CHRISTOPH WITTICH’S ANTI-SPINOZA
Alexander Douglas
Anti-Spinoza, written by the theology professor at Leiden University and promoter of Cartesian
philosophy, Christoph Wittich, is one of the earliest and most detailed critiques of Spinoza’s
Ethics. Published in 1690, it is both a criticism of the Ethics and, by way of favourable
comparison, a piece of advocacy for a Cartesian method in philosophy. Christiane Hubert
describes Anti-Spinoza as a work ‘intended no doubt to become a sort of teaching manual –
precise, meticulous, exhaustive.’1 Theo Verbeek speculates that it was compiled out of notes that
Wittich had prepared in order to give private tuition on Cartesian philosophy.2 For this purpose,
Spinoza’s Ethics is employed as an object lesson in how not to do philosophy and subjected to a
point-by-point criticism and commentary.
Anti-Spinoza should therefore be of interest to scholars of both Spinoza and Descartes. But,
with the exception of Hubert and Verbeek, it has been largely overlooked by recent scholars. In
this article I aim to critically address Wittich’s critique of Spinoza’s philosophical method. Not
only does Anti-Spinoza begin with a discussion of method, many of its later arguments against the
Ethics return to the question of methodology. For example, Wittich’s criticism of the third part of
the Ethics questions the methodological soundness of Spinoza’s proposal to examine human
actions and appetites in more geometrico.3 Central to Wittich’s major critique is a discussion of
the role of second notions (notiones secundae) in Spinoza’s philosophical demonstrations. Wittich
takes the key terms in the Ethics, such ‘substance’, ‘attribute’, ‘accident’, ‘mode’, ‘cause’,
‘effect’, and so on, to refer to second notions, and criticizes Spinoza for drawing metaphysical
conclusions from their application to various objects existing outside the intellect. Second notions,
according to Wittich, describe our ways of conceiving of such objects and not to the objects
themselves. To draw inferences about how the objects really are from facts about how they are
conceived involves an obvious fallacy. This criticism is valid as an argument. But, I shall argue, it
is unsound because its first premise is false. Spinoza’s key terms do not refer to second notions.
They refer, rather, to what Spinoza calls ‘common notions’, which I shall endeavour to explain.
In what follows I shall closely follow Verbeek’s analysis both of Wittich’s argument and its
various ambiguities and difficulties, with one important difference that I shall explain. I shall then
provide an interpretation of a section of the Ethics in order to reveal how Wittich’s critique misses
its target. Finally, I shall use a crucial example to show how the demonstrations of the Ethics,
interpreted in what I take to be the correct way, do not appear formally very different from those
used by Descartes in the Meditations. Wittich’s attempt to draw a sharp and salutary distinction
between the methods of Spinoza and Descartes is, to this extent, a failure.
ANALYTIC VS. SYNTHETIC METHOD
Anti-Spinoza opens with a general discussion of philosophical method, beginning with a
distinction between analytic and synthetic method – the latter being, we are told, the one
1
C. Hubert, Les premières réfutations de Spinoza : Aubert de Versé, Wittich, Lamy (Paris: Presses de
l'Université de Paris Sorbonne, 1994), 27.
2
T. Verbeek, ‘Wittich's Critique of Spinoza,’ in Receptions of Descartes: Cartesianism and AntiCartesianism in Early Modern Europe, edited by T. M. Schmaltz (London: Routledge, 2005), 114.
3
C. Wittichius, Anti-Spinoza; sive examen Ethices B. de Spinoza, et Commentarius de Deo et ejus
attributis. (Epistolæ.) (Amsterdam, 1690) 159. ‘An quoque possumus more Geometrico demonstrare
connexiones cogitationum, quae sunt in diversis hominibus de eodem objecto? An eas, quae sunt in eodem
homine de eodem objecto diverso tempore? Unde verò est, quod sermonem habentes de eadem re tàm
diversa de illâ dicant? An quis demonstrabit more Geometrico morborum naturam eorumque symptomata
tàm varia tam diversa in diversis hominibus, & in eodem homine diversis temporibus? Cùm omnia ista
dependeant à diversis circumstantiis particularibus & dispositione ac qualitate corporis, quae est in
homine?’
2 A. DOUGLAS
employed by Spinoza in the Ethics.4 The opening passage is a very close paraphrase of a section
of Descartes’s replies to the objections posed by Marin Mersenne to the Meditations (the Second
Replies).5 I shall begin with a discussion of the latter, which helps to highlight a point that
becomes important for Wittich.
In his objections, Mersenne asks Descartes why, in the Meditations, he did not treat his subject
‘in geometrical order’, the order he employed so effectively in his mathematical studies.6
Descartes replies that ‘the geometrical manner of writing’ involves two things – order (ordo) and
method (ratio). Order is the requirement that:
[T]he items which are put forward first must be known entirely without the aid of what
comes later; and the remaining items must be arranged in such a way that their
demonstration depends solely on what has gone before.7
Descartes asserts that he did follow order in the Meditations. But as far as the method of
demonstration goes, there are two kinds: the analytic and the synthetic. In the Meditations, he
asserts, he favoured the analytic method. Analytic method, he explains:
[...] shows the true way by means of which the thing in question was discovered
methodically and as it were a priori [tanquam a priori], so that if the reader is willing to
follow it and give sufficient attention to all points, he will make the thing his own and
understand it just as perfectly as if he had discovered it for himself. 8
Synthetic method, meanwhile:
[…] demonstrates the conclusion clearly and employs a long series of definitions,
postulates, axioms, theorems, and problems, so that if anyone denies one of the
conclusions it can be shown at once that it is contained in what has gone before. 9
Descartes adds that synthetic method proceeds ‘as it were, a posteriori [tanquam a posteriori]
(though the proof itself is often more a priori than it is in the analytic method).’10 It is not obvious
what Descartes means by the terms a priori and a posteriori here. John Cottingham’s footnote to
the translation of this passage notes that Descartes’s use of the term a priori in his discussion of
analytic method:
[...] seems to correspond neither with the post-Leibnizian sense (where a priori truths are
those which are known independently of experience), nor with the medieval, Thomist
sense (where a priori reasoning is that which proceeds from cause to effect). What
Descartes may mean when he says that analysis proceeds ‘as it were a priori (tanquam a
priori) is that it starts from what is epistemically prior, i.e. from what is prior in the ‘order
of discovery’ followed by the meditator.11
However, when Descartes makes the point about synthetic method quoted above, Cottingham
speculates that:
Wittichius, Anti-Spinoza ‘De Methodo Demonstrandi’, sig. Ar. There are no page numbers in this
introductory section, thus I refer to the collating figures.
5
R. Descartes, Oeuvres de Descartes, edited by C. Adam and P. Tannery, 12 vols. (Paris: Vrin, 1974), vol.
7, 155-9. R. Descartes, The Philosophical Writings of Descartes, translated by J. Cottingham, R. Stoothoff,
and D. Murdoch, 3 vols. (Cambridge: Cambridge University Press, 1985), vol. 2, 100-113.
6
Descartes, Oeuvres, vol. 7, 128; Descartes, Philosophical Writings, vol. 2, 92.
7
Descartes, Oeuvres, vol. 7, 155; Descartes, Philosophical Writings, vol. 2,110.
8
Descartes, Oeuvres, vol. 7, 155; Descartes, Philosophical Writings, vol. 2, 110.
9
Descartes, Oeuvres, vol. 7, 156; Descartes, Philosophical Writings. vol. 2, 111.
10
Descartes, Oeuvres, vol. 7, 156; Descartes, Philosophical Writings, vol. 2, 110-111.
11
Descartes, Philosophical Writings, vol. 2, 110, fn.2.
4
WITTICH’S ANTI-SPINOZA 3
[...] Descartes may mean that though the proofs involved are a priori (viz., in the
traditional, Thomist sense), the method of synthesis starts from premisses which are
epistemically posterior – i.e., which are arrived at later in the order of discovery. 12
Cottingham has a strong rationale for suspecting that when Descartes speaks of methods
proceeding ‘as it were’ a priori and a posteriori he cannot mean simply a priori/posteriori in the
post-Leibnizian or the Thomistic sense. To claim that Descartes means a priori/posteriori in the
post-Leibnizian sense would be historically absurd. As for the Thomistic sense, the relevant
criterion, according to Descartes, for demarcating a demonstration as tanquam a priori or
tanquam a posteriori is whether or not it follows the order of discovery, which, at least prima
facie, seems independent of the criterion for judging whether or not a demonstration is a priori or
a posteriori in the Thomistic sense, namely whether it proceeds from cause to effect. In fact, I
would generalize Cottingham’s explanation and say that a priori reasoning in the Thomistic sense
proceeds from whatever is the real reason for something to whatever it is the real reason for.13 The
term ‘cause’ is often used in this very broad sense to mean simply the reason for things being the
way they, as in fact Wittich points out at a certain point.14
This, however, means that Cottingham’s explanation of the order of discovery as beginning
with what is ‘epistemically prior’, creates a certain ambiguity by the use of the latter modern term.
Anybody who had any formal training in philosophy in the seventeenth century – and this
certainly includes both Descartes and Wittich – would presumably be aware of the Aristotelian
distinction between things that are better known by us and those that are ‘better known by nature’
(γνωριμωτέρων φύσει). The distinction is made, for instance, near the beginning of the Physics.15
Things better known by us are those whose concepts are most familiar to us, or those whose
concepts we are capable of forming first. Things better known by nature are those that are the real
reasons for whatever they are invoked to explain. Does ‘epistemically prior’ mean better known
by us or better known by nature? If it means the latter, then the order of epistemic priority, that is,
the order of discovery, may well be the same as the order of causation (or explanation) in the
Thomistic sense.
Descartes’s proposal that demonstrations following the order of discovery allow the reader to
know things as though he is discovering them for himself suggests that such demonstrations begin
with what is best known to us. Moreover, when he prepares a synthetic presentation of the
Meditations at Mersenne’s request, the definitions and axioms with which he begins involve many
concepts that appear to be explanatorily fundamental but are also highly abstract and unfamiliar to
an unprepared reader, such as ‘objective reality’, ‘formal and eminent existence’, and so on. Such
concepts may be best known by nature but are certainly not best known by us. Thus Descartes
seems to be implying that the key criterion distinguishing a demonstration tanquam a priori from
one tanquam a posteriori is whether or not it begins with what is best known by us, not best
known by nature. Thus the distinction is different from that between a priori and a posteriori
demonstrations in the Thomistic sense. As for his comment that synthetic demonstrations are
often more a priori than analytic ones, clearly ‘a priori’ cannot here mean the same as ‘tanquam a
priori’ earlier, and Cottingham’s proposal that here Descartes is at this point using ‘a priori’ in
the Thomistic sense, to refer to reasoning that proceeds from cause to effect, is highly plausible.
An examination of Descartes’s geometrical presentation and some of his other comments in
the Second Replies suggests a certain interpretation of what he means by this. Earlier in these
replies, while elaborating the cogito argument, he points out that when somebody infers his own
12
Descartes, Philosophical Writings, vol. 2,111 fn.1.
This seems to line up fairly well with Aquinas’ own way of thinking, as the following passage shows:
‘[…] duplex est demonstratio. Una quae est per causam, et dicitur propter quid, et haec est per priora
simpliciter. Alia est per effectum, et dicitur demonstratio quia, et haec est per ea quae sunt priora quoad
nos, cum enim effectus aliquis nobis est manifestior quam sua causa, per effectum procedimus ad cognitionem causae.’ T. Aquinas, Summa Theologica (Rome: Forzani, 1894) 1.2.2.
14
At least, it is used in this wide sense by geometers: ‘Et sane apud Geometros vox Causa adeo late
sumitur, ut quamvis designet rationem, propter quam talis proprietas, alicui rei competat [...].’ Wittichius,
Anti-Spinoza , 8.
15
Aristotle, Aristoteles Opera Graece ex recensione Imannuelis Bekkeri, edited by Academia Regia
Borrusica, (Berlin: Georg Reimer, 1831), 184a116-118.
13
4 A. DOUGLAS
existence from the fact that he is thinking, he is not reasoning by way of any syllogism. If he were
reasoning syllogistically, he would need to know the major premise: ‘Everything which thinks is,
or exists’.16 But, Descartes explains, such a person in fact arrives at his conclusion ‘from
experiencing in his own case that it is impossible that he should think without existing.’ Descartes
goes on: ‘It is in the nature of our mind to construct general propositions on the basis of our
knowledge of particular ones.’17 Thus, when Descartes presented the cogito analytically in the
Meditations, he began with a first-person account of his own intuition about his own particular
case, and not with the general proposition that everything that thinks exists. When he later
presents his reasoning synthetically, however, he begins with general propositions, for instance
the proposition that ‘a real attribute cannot belong to nothing’.18 This is even more general than
the proposition that everything that thinks exists; indeed, the latter can be derived from it.
Here then, perhaps, is the reason for Descartes’s belief that synthetic demonstrations are
tanquam a posteriori: they do not follow the order of discovery because they begin from general
propositions and derive particular ones from these, whereas ‘it is the nature of our mind to
construct general propositions from particular ones’. On the other hand, synthetic demonstrations
may often be more a priori than analytic ones, since general propositions are often the real
explanations for particular ones. For instance: I may learn that thinking entails existing first in my
own case and later in the general sense, but the general rule explains my particular case and not
the other way around.19
The reason this is important here is that Wittich seems to suppose that the analytic method is
more a priori than the synthetic, not simply tanquam a priori. Close as his discussion of analytic
and synthetic method is to that of Descartes, it diverges from its model in claiming that the
analytic method ‘follows the order of nature accurately, proving nothing except by the true causes
and true principles of things’,20 whereas:
The synthetic method often deviates from the order of nature, supposing it to be
unnecessary to order propositions and demonstrations in the way it would have to if it
followed that order; thus things are often demonstrated in quite difficult ways, not as
things are really discovered, but rather by far-fetched, forced, and artificially contrived
and structured means, of which many examples can be found in Euclid. 21
To say that only the analytic method always follows the order of nature, that is, the order of ‘the
true causes and true principles of things’ is, in effect, to say that only the former always proceeds
a priori in the Thomistic sense.22 This does not accord with Descartes’s own views, at least as I
have interpreted them. For him, as we have seen, the synthetic method is often more a priori than
the analytic, since it deduces particular propositions from general ones, which often means that it
follows the order of real explanatory priority, though not the order of discovery. Does Wittich,
16
Descartes, Oeuvres, vol. 7, 140; Descartes, Philosophical Writings, vol.2, 100
Descartes, Oeuvres, vol. 7, 140 ; Descartes, Philosophical Writings, vol. 2, 100
18
Descartes, Oeuvres, vol. 7, 161; Descartes, Philosophical Writings, vol. 2, 114
19
That is to say, ‘all As are Bs’ explains why ‘if x is an A then x is a B’, whereas the latter particular
proposition cannot explain why the former general one holds true. This is different from saying that ‘all As
are Bs’ explains ‘this A is a B’. In that case one may take the former, general proposition as being nothing
more than a conjunction of propositions of the latter, particular kind, meaning that the latter (along with
others of the same kind) explains the former. The difference is that in the first case the latter proposition
contains an entailment, which requires further explanation, whereas in the second case the latter proposition
is, arguably, explanatorily basic: there is nothing to be said to explain why ‘this A is a B’ besides the fact
that this A is a B.
20
Wittichius, Anti-Spinoza , ‘De Methodo Demonstrandi’, sig. A2r. [...] Methodus Analytica ordinem
naturae accurate observat, nihil probat nisi per rerum veras causas atque vera principia [...].’
21
Wittichius, Anti-Spinoza ‘De Methodo Demonstrandi’, sig. Ar-A2r. ‘Sic vero Methodus Synthetica saepe
multum deviat ab ordine naturae, dum supponit, non opus esse, ut alius servetur ordo, quam ut priores
propositiones possint inservire ad demonstrandas sequentes, unde saepe res demonstrantur per vias
admodum difficiles, & non natura duce repertas, sed longe petitas, coactas, & non nisi artificio aliquo
paratas & structuras, cujus rei varia exempla in Euclide licet reperire.’
22
Verbeek makes roughly the same claim about Wittich’s divergence from Descartes on this point.
Verbeek, ‘Wittich's Critique of Spinoza’, 122.
17
WITTICH’S ANTI-SPINOZA 5
then, reject Descartes’s view that general propositions are often prior to particular ones? Yet who
could seriously deny that the order of explanation often runs from propositions like ‘everything
that thinks exists’ to propositions like ‘if I think then I exist’?23
One need not make such a radical proposal about Wittich as long as one gives due weight to an
explanation he later gives of why synthetic method diverges from the order of nature. This
explanation makes specific reference to second notions. I turn to it now.
SPINOZA’S KEY TERMS AS SECOND NOTIONS
At the beginning of the main part of Anti-Spinoza, Wittich again refers to the order of nature, this
time while criticizing Spinoza’s methodological choice to begin with a consideration of second
notions. ‘If we are to follow the order of nature’, he insists ‘second notions cannot be examined
before primary notions’.24 As examples of second notions, Wittich gives ‘substance’, ‘essence’,
‘subject’, ‘attribute’, ‘mode’, ‘cause’, and ‘effect’.25 Second notions, he claims:
[…] are simply general ways of considering things […]. They do not fall under the
common notions and intellect of men, but rather are invented by philosophers who
establish a collection of really existing things, or of things which exist only in thought or
as sensory appearances, to be compared amongst each other. 26
This vague description does little to explain precisely what Wittich means by ‘second notions’.
Verbeek aims to explain this further by tracing the origins of the term to Zabarella and the
nominalist tradition in Scholastic philosophy.27 Verbeek quotes Zabarella:
[…] first notions are names [nomina] which immediately signify things by means of a
concept of the mind, like ‘animal’ and ‘man,’ that is, the very concepts of which those
names are signs; second notions are other names imposed upon the first, like ‘genus,’
‘species,’ ‘name,’ ‘verb,’ ‘proposition,’ ‘syllogism,’ and so on, or rather on the concepts
signified by those (first) names.28
Second notions are, Verbeek suggests, ‘the elements of what we would call a “meta-language”’.29
Like Wittich in the above quotation, Zabarella claims that ‘second notions are our own work and
fictions of our minds.’30 Verbeek’s exegesis of Zabaralla continues:
If anything, it is Descartes’s contention that the order of discovery runs from the particular to the general
that is somewhat controversial. Ramus, for instance, rejects this view as expressed by Aristotle. See Petrus
Ramus, ‘Qvod Sit Vnica Doctrinae Institvendae Methodvs, Locvs E IX,’ in Petri Rami Veromandui
animadversionum aristotelicarum libri XX. Ad Carolum Lotharingum, cardinalem Guïsianum (Paris, 1548),
3-4.
24
Wittichius, Anti-Spinoza, 7: ‘Notiones secundae non possunt spectari ante primas, si ordinem naturae
sequamur [...].’
25
Wittichius, Anti-Spinoza 7.
26
Wittichius, Anti-Spinoza, 7: ‘Notiones secundae [...] sunt enim tantum modi quidam considerandi
generales [...] neque cadunt sub communem notitiam & intellectum hominum, sed inventae sunt à
Philosophis collatione institutâ rerum particularium sive extantium inter se invicem, vel etiam cum rebus
non extantibus nisi in cogitatione vel apparentia sensum.’
27
The details of this historical account can be found in Verbeek, ‘Wittich's Critique of Spinoza,’ 118-119.
28
J. Zabarella, ‘De natura logicae,’ in Opera Logica, anastatic reprint of the 1597 Cologne edition, edited
by W. Risse) (Hildesheim: Georg Olms, 1966), I.3, 6., cit. Verbeek, ‘Wittich's Critique of Spinoza,’ 118119.
29
Verbeek, ‘Wittich's Critique of Spinoza,’ 119. Verbeek does not explicitly say that this is what second
notions are. He claims that logic, according to Zabarella, deals in the elements of meta-language, with the
implicit understanding that logic deals exclusively in second notions. Thus I hope I have interpreted him
rightly with what I say in the main text.
30
Zabarella, ‘De naturae logicae’, I. 3, 6, cit. Verbeek, ‘Wittich's Critique of Spinoza,’ 119.
23
6 A. DOUGLAS
[…] since logic deals with second notions only, there is no good reason why it should be
called a science, given the fact that ‘science relates only to things that are necessary’.
Dealing with man made things logic is more like an art or technique. It is not knowledge
of things although it helps us to produce knowledge of things.31
What Verbeek seems to mean in saying that second notions are elements of meta-language is that
they refer to our ways of conceiving of things rather than to things themselves. To designate a
word by the meta-linguistic term ‘verb’ is to define the way in which that word is used by
competent speakers of the language; likewise, to designate an idea by a second notion is to define
the conceptual role that that idea plays within a certain scheme. The reason that logic is said by
Zabarella not to be a science relating to ‘things that are necessary’ is, presumably, that how we
choose to conceive of things is up to us, whereas the way that things outside of our intellects are is
not up to us and is ‘necessary’ in this sense.
One consequence of this is that logical propositions, which apply second notions, cannot by
themselves support any conclusions about how things are, independently of our ways of
conceiving of them. This is for the simple reason that such propositions merely describe our ways
of conceiving. The proposition, for example, that every species subsists within a genus does not
tell us anything about the things thus classified, except that they are thus classified.
Wittich’s claim that synthetic demonstrations reject the order of nature is therefore likely to be
based upon the belief that such demonstrations begin with the application of second notions. Most
would agree that it is things being the way they are that (at least partly) causes us to conceive of
them in certain ways, certainly not our conceiving of them in certain ways that causes them to be
the way they are. To say the latter would be to deny the order of nature, taking the effect for the
cause or the explanandum for the explanans. But to begin a demonstration with a consideration of
second notions is to risk implying that our ways of conceiving of things are explanatorily
fundamental, thus denying the order of nature. For this reason Wittich claims that:
[…] if Spinoza had wanted to proceed distinctly and in an orderly fashion, he should have
first considered those things which fall under the common intellect – God, Mind, etc. –
and only after observing their real attributes formed the various second notions – Cause,
Substance, etc.32
It appears, then, that Wittich’s basic methodological critique of the Ethics is that Spinoza reverses
the order of nature by concluding facts about how things are from facts about how they are
conceived of, the latter expressed in second notions.
This helps, for instance, to explain Wittich’s response to proposition 5 of Part 1 of the Ethics,
which states that ‘in nature there cannot be two or more substances of the same nature or
attribute’. Wittich points out that:
This proposition is false if ‘of the same nature or attribute’ is understood to refer to
specific identity rather than numerical identity. […] To it I respond, that two substances
could be distinguished by the diversity of their individual attributes, which may be
considered as affections or modes of the substances as generally conceived. Thus the
minds of Peter and Paul [are distinguished], likewise two bodies. 33
Yet Spinoza seems to have pre-empted this response by claiming that ‘substance is prior in nature
to its affections [so that] if the affections are put to one side and [the substance] is considered in
Zabarella, ‘De naturae logicae’, I.3, 6F-7A, 7D-E, 8B-C, 8E. cit Verbeek, ‘Wittich's Critique of Spinoza,’
119.
32
Wittichius, Anti-Spinoza, 8: ‘Quod si igitur distincte & ordine voluisset progredi Spinoza particulares res
prius debuisset considerasse, quae cadunt sub intellectum communem, Deum, Mentem, &c. & post realia
observata attributa etiam notiones secundas varias Causae, Substantiae, &c. formare.’
33
Wittichius, Anti-Spinoza, 49: ‘Falsa est propositio si intelligatur illa eadem natura sive attributum de
identitate specifica, non vero de identitate numerica. [...] Ad quæ respondeo, duas substantias posse
distingui ex diversitate attributorum individualem, quæ possunt considerari tanquam affectiones sive modi
substantiæ prout generatim concipitur. Ita mentes Petri & Pauli. Ita duo corpora.’
31
WITTICH’S ANTI-SPINOZA 7
itself […] one cannot be conceived to be distinguished from another.’34 An explanation of why
Wittich is unmoved by this argument is that he interprets the proposition ‘substance is prior in
nature to its affections’, as merely describing how we conceive of things, with ‘substance’ and
‘affections’ as second notions. It may mean, for instance, that we predicate affections of
substances and not the other way around. Such a proposition cannot entail any fact about the real
identity conditions of things outside of our intellect, for no such fact could follow simply from
how we conceive of things.35
A clearer example of the same line of thinking is found in Wittich’s objection to proposition7,
Part 1, which states that ‘it pertains to the nature of a substance to exist.’ Wittich reads this as
expressing the standard view that a substance is by definition something that exists per se, and
comments as follows:
When […] a substance is said to be in itself and by itself, this is understood to be in
thought – in the intellect. That is, we understand a substance to be this way, but it cannot
be determined by our having this concept of it that it exists. Likewise if we say a whole is
that which is composed of all its parts, this does not allow us to infer that therefore the
whole exists, since while it shows us what a whole is, this is understood to mean how it is
in our intellect, that is, how a whole exists in our intellect. Likewise, however much we
truly have a clear and distinct idea of what substance is like, we cannot thereby say that a
substance exists in that way. If we have a clear and distinct idea of an Angel – that which
is a mind separated from a body – we are not from this idea allowed to conclude that an
Angel exists in such a way.36
In other words, the assertion that something is a substance does not concern the thing itself; it
concerns how the thing is ‘in our intellect’, that is, how we conceive of it. To say that something
is a substance is to say that we conceive of it as existing per se, just as to call something a whole
is to assert that we conceive of it as something composed of its parts, or to call something an
angel is to claim that we believe it to be a mind separated from a body. But to conceive of things
thusly does not make them so. This is important because Spinoza’s first proof of God’s existence,
the demonstration for proposition 11, Part 1 depends almost entirely on proposition 7, Part 1. It
states, in effect, that since God is a substance, and since it pertains to the nature of substance to
exist, God must exist. For Wittich, this must be the wrong way around: to say that God is a
substance is to say that we consider him to exist per se; one reason for this is, presumably, that he
does exist per se. But certainly our considering him to exist in that way is not the reason for him
doing so. Thus Spinoza’s proof takes the effect for the cause.
Wittich’s objection would not hold if Spinoza’s claim was that a substance is something that
does exist per se, or, as Spinoza puts it, something to whose nature existence pertains, rather than
simply something that we consider to be per se. The proof for proposition 11 would then be,
roughly, that since God is something to whose nature existence pertains, he must exist necessarily.
Such a proof has many problems, apparent circularity not least among them, but it is not subject to
34
B. de Spinoza, Ethics, translated by E. Curley (Princeton, NJ: Princeton University Press, 1985) Part 1,
proposition 5, demonstration, p.48. The page numbers given for this work throughout the article refer to
Volume 2 of B. de Spinoza, Opera, edited by C. Gebhardt (Heidelberg: Carl Winters, 1924). Curley’s
translation includes these page numbers as marginal notes.
35
For that matter, even if Spinoza were right that two substances that differed only in their affections could
not be conceived to be distinct, it still would reverse the order of nature to conclude on this basis that those
substances cannot be distinct. Wittich would not bother to make this argument, however, since, as the above
quotation shows, he believes that substances in such a case can be conceived as distinct, and indeed often
are.
36
Wittichius, Anti-Spinoza, 49: ‘Quando ergo substantia dicitur esse in se & per se, intelligitur sic esse in
cogitatione, in intellectu, hoc est, nos per substantiam tale quid intelligere quod an existat nec ne ex hoc
nostro conceptu non possumus determinare, quemadmodum, si quis dicat, totum est id quod componitur ex
omnibus suis partibus non licet inde inferre: Ergo existit totum, quia dum sic esse tribuimus toti, id tantum
intelligendum est de nostro intellectu, quod scilicet sic totum sit in nostro intellectu. Talem veram h.e.
claram & distinctam ideam habemus de substantia, non tamen possumus propterea dicere, quod substantia
talis existat. Sic habemus ideam claram & distinctam de Angelo, quod sit mens separata à corpore, nec
tamen ex illa ipsa idea licet colligere, quod Angelus existat.’
8 A. DOUGLAS
Wittich’s critique.37 For it draws its inference from something that is true about God, not from
something that is conceived of God. When, in the Fifth Meditation, Descartes proves God’s
necessary existence from the inconceivability of his nonexistence, he goes on:
It is not that my thought makes it so, or imposes any necessity of the thing itself; on the
contrary, it is the necessity of the thing itself, namely the existence of God, which
determines my thinking in this respect.38
If Spinoza is not using ‘substance’ as a second notion, there is no reason he would not be entitled
to say the same thing. Here I diverge from Verbeek, who construes Wittich’s critique as follows:
[…] by simply reflecting on the notion of substance one cannot legitimately claim that a
substance exists because the notion of substance implies necessary existence (or,
inversely, that if we find something that exists necessarily it is a good candidate for being
categorized as a substance).39
This is, it seems to me, a less charitable way of interpreting Wittich’s argument. Certainly one
cannot conclude ‘by reflecting on the notion of substance’ that a substance exists. But Spinoza
does not attempt any such inference. He draws his conclusion partly from the consideration that
God is a substance, which is to say that he reflects not only on the notion of substance but also on
the notion of God. Wittich, who has clearly read the Ethics with great care, is surely aware of this.
Moreover, Verbeek’s way of construing the argument does not require it to depend on the view
that ‘substance’ is a second notion. Even if ‘substance’ means ‘thing existing per se’ rather than
‘thing conceived of as existing per se’, it would remain the case that simply reflecting on this
definition could not prove the existence of any substance. No existential conclusions can be drawn
from a mere definition. If this were all Wittich meant, it would not explain why he emphasizes the
fact that the definitions of ‘substance’ and ‘whole’, and the clear and distinct idea of an angel,
express only how such objects exist ‘in our intellect’. This is to say that such ideas express how
we conceive of things; it is not to say that such ideas describe how things must be if they do exist.
For only the first statement concerns our intellect; the second concerns what is possible,
irrespective of our intellect.
Indeed, Verbeek might not really disagree with my construal of Wittich’s argument, since he
claims that ‘[a]ll Wittich’s arguments – those at any rate which I have discussed – can be reduced
to an argument about the difference between first notions and second notions.’40 This will be true
if Wittich’s argument is what I construe it as: that because ‘substance’ is a second notion, calling
God a substance is simply to say that he is conceived of in a certain way, and no fact about God
(including the fact of his existence) could follow from a mere conception in our intellect.
My construal also explains why, as Verbeek notes, Wittich believes that synthetic
demonstrations aim at ‘persuasion rather than illumination’.41 While the fact that I conceive of
something as being a certain way certainly does not explain its being that way, somebody hoping
Hubert gives the impression that other early critics of Spinoza also took the word ‘substance’ to describe
how a thing is being conceived, rather than simply the way it is. Dom François Lamy, for example,
criticizes Spinoza for not realizing that ‘to say that a substance is conceived in itself is to say no more than
that its concept excludes all others.’ Confusion on this point is said to be the basis of many of Spinoza’s
demonstrations. Aubert de Versé gives a specific example: a substance is not conceived along with its
cause, but Spinoza illegitimately concludes from this that a substance must be uncaused. However, ‘my
spirit can consider a substance simply in itself as substance, as a thing existing in itself, by ignoring
absolutely whether it was produced or not.’ That is to say, a substance is conceived independently of any
cause, but it does not follow that it is independent of any cause. Thus, again, Spinoza is said by these critics
to have illegitimately drawn conclusions about the way a substance must be from the way we conceive of
substance. This is some evidence that ‘substance’ was generally taken to be a second notion, at least to an
extent. Hubert, Premières Réfutations, 52.
38
Descartes, Oeuvres, vol. 7, 67; Descartes, Philosophical Writings, vol. 2, 46
39
Verbeek, ‘Wittich’s Critique of Spinoza’, 121.
40
Verbeek, ‘Wittich’s Critique of Spinoza’, 121.
41
Verbeek, ‘Wittich’s Critique of Spinoza’, 115. See Verbeek’s footnote here.
37
WITTICH’S ANTI-SPINOZA 9
to persuade me that it is really thus could do well to show how this is entailed by my own
conceptions. For instance, if I conceive of God as a being per se, and if I can be shown that it
follows from this that God exists necessarily, somebody could use a synthetic demonstration
based on the fact that God is a substance (in Wittich’s second-notion sense of ‘substance’) to
convince me that God exists necessarily. Such a demonstration would not really prove anything. It
would draw out the implications of my own conceptions without showing them to be correct. But
it would admirably serve the purpose of ‘persuasion rather than illumination’. This may be what
Wittich means when he says that, since the synthetic method does not follow the order of nature
in the way the analytic method does, demonstrationes ad absurdum appear frequently in the
former but not in the latter. For, Wittich proposes, ‘such [demonstrations] can convince a mind,
but do not bring any clarity.’42 Demontrationes ad absurdum, that is to say, may reveal what one
must concede in order to hold consistent beliefs, but do not in themselves show what is really the
case.
Even as pieces of persuasion, however, synthetic demonstrations based on second notions will
only work if the target of the persuasion agrees to conceive of things in the manner described by
the second notions. Wittich believes that Spinoza’s proofs do not even meet this standard. Spinoza
himself, Wittich argues, distinguishes between a true definition referring to a real object and a
definition that merely expresses how a certain term is understood. This, Wittich says, corresponds
to ‘the distinction that others make between real definitions and nominal definitions’. 43
Definitions in terms of second notions can only be nominal. But people would not generally agree
with many of Spinoza’s definitions.44 Thus his synthetic demonstrations, though they would in
any case not constitute genuine proofs, cannot even serve the purpose of persuasion.
I shall now endeavour to show that Spinoza’s key terms do not, in fact, refer to second notions,
and that Wittich’s objections are therefore misdirected.
SPINOZA’S KEY TERMS AS COMMON NOTIONS
Spinoza is, perhaps, not sufficiently careful to avoid giving the impression that ‘substance’, as he
defines it, is a second notion. The definition he gives is: ‘By substance I understand what is in
itself and is conceived through itself, that is, that whose concept does not require the concept of
another thing from which it must be formed.’45 In stating that a substance is not only something
conceived through itself but also what is in itself, he seems to be blocking Wittich’s reading of
‘substance’ as a second notion. But as he elaborates on this, he returns to speaking only about the
way things are conceived, again inviting the reading that ‘substance’ is a second notion.46
Nevertheless, the damage that this reading does to his central proofs is enough to make it
questionable, simply as a matter of interpretative charity.
The issue is, however, made more complex by the fact that Spinoza excuses himself in the
Ethics from discussing ‘those notions they call Second’, since he has ‘set these aside for another
treatise, and [does] not wish to give rise to disgust by too long a discussion’.47 There is no record
that this other treatise was ever written. Nevertheless, this statement in itself tells us something. It
seems improbable that Spinoza would find a discussion of second notions peripheral to the Ethics
if his crucial demonstrations were really based upon them. Moreover, at the same place, he lists
second notions and ‘the axioms founded on them’ alongside ‘other things I have thought about
Wittichius, Anti-Spinoza, ‘De Methodo Demionstrandi’, sig. A2 v ‘Atque haec etiam est causa, quare
Methodus Analytica non ita utatur demonstrationibus ad absurdum, quae frequentissime locum habent in
Methodo Synthetica, quia tales mentem quidem convincere, sed nulla claritate affundere queunt.’
43
Wittichius, Anti-Spinoza, sig. A3r ‘Haec est eadem distinctio cum ea, qua alii utuntur inter Definitionem
rei & definitionem vocis.’ Wittich quotes Spinoza’s 1663 letter to Simon de Vries, Epistola VIII, in
Spinoza, Opera, vol. 4, 39-40.
44
Examples are given at: Wittichius, Anti-Spinoza ‘De Methodo Demonstrandi’ A3r-v, 14-15, 16-17.
45
Spinoza, Ethics, Part I, definition 3, p.45.
46
Hubert shows that other critics of Spinoza also paid exclusive attention to the ‘conceived through itself’
part of his definition of ‘substance’ and ignored the ‘in itself’ part. See footnote 37 above.
47
Spinoza, Ethics, Part 2, proposition 40, first scholium, p.120.
42
10 A. DOUGLAS
from time to time [et alia, quae circa haec aliquando meditatus sum]’. It would be somewhat odd
to speak of issues of direct importance to the argument one is currently constructing as ‘things I
have thought about from time to time’.
What makes the matter even more confusing is that Spinoza goes on, ‘in order not to omit
anything that it is necessary to know’, to discuss what he calls ‘Transcendental terms’ and
‘Universal terms’, leaving open the possibility that he regards these as species of second notions.
This would mean he understands the latter very differently from Wittich, since it is clear that he
regards transcendental and universal terms as terms referring to universals – classes of things
(albeit rather vaguely defined) – rather than to ways of conceiving of things. It is also highly
unlikely that he would deliberately base his reasoning on these ideas, since he regards them as the
product of confusion. But since they are not what Wittich calls ‘second notions’, it is possible that
the latter are in fact the foundations of his reasoning. There is, however, as I shall now endeavour
to show, a more plausible candidate for what Spinoza takes his key terms such as ‘substance’,
‘attribute’, and ‘mode’ to be, namely what he calls ‘common notions’.
In explaining these, Spinoza asserts, first, that ‘[t]hose things which are common to all things,
and which are equally in the part and in the whole, can only be conceived adequately’, 48 and next
that:
[I]f something is common to, and peculiar [proprium] to, the human body and certain
external bodies by which the human body is usually affected, and is equally in the part
and in the whole of each of them, its idea will also be adequate in the mind. 49
Spinoza goes on to assert that ‘[w]hatever ideas follow in the mind from ideas which are adequate
in the mind are also adequate.’50 Having said all this, Spinoza concludes: ‘With this I have
explained the cause of those notions which are called common, and which are the foundations of
our reasoning.’51 This implies that ideas of whatever is common to all bodies, or peculiar to the
human body and the external bodies by which it is usually affected, and whatever ideas are
formed by seeing what follows from such ideas are all common notions. It also implies that such
ideas, being ‘the foundations of our reasoning’, are the basis of the demonstrations in the Ethics;
that is at least a plausible way of understanding what Spinoza means by ‘our reasoning’. It is
likely, therefore, that ‘substance’, ‘attribute’, and so on are common notions for Spinoza. I shall
now elaborate this point.
Clearly one example of a common notion is the idea of God’s essence. The idea of this can
only be adequate, Spinoza explains, because ‘[e]ach idea of each body, or of each singular thing
which actually exists, necessarily involves an eternal and infinite essence of God.’52 Thus ‘what
gives knowledge of an eternal and infinite essence of God is common to all, and is equally in the
part and in the whole.’53 It fits the definition of a common notion, and therefore must be adequate
for Spinoza.
By the same reasoning, however, ‘substance’, ‘attribute’, and ‘mode’ are likely to be the
objects of common notions. It is a common feature of all bodies that they are modes of the
attributes of a single substance.54 If God is the object of a common notion because every singular
thing involves an eternal and infinite essence of God, the same must surely be said of God’s
attributes, since, Spinoza says, the ideas of singular things ‘must involve the concept of their
attribute.’55 If we can form a common notion of God’s essence, we must likewise be able to form
a common notion of God’s attributes; an attribute, after all, is simply ‘what the intellect perceives
of a substance, as constituting its essence.’56 But it is difficult to imagine forming an adequate
48
Part 2, proposition 38, p.118.
Part 2, proposition 39, p.119.
50
Part 2, proposition 40, p.120.
51
Part 2, proposition 40, first scholium, p.120.
52
Part 2, proposition 45, p.127.
53
Part 2, proposition 46, demonstration, p.127-8.
54
Part 2, definition 1, p.84; Part 1, proposition 25, corollary, p.68; and passim.
55
Part 2, proposition 45, demonstration, p.127.
56
Part 1, definition 3, p.45.
49
WITTICH’S ANTI-SPINOZA 11
idea of God’s attributes without knowing what an attribute is, that is, having an adequate idea of
‘attribute’ in general. Indeed, it is difficult to imagine forming an adequate idea of God’s essence
without having an adequate idea of both ‘substance’ and ‘essence’. A substance, after all, is what
God is; an essence is what his essence is.
By parity of reasoning, since all singular things are alike in being modes, the property of being
a mode is also something ‘peculiar to the human body and certain external bodies by which the
human body is usually affected’, and will also therefore be the object of a common notion.
Likewise, the ideas of cause and effect can be common notions, since Spinoza believes that it is a
universal property of all bodies that, as finite modes, they have a cause, and that they cause other
things – that is, they all possess both the property of being a cause and the property of being an
effect.57
The supposition that such ideas are common notions provides an attractive alternative to
Wittich’s view that they are second notions. Rather than referring to our ways of conceiving of
things, such terms might refer to things of particular kinds, or to the properties that define the
kinds of things they are. ‘Substance’, for example, might be the idea of a thing to whom existence
pertains, or of the property of being such a thing, rather than the idea of our way of conceiving of
certain things. This is likely to be what Spinoza believes, since he regards common notions as the
foundations of our reasoning. Given the presence of this alternative hypothesis, one that renders
Spinoza’s arguments a great deal more consistent, plausible, and intelligible, there seems no
motivation to persevere with Wittich’s hypothesis that Spinoza’s reasoning is grounded on second
notions. Wittich is simply wrong to suppose that Spinoza’s key terms refer to concepts that ‘do
not fall under the common notions and intellect of men’.58
CONCLUSION
I have proposed an interpretation of Wittich’s methodological argument against Spinoza, running
roughly as follows: Synthetic demonstrations in metaphysics tend to violate the order of nature
because they begin with the definitions of second notions. Since second notions describe our ways
of conceiving things, placing them at the start of a demonstration invites the reader to suppose that
our ways of conceiving of things explain how things are in themselves, rather than the other way
around. Such demonstrations may be persuasive, but they are invalid as genuine proofs. All
Spinoza’s key metaphysical propositions are demonstrated in this invalid way (and they fail even
at persuasion).
Since Spinoza’s key terms – ‘substance’, ‘attribute’, ‘mode’, ‘cause’, ‘essence’, and so on –
refer to common notions rather than second notions, Wittich’s argument misses its target. To
return to an earlier example, Spinoza’s proof of God’s existence, based on the fact that God is a
substance, is not vulnerable to Wittich’s critique if ‘substance’ is taken to refer to a common
notion meaning ‘thing per se’, rather than to a second notion meaning ‘thing conceived of as
being per se’. The inference that God exists necessarily because he is a substance does not reverse
the order of nature by deducing a fact about an object outside the intellect from a fact about the
way we conceive of it.
As I mentioned, there are problems with this proof even when it is interpreted rightly. Since
the proposition that God is a substance (in what I take to be Spinoza’s sense of ‘substance’) seems
to be very nearly equivalent to the proposition that God exists necessarily, the proof appears to be
circular, using its conclusion as a premise. Nevertheless, it does not seem worlds away from
Descartes’s proof in the Fifth Meditation. Descartes, as I mentioned, argues that although his
proof of God’s existence infers that God necessarily exists from the fact that one cannot conceive
of God not existing, this fact about conceivability is not given as the reason for God’s necessary
existence, rather the converse:
57
58
Part 1, proposition 16, p.60 and proposition 28, p.69.
Wittichius, Anti-Spinoza , 7.
12 A. DOUGLAS
It is not that my thought makes it so, or imposes any necessity of the thing itself; on the
contrary, it is the necessity of the thing itself, namely the existence of God, which
determines my thinking in this respect.59
Conceivability is determined by possibility and is therefore a guide to possibility. It does not
determine possibility. Likewise, Spinoza’s first proof at proposition 11, Part 1 explains the fact
that God cannot be conceived of except as existing by the fact that God is a substance – a thing to
whose nature existence pertains. He invites the reader to suppose that God does not exist, then
shows that because God is a substance, such a supposition leads to absurdity: the necessity of
God’s existence determines the absurdity rather than the absurdity determining the necessity.
What Spinoza does not give is an argument for the claim that God is a substance, that is, that
definition 6 of Part 1 is correct. But, as we have seen, Spinoza believes that we necessarily
possess an adequate idea of God; thus, because our idea of God is an idea of a substance, and we
know that this idea is adequate, we perhaps have the required premise. Similarly in the Fifth
Meditation we are assumed at the beginning of the proof to know by a clear and distinct idea that
God is an entia summe perfecti.60 This is where circularity appears in both proofs: Descartes,
arguably, cannot assume the truth of all clear and distinct ideas without assuming the existence of
God, while Spinoza’s proposition that the human mind necessarily forms an adequate idea of God
arguably depends for its demonstration upon the presumed truth of a metaphysical theory in
which God plays an integral part.61
Naturally these are all topics for further discussion, but the parallel is clear enough to suggest
that, minus the unfounded assumption that Spinoza’s key terms refer to second notions, Spinoza’s
reasoning and that of Wittich’s favoured philosopher are not wildly dissimilar. According to
Verbeek, Wittich’s ‘critique of Spinoza helps us to neutralize the unfruitful and uninteresting
textbook idea of Spinoza as being in some way a “radical Cartesian”’.62 If Wittich’s analysis of
the Ethics were correct, Spinoza would be committed to a methodological mistake that Descartes
seems to avoid, and this would provide a good reason to reject the characterization of Spinoza as a
Cartesian. But as things stand, the issue is rather less clear. There may be many things wrong with
what Verbeek calls the textbook idea of Spinoza, but the objection following from Wittich’s
methodological critique cannot be one of them.
Jacobsen Fellow, Institute of Philosophy and King’s College, London.
59
Descartes, Oeuvres, vol. 7, 67; Descartes, Philosophical Writings, vol. 2, 46
Descartes, Oeuvres, vol. 7, 65; Descartes, Philosophical Writings, vol. 2, 45
61
The charge of circularity is levelled against Descartes in both the second and fourth objections – see
Descartes, Oeuvres, vol. 7, 125; Descartes, Philosophical Writings, vol. 2, 89 and Descartes, Oeuvres, vol.
7, 214; Descartes, Philosophical Writings, vol. 2, 150. Descartes’s not entirely illuminating reply is at
Descartes, Oeuvres, vol. 7, 140-146; Descartes, Philosophical Writings, vol. 2,100-105. A fallacy in
Descartes’s reply is presented in A. Kenny, ‘The Cartesian Circle and the Eternal Truths’, Journal of
Philosophy, 67:19 (1970), 685-700. Some replies to Kenny are: F. Feldman and A. Boyd Levison, ‘Anthony
Kenny and the Cartesian Circle,’ Journal of the History of Philosophy, 9:4 (1971), 491-6; D. Odegard,
‘Escaping the Cartesian Circle’, American Philosophical Quarterly, 21:2 (1984), 167-173. Kenny replies to
Feldman and Levison in A. Kenny, ‘A Reply by Anthony Kenny’, Journal of the History of Philosophy, 9:4
(1971), 497-8. I do not know of any sources discussing the more subtly embedded but similar circle in
Spinoza’s reasoning.
62
Verbeek, ‘Wittich's Critique of Spinoza,’ 125.
60