursuance of this task. It was this effort to understand
what was already accepted on faith that lead to the serious and sustained
12
See also Aquinas 1986a, q. 5, a. 3.
13
This is not the place to explore these tensions in Aquinas. But it would appear that
Aquinas uses the principle of separability only when the conclusions it supports can be
arrived at through other means. Thus he resists the line of thought sketched below by
insisting that what is possible for God is unknown to us, and so such possibilities cannot be
used in philosophical argument. See Summa Contra Gentiles, bk. 3, chaps. 56 and 59 (in
Aquinas 1997).

2011 The Author

Metaphilosophy 2011 Metaphilosophy LLC and Blackwell Publishing Ltd
 THE MEDIEVAL ORIGINS OF CONCEIVABILITY ARGUMENTS 631

effort on the part of the Scholastics to engage with and employ the work
of Aristotle once his entire corpus had been recovered. In the hands of
Aquinas and other Scholastics this developed into the concerted effort to
synthesise the theology of Augustine with the philosophy of Aristotle.
This project was not without its successes; but fundamental tensions
between Aristotelian and Catholic teaching could not be papered over.
Tensions eventually came to a head in 1277 when the bishop of Paris
issued a list of 219 condemned propositions, that is, propositions that
could not be taught within the University of Paris on pain of excommu-
nication. The condemned propositions of particular interest for present
purposes were those that threatened Gods omnipotence. The following is
a representative sample:

(a) That what is impossible absolutely speaking cannot be brought

about by God or by another agent. This is erroneous if we mean
what is impossible according to nature.
(b) That God cannot be the cause of a new-made thing and cannot
produce anything new.
(c) That God cannot move anything irregularly, that is, in a manner
other than that in which he does, because there is no diversity of
will in Him.
(d) That God cannot multiply individuals of the same species without
matter.
(e) That God could not make several intelligences of the same species
because intelligences do not have matter.
(f) That God could not move heaven in a straight line, the reason
being that He would then leave a vacuum.
(g) That God cannot produce the effect of a secondary cause without
the secondary cause itself.14

These propositions had been entertained by various thinkers taking Aris-

totle as their point of departure, and all posit a limit on Gods power. In
the wake of the condemnations of 1277, such thoughts were no longer
tolerated. And since on standard interpretations of omnipotence Gods
power is limited only by the principle of non-contradiction, it was incum-
bent upon all to accept as physically and metaphysically possible any
proposition that did not contain or entail a logical contradiction.15 It was
incumbent upon all to accept as physically and metaphysically possible any
proposition that did not contain or entail a logical contradiction. As
14
These are taken from Klima 2007, 18284.
15
Aquinas would prefer one to say that Gods omnipotence amounts to the ability to do
anything that is possible. If God cannot bring about a logical contradiction it is not because
of any incapacity in His power but because of the incapacity on the part of the logically
impossible to be brought into existence. See Summa Theologica, bk. 1, q. 25, a. 3 (in Aquinas
1997).

2011 The Author

Metaphilosophy 2011 Metaphilosophy LLC and Blackwell Publishing Ltd
632 STEPHEN BOULTER

Nicholas of Autrecourt would put it sixty years later, and more than two
hundred years before Descartes: Every being which does not contain an
incompatibility in its concept is possible (1971, 42).16 The upshot of the
condemnations for philosophy were: (1) that the notions of metaphysical
and physical necessity were ruled out as doctrinally unsound; and (2) since
strict scientific knowledge is only of what is necessary, knowledge is only
possible of logically necessary truths. The implications for Aristotelian
metaphysics and epistemology became clear in due course.
Soon after the condemnations Scholastics begin employing the princi-
ple of separability explicitly and consistently, and citing Gods omnipo-
tence in justification. Moreover, they begin to formulate theses commonly
associated with philosophers from the early modern period. Some clear-
cut examples are worth noticing in this regard. If one compares the views
of William of Ockham (c. 12851347) and Nicholas of Autrecourt (c. 1300
to after 1347) with the views of Aristotle on the one hand and Descartes on
the other, the considerable distance travelled in the direction of modernity
is obvious. Consider the central principles of the Ockhamist world view,
many of which are simply adopted from the condemnations:

1. All things are possible for God, save such as involve a logical con-
tradiction.17
2. Whatever God produces by means of secondary causes he can
produce and conserve immediately without their aid.18
3. God can save, conserve, and produce every reality, be it substance or
accident, apart from any other reality.19
4. We must not affirm that a proposition is true, or that something is
necessarily required for the explanation of an effect, if we are not
lead to this by a reasoning proceeding either from a truth known by
itself or from an experience that is certain.20

16
It is highly significant that in this same work Autrecourt writes: I declare that, neither
in this treatise nor in others, do I wish to say anything which is against the articles of faith,
or against the decision of the Church, or against the articles the opposite of which were
condemned at Paris (1971, 33).
17
This is a constant refrain in Ockham, and is usually grounded on the claim that I
believe in God the Father Almighty, the opening line of the Apostles Creed. See, for
example, the Fourth Quodlibet, Question 22, in Ockham 1991.
18
See Ockham 1991, Fourth Quodlibet, Question 22. An implication of this claim is that
God can cause one to perceive a tree, say, without the perception being caused by a tree. It
is striking that Ockham uses the term demon in an argument for scepticism based on the
consideration expressed in (2). See Ockhams Ordinatio I.27.3, in Pasnau 2002, 235.
19
See Ockham 1991, Fifth Quodlibet, Question 18. This point was particularly sensitive in
the context of the Eucharist. On Ockhams view that the bread and wine can become the flesh
and blood of Christ, making Christ literally present despite the fact that one continues to see
bread and wine. This is metaphysically possible because there is no contradiction in the claim
that the underlying substances can change while the accidents do not.
20
See Ockham 1991, Reportatio, II, Question 150.

2011 The Author

Metaphilosophy 2011 Metaphilosophy LLC and Blackwell Publishing Ltd
 THE MEDIEVAL ORIGINS OF CONCEIVABILITY ARGUMENTS 633

5. Everything that is real, and different from God, is contingent to the

core of its being.21

Consider also the views expressed by Nicholas of Autrecourt in his first

and second letters to Bernard of Arezzo, in which he draws out the
consequences of a post-1277 environment for Aristotelian physics and
metaphysics.22 Perhaps the most important feature of these letters is
Autrecourts use of a new inference pattern, a pattern that would have
struck Descartes as entirely natural. Autrecourt writes: I propose this
inference: It is possible, without any contradiction following therefrom,
that it will appear to you to be the case, and yet it will not be so. Therefore
you will not be evidently certain that it is the case (2006, 454). Here
Autrecourt has raised the bar to genuine knowledge in a fashion that
Descartes himself would have accepted. Only what cannot possibly be
false can be accepted as true. The road to sceptical conclusions is then
made easyone simply has to show that the contrary of a proposition is
logically possible to establish that the proposition cannot be known with
certainty. Autrecourt puts this principle to work, claiming: From the
fact that some thing is known to be, it cannot be inferred evidently, by
evidentness reduced to the first principle [i.e., the principle of non-
contradiction], that there is some other thing (455). This is a view one
now associates most readily with Hume, and Autrecourts reasoning is
essentially the same as the Scots: In such an inference in which from one
thing another thing would be inferred, the consequent would not be
factually identical with the antecedent, nor with part of what is signified by
the antecedent. It therefore follows that such an inference would not be
evidently known with the aforesaid evidentness of the first principle
(455).
In a particularly revealing example of this mind-set at work, Autrecourt
argues that one cannot know with certainty that individual substances
underlie observed properties. He writes: When a log or a stone has been
pointed out, it will be most clearly deduced that a substance is there, from
a belief accepted simultaneously. But this cannot be inferred from a simul-
taneous belief evidently. For, even if all kinds of things are perceived prior
to such discursive thought, it can happen, by some power, namely the
divine, that no substance is there. Therefore in the natural light it is
not evidently inferred from these appearances that a substance is there

21
This summary of Ockhams fundamental assumptions is taken from Boehner
1990.
22
Autrecourt does not necessarily endorse them, however. In fact he is explicitly asking
for advice on how to avoid these consequences, which he thinks inevitably follow from
positions sanctioned by the 1277 condemnations. He might be drawing out the implica-
tions of the condemnations of 1277 with a view to exploiting these consequences as a
reductio of positions popular among his contemporaries in Paris. See Zupka 1993, 191
221.

2011 The Author

Metaphilosophy 2011 Metaphilosophy LLC and Blackwell Publishing Ltd
634 STEPHEN BOULTER

(457).23 Having thus denied Aristotle the right to claim that properties
require an underlying subject, Autrecourt also shows in his first letter that
one cannot be sure the external world matches our sensory experiences in
any fashion whatsoever: Every impression we have of the existence of
objects outside our minds can be false, since . . . it can exist, whether or not
the object is (in Klima 2007, 134). The upshot of his reflections is that
Aristotle in his entire natural philosophy and metaphysics possessed such
certainty of scarcely two conclusions, and perhaps not even one. . . . I have
an argument that I am unable to refute, to prove that he did not even
possess probable knowledge (Autrecourt 2006, 457). In fact, according to
Autrecourt, the only substance Aristotle ever had any evident knowledge
about was his own soul, precisely the point Descartes would make nearly
three centuries later in the Second Meditation.24
I take it that elements of the Cartesian mind-set can be found at least in
germ form in earlier generations of thinkers, and that these are plausibly
linked with events that came to a head in 1277. It remains to be determined
whether there might have been a connection between these fourteenth-
century thinkers and Descartes. It is possible, of course, that Descartes
independently rediscovered many of these principles. But if there is reason
to believe that there was a link, however indirect, then the most natural
explanation of the similarity between their respective views is that the
former influenced the latter. And there is good reason to think that
Ockham, Autrecourt, and others working along the same lines managed
to influence latter generations of thinkers, if not directly, then at least
indirectly through the work of John Buridan. Buridan (13001358) was a
nominalist in the tradition of Ockham, and he discussed Autrecourts
arguments in works widely disseminated across Europe that formed an
important part of the curriculum in many universities well into the six-
teenth century.25 For example, in Quaestiones in Aristotelis Metaphysicam
(bk. 2, q. 1), where Buridan discusses whether scientific knowledge is
possible, one finds a series of sceptical arguments that are strikingly remi-
niscent of Descartes First Meditation and much of Hume as well, and
these arguments turn on points insisted upon by Ockham and Autrecourt.
There are traditional arguments against the reliability of the senses, which
Buridan then claims are made even more difficult by the articles of faith,
namely, that God can cause us to have visual experiences of objects

23
In 133031 Crathorn took a line similar to Ockham and Autrecourts. In his Questions
on the First Book of Lombards Sentences, Question 1, he argues that because the accidents of
the consecrated Host are indistinguishable from those of unconsecrated bread, one cannot
tell on the basis of sense experience whether ones visual experience of bread is caused by
bread, or the body of Christ, or, indeed, nothing at all. Crathorn 2002, 290.
24
The connection between the condemnations of 1277 and Cartesian scepticism has been
noted before, but the importance of the point has not been recognised. See Bermudez 2000
and Frede 1988.
25
See the preface to Pinborg 1976 and Markowski 1984 for able discussions of this point.

2011 The Author

Metaphilosophy 2011 Metaphilosophy LLC and Blackwell Publishing Ltd
 THE MEDIEVAL ORIGINS OF CONCEIVABILITY ARGUMENTS 635

that do not exist because this is consistent with divine omnipotence. He

writes:

For God can form in our senses the species of sensible things without these
sensible things, and can preserve them for a long time, and then . . . we would
judge those sensible things to be present. Furthermore, you do not know
whether God, who can do such and even greater things, wants to do so. Hence,
you do not have certitude and evidentness about whether you are awake and
there are people in front of you, or you are asleep, for in your sleep God could
make sensible species just as clear as, or even a hundred times clearer than,
those that sensible objects can produce; and so you would formally judge that
there are sensible things in front of you, just as you do now. Therefore, since
you know nothing about the will of God, you cannot be certain about anything.
(In Klima 2007, 143)

It is easy to see this passage as an early version of the infamous malin gnie
hypothesis of the First Meditation.26
One also finds in Buridan an early version of Humes problem of
induction:

Experiences do not have the force to conclude a universal principle, unless by

means of induction over many [singular cases]. But a universal proposition
never follows by induction, unless induction covers all singular cases of that
universal, which is impossible. Indeed, let us assume that whenever you touched
fire, you always felt it to be hot; therefore, by experience you judge the next fire,
which you have never touched, to be hot too, etc., and so finally you judge that
every fire is hot. Let us assume, then, by Gods will whenever you touched a
piece of iron, then you felt it to be hot. It is clear by parity of reasoning that
when you next see a piece of iron that is in fact cold, you will judge it to be hot.
And these would be false judgements, although at that point you would have
just as much experience about iron as you now in fact have about fire. (In
Klima 2007, 144).

Finally, it is worth noting that Autrecourts point about the inadmissibil-

ity of inferring causes from effects is given an airing in Buridans
discussion:

Again, a conclusion or an effect cannot be known through its cause, or a cause

through its effect, because the cause is not contained essentially or virtually in
its effect. . . . [So] it seems that we can never have evident knowledge of one

26
Consider also Spinozas remark in his Principles of Cartesian Philosophy regarding the
origins of the hypothesis. He writes: And there was a particularly strong reason, an ancient
belief, fixed in his [Descartes] mind, that there was an all-powerful God who had created
him, and so may have caused him to be deceived even regarding those things that seemed
very clear to him. This then is the manner in which he called everything into doubt (1998,
8).

2011 The Author

Metaphilosophy 2011 Metaphilosophy LLC and Blackwell Publishing Ltd
636 STEPHEN BOULTER

thing through another, because there is no evidentness, except by reduction to

the first principle, which is grounded in contradiction. However, we can never
have a contradiction concerning two diverse things: for let us assume that they
are A and B; then it is not a contradiction that A exists and B does not exist, or
that A is white and B is not white. Therefore, there will never be an evident
inference concluding that A exists from the fact that B exists, and so on for
other cases. (In Klima 2007, 144)

This is simply a repetition of Autrecourts fundamental point on which he

bases his critique of Aristotles claims to know anything at all. But the
point to note here is that the views of Ockham and Autrecourt did get an
airing well into the sixteenth century and beyond, views that gained wide-
spread currency only after the condemnations of 1277.
A further point of interest: Buridan rehearsed these arguments, but he
did not accept their sceptical conclusions. He agreed that acceptance of
Gods omnipotence does indeed lead to the consequences Autrecourt so
ably drew. But Buridan maintained that the philosopher could afford to
ignore these results. He writes: It is in this way [i.e., on the assumption
that things obey the common course of nature] that it is evident to us that
every fire is hot or that the heavens are moving, although the opposite is
possible by Gods power. And this sort of evidentness is sufficient for the
principles and conclusions of natural science (in Klima 2007, 14546).
Indeed, he rebukes those who do not settle for this degree of evidentness:
Therefore we conclude as a corollary that some people speak very
wrongly, wanting to destroy the natural and moral sciences on the
grounds that their principles and conclusions are often not absolutely
evident, but can be falsified by supernatural possibilities. Because such
sciences do not require absolute evidentness, but the above mentioned
kinds of non-absolute evidence or conditional evidentness suffices (in
Klima 2007, 146). Apparently Buridan did not feel the need to meet
sceptical arguments based on supernatural possibilities. They might
command the attention of theologians, he implies, but they need not detain
natural philosophers.
One final link between the condemnations of 1277 and Descartes is
worth mentioning. Descartes does not betray any direct acquaintance with
the works of Ockham, Autrecourt, or Buridan, but the first philosophical
author [he] came across was certainly familiar with this Scholastic herit-
age.27 And in the course of his Metaphysical Disputations, Suarez relies
on the cluster of assumptions whose history we have been tracing. For
example, in section 6 of Disputation 31 we see the necessity claim at work
in the blurring of the distinction between logical and metaphysical neces-
sity: A thing capable of being created implies merely the lack of contra-
diction, or logical possibility (in Ariew, Cottingham, and Sorell 1998,
27
This is what Descartes says of Francis Suarez in the course of his fourth set of replies
(2008, 164).

2011 The Author

Metaphilosophy 2011 Metaphilosophy LLC and Blackwell Publishing Ltd
 THE MEDIEVAL ORIGINS OF CONCEIVABILITY ARGUMENTS 637

45). And when explicating the difference between modal and real distinc-
tions Suarez relies again on the powers of God to ground the difference. If
two putative entities are only modally distinct, they cannot be separated
even by Gods infinite power, which, again, is limited only by the principle
of non-contradiction: The reason for this is that if one of the two terms
is such that it cannot, even through the absolute power of God, be con-
served without the other, this is a great argument for its being only a mode
and not a true entity. For if it were a true entity, it could not have such an
intrinsic dependence on another entity that God could not make up for it
by his own infinite power (Disputation 7, in Ariew, Cottingham, and
Sorell 1998, 49). Just how well Descartes knew the works of Suarez is a
matter of some debate; but there is no doubt that Suarez deeply influenced
Descartes teachers at La Flche, and so a line of influence can be traced
from the condemnations of 1277 to Descartes. What is not so clear is why
Descartes did not follow in the footsteps of Buridan but instead adopted
the theologically motivated course taken by Ockham and Autrecourt.28
Whatever the answer to that question might be, by insisting on absolute
certainty, by assuming that the only necessity is logical necessity, and by
accepting the principle of separability, Descartes imported into the natural
sciences and philosophy assumptions and inference patterns more natu-
rally at home in the theology of the thirteenth and fourteenth centuries. In
short, the father of modern philosophy did not leave the Middle Ages
behind but continued to operate with the theologically grounded assump-
tions of the Augustinian reactionaries of 1277.

5
It is time to consider some objections. What can be concluded on the basis
of this history? First, and perhaps most obviously, one is likely to wonder
whether this history is philosophically significant at all. After all, if the
ground for a favoured belief is rendered problematic, one can always hope
to support that belief on other grounds.
In reply I should argue, first, that philosophical progress is made if one
comes to hold a true belief for the right reasons. So if a historical study
prods a philosopher along this road, inducing him to substitute good
reasons for bad, then that study has philosophical value. So while one
might very well find non-theological grounds for upholding the principle
of separability, I would claim that this in and of itself is no small gain.
Moreover, the very fact that one now recognises the need to find
any grounds at all for the cluster of assumptions in question constitutes
28
It is more than possible that the explanation has something to do with the fact that
Descartes is even more extreme than I have portrayed him here, in that he denies even the
existence of logical necessity. However, as was kindly pointed out by an anonymous referee
for Metaphilosophy, it is difficult to reconcile Descartes view that the only necessity is
conformity to the will of God with his, or any, philosophical practice.

2011 The Author

Metaphilosophy 2011 Metaphilosophy LLC and Blackwell Publishing Ltd
638 STEPHEN BOULTER

philosophical progress. Very often philosophical success amounts to little

more than shifting the burden of proof onto the opposition. If a historical
study allows one to shift the burden of proof, then clearly that historical
study has philosophical value. And what is quite remarkable about the
principle of separability and the related assumptions is how often they are
simply taken for granted.
One might also consider, if only for a moment, the prospects of finding
alternative grounds for the Cartesian mind-set. What, apart from the
intellectual inertia associated with periods of normal science, could induce
one to accept the metaphysical assumption that there is no necessity
outside of logic, or that our imagination is a guide to all real possibilities?
It is impossible to enter into this question fully here, but it is worth noting
that there is no obvious answer. In fact most of our pre-theoretical intui-
tions are likely to be more at home in the pre-1277 mind-set. Do we really
believe that properties can float free of the substances in which they
inhere? Do we really believe that iron could fail to be hard at room
temperature? Do we really believe that there is no distinction between
accidental regularities and genuine laws of nature, and that the latter are
underwritten by the genuinely causal properties of real things? And does
any of this really have anything to do with logic? These rhetorical ques-
tions are all that space allows, and are only the start of what needs to be
a much longer conversation; the point for now is merely that there are no
immediately obvious non-question-begging grounds on which to build a
case for the claim that the only necessity is logical necessity.
Perhaps it will be objected that philosophers have been able to subject
the assumptions of the Cartesian mind-set to serious scrutiny, and in many
cases to reject them, without the benefit of any historical study. So is this
historical study really necessary? There are several responses to this line of
questioning. The first is simply to grant that while there are many routes
to the same conclusion, it is nonetheless significant that philosophical
conclusions can be reached on historical grounds. There is no question
here of historical methods replacing traditional techniques; all that is being
asserted is that historical studies should be included in the philosophers
toolkit. Secondly, while it is true that elements of the Cartesian mind-set
have been scrutinised and rejected by various philosophers without the
benefit of historical studies, what this case study allows one to do is place
a question mark over the entire mind-set at once by seeing it as a self-
contained whole with a common historical origin. Rather than having to
rest content with piecemeal attacks, the case study allows for the consid-
eration of a more systematic rejection of an entire mind-set that has
animated philosophy for centuries.
Finally, there is that apparently most un-philosophical of questions. Is
the history sketched in section 4 true? To raise this question in a philo-
sophical context, however, is to concede that purely historical facts might
have strictly philosophical significance, and so my methodological point

2011 The Author

Metaphilosophy 2011 Metaphilosophy LLC and Blackwell Publishing Ltd
 THE MEDIEVAL ORIGINS OF CONCEIVABILITY ARGUMENTS 639

has been made. But on the purely factual question, the honest answer is
that the historical jury is still out. Much remains to be uncovered about
the late Scholastic period in general and the lines of influence extending
into the early modern period in particular. Section 4 has only traced
connections between (1) pervading views regarding the nature of necessity
in the ancient world; (2) ambiguities on such matters in Aquinas just prior
to the condemnations of 1277; (3) the background to the condemnations
themselves, and their injunctions on matters relating to the metaphysics of
modalities; (4) the link between the injunctions of the condemnations and
the surprisingly modern theses defended by Ockham and Autrecourt; and
(5) the discussion of those theses in the works of Buridan, works widely
available in the sixteenth century and beyond, which were discussed by
thinkers known to Descartes. Whether this history will withstand scrutiny
is a question with important philosophical implications. But it will only be
answered by further historical research on the long neglected period domi-
nated by the Scholastics.

Department of History, Philosophy and Religion

Oxford Brookes University
Harcourt Hill Campus
Oxford OX2 9AT
United Kingdom
sboulter@brookes.ac.uk

References
Aquinas, Thomas. 1986a. The Divisions and Methods of the Sciences.
Fourth edition. Translated by Armand Maurer. Toronto: Pontifical
Institute of Medieval Studies.
. 1986b. On Being and Essence. Translated by Armand Maurer.
Toronto: University of Toronto Press.
. 1995. Commentary on Aristotles Metaphysics. Translated by John
P. Rowan. Notre Dame: Dumb Ox Books.
. 1997. Basic Writings of Saint Thomas Aquinas. Volume 2. Edited by
Anton Pegis. English Dominican Translation corrected by Anton
Pegis. Indianapolis: Hackett.
Ariew, Roger, John Cottingham, and Tom Sorell (eds.). 1998. Descartes
Meditations: Background Source Materials. Cambridge: Cambridge
University Press.
Aristotle. 1941. The Basic Works of Aristotle. Edited by Richard McKeon.
New York: Random House.
Autrecourt, Nicholas. 1971. The Universal Treatise. Translated by
Leonard Kennedy, Richard Arnold, and Arthur Millward. Milwaukee:
Marquette University Press.

2011 The Author

Metaphilosophy 2011 Metaphilosophy LLC and Blackwell Publishing Ltd
640 STEPHEN BOULTER

. 2006. The Second Letter to Bernard of Arezzo. In Basic Issues in

Medieval Philosophy, second edition, edited by Richard Bosley and
Martin Tweedale, 45357. Peterborough: Broadview Press.
Bermudez, Jos Luis. 2000. The Originality of Cartesian Scepticism: Did
It Have Ancient or Medieval Antecedents? History of Philosophy
Quarterly 17: 33360.
Boehner, Philotheus. 1990. Ockham: Philosophical Writings. Indianapolis:
Hackett.
Braddon-Mitchell, David, and Frank Jackson. 2007. Philosophy of Mind
and Cognition: An Introduction. Oxford: Blackwell.
Chalmers, David. 1996. The Conscious Mind. Oxford: Oxford University
Press.
. 2002. Does Conceivability Entail Possibility? In Conceivability
and Possibility, edited by Tamar Gendler and John Hawthorne, 145
200. Oxford: Oxford University Press.
Crathorn, William. 2002. Questions on the First Book of Lombards Sen-
tences. In Pasnau 2002, 245301.
Descartes, Ren. 1990. The Philosophical Writings of Descartes. Volume 1.
Translated by John Cottingham, Robert Stoothoff, and Dugald
Murdoch. Cambridge: Cambridge University Press.
. 2006. A Discourse on Method. Translated by Ian Maclean. Oxford:
Oxford World Classics.
. 2008. The Philosophical Writings of Descartes. Volume 2. Trans-
lated by John Cottingham, Robert Stoothoff, and Dugald Murdoch.
Cambridge: Cambridge University Press.
Frede, Michael. 1988. A Medieval Source of Modern Scepticism. In
Gedankenzeichen: Festschrift fr Klaus Oehler zum 60. Geburtstag,
edited by R. Claussen and R. Daube-Schackat, 6570. Tbingen:
Stauffenburg.
Friedman, Russell, and Lauge O. Nielson. 2010. The Medieval Heritage in
Early Modern Metaphysics and Modal Theory, 14001700. Dordrecht:
Kluwer Academic.
Gendler, Tamar, and John Hawthorne (eds.). 2002. Conceivability and
Possibility. Oxford: Oxford University Press.
Hume, David. 1985. Enquiries Concerning Human Understanding. Third
edition. Edited by J. A. Selby-Bigge, with text revised and notes by
P. H. Nidditch. Oxford: Clarendon Press.
. 1989. A Treatise of Human Nature. Second edition. Edited by J. A.
Selby-Bigge, with text revised and variant readings by P. H. Nidditch.
Oxford: Clarendon Press.
Inwood, Brad, and L. P. Green (eds.). 1997. Hellenistic Philosophy.
Second edition. Indianapolis: Hackett.
Kant, Immanuel. 1933. Critique of Pure Reason. Second edition. Trans-
lated by Norman Kemp-Smith. London: Macmillan.
Klima, Gyula (ed.). 2007. Medieval Philosophy. Oxford: Blackwell.

2011 The Author

Metaphilosophy 2011 Metaphilosophy LLC and Blackwell Publishing Ltd
 THE MEDIEVAL ORIGINS OF CONCEIVABILITY ARGUMENTS 641

G. S. Kirk, J. E. Raven, and M. Schofield. 1991. The Presocratic Philoso-

phers. Cambridge: Cambridge University Press.
Markowski, M. 1984. Linfluence de Jean Buridan sur les universits
dEurope centrale. In Preuve et raisons luniversit de Paris, edited by
Z. Kaluza and P. Vignaux, 14963. Paris: Vrin.
Moore, G. E. 1993. Principia Ethica. Cambridge: Cambridge University
Press.
Nagel, Thomas. 1974. What Is It Like to Be a Bat? Philosophical Review
83: 43550.
Ockham, William of. 1991. Quodlibetal Questions. Edited by Alfred
Freddoso and Francis Kelley. New Haven: Yale University Press.
Pasnau, Robert (ed.). 2002. The Cambridge Translations of Medieval
Philosophical Texts. Volume 3, Mind and Knowledge. Cambridge: Cam-
bridge University Press.
Pinborg, Jan (ed.). 1976. The Logic of John Buridan: Acts of the European
Symposium on Medieval Logic and Semantics. Copenhagen: Museum
Tusculanum.
Plato. 1987. Timaeus. In Collected Dialogues, edited by Edith Hamilton
and Huntington Cairns, translated by Benjamin Jowett. Princeton:
Princeton University Press.
Quine, Willard van Orman. 1953. Reference and Modality. In From a
Logical Point of View, 13959. Cambridge, Mass.: Harvard University
Press.
Spinoza, Baruch. 1998. Principles of Cartesian Philosophy. Translated by
Samuel Shirley. Indianapolis: Hackett.
Williamson, Timothy. 2007. The Philosophy of Philosophy. Oxford:
Blackwell.
Wittgenstein, Ludwig. 1986. Tractatus Logico-Philosophicus. Translated
by C. K. Ogden. London: Routledge and Kegan Paul.
Yolton, John. 1983. Thinking Matter: Materialism in Eighteenth-Century
Britain. Minneapolis: University of Minnesota Press.
Zupka, Jack. 1993. Buridan and Scepticism. Journal of the History of
Ideas 31: 191221.

2011 The Author

Metaphilosophy 2011 Metaphilosophy LLC and Blackwell Publishing Ltd