1. Introduction
Spinoza believes that God or Nature, the only substance, has several distinct essential and infinite attributes, such as Thought or Extension. The question is how they are distinguished. The nature of the distinction between attributes has been a matter of persistent debate in Spinoza literature. It is commonly believed that for Spinoza each essential attribute is one way of conceiving the essence of the only substance, and hence that for him attributes must be numerically identical. This reading has profound implications. Since Spinoza believes that the mind is a modification of the attribute of Thought, and the body is a modification of Extension, if Thought and Extension are numerically distinct attributes, the mind and the body are one and the same thing conceived in different ways.
Spinoza did not, in my view, hold that different attributes are, in fact, numerically identical but here I will only show that a main argument of Spinoza's commentators to the effect that he held that view is mistaken. This argument is based on the claim that when accounting for the distinction between a substance and its essential attributes Spinoza has in mind Francisco Suárez's distinction of reason, a distinction that does not exist in reality and that somehow must involve an intellect.Footnote 1 Let us call this principle:
(Distinction of Reason Between Essential Attribute and Substance): The distinction between any single essential attribute of a substance and that substance must be a distinction of reason.
In turn, three issues seem to back up (Distinction of Reason Between Essential Attribute and Substance):
(1) The Metaphysical Thoughts. In the Metaphysical Thoughts (CM), one of his early works, Spinoza seems to accept this principle when he posits that: ‘that distinction is said to be of reason which exists between substance and its attribute, as when duration is distinguished from extension. And this is also known from the fact that such a substance cannot be understood without that attribute’ (CM, II.5).Footnote 2
(2) Descartes’ Shadow. In the CM Spinoza follows Descartes in positing that there are only three ways in which two things can be distinguished: a modal distinction, a real distinction, and a distinction of reason (CM, II.5). Two essential attributes, such as Thought or Extension, cannot be modally distinct because a modal distinction requires that at least one of its involved relata is a modification or affection of an attribute. Examples of a modal distinction are the distinction between my mind – an affection or mode of the attribute of Thought – and my body – a mode of Extension –, or the distinction between the attribute of Thought and one of its modes, such as my mind. Consequently, Thought and Extension must either be really distinct or else they must be rationally distinct. But they cannot be really distinct, since for Spinoza this distinction only applies between substance and for Spinoza there is only one substance. Hence, they must be rationally distinct.
(3) The Intellect in the Definition of Attributes. Finally, although Spinoza does not mention the distinction of reason in works other than the CM, in the Ethics he defines attributes as ‘what the intellect perceives of a substance as constituting its essence’. Since the intellect is involved in a distinction of reason, this definition is commonly seen as evidence that he still believes that attribute and substance are distinguished by reason.
And yet, it is a mistake to hold that Spinoza accepts (Distinction of Reason Between Essential Attribute and Substance). In this paper, I contend that Spinoza does not accept this principle because for him a substance and its essential attributes cannot be distinguished by the only two forms that a distinction of reason can take. I contend that this does not mean that Spinoza is departing from Descartes in his understanding of how things or a substance and its attributes are distinguished. I show that for Spinoza and Descartes the modal, real, and rational distinctions only apply to the realm of things, and not to the realm of essences. Finally, I turn to Spinoza's definition of attributes in the Ethics. I defend that the presence of the intellect in this definition indicates that for Spinoza the distinction between a thing and its essence must be performed by the intellect. But I also show that this does not entail that this must be a distinction of reason.
2. Can Essential Attributes be Distinguished by a Distinction of Reason?
Although it is commonly believed that Spinoza accepts the Principle of the Distinction of Reason, there is disagreement over how he understands the distinction involved in this principle. Like Descartes’, Spinoza's understanding is closely related to Francisco Suárez's views.Footnote 3 For Suárez, a distinction of reason is just two ways of conceiving one and the same thing. But note that for Suarez, and hence for his followers, every distinction of reason must be either
(i) a distinction of reasoning reason; or
(ii) a distinction of reasoned reason.
I shall next examine Suárez's account of these two distinctions, laying out the necessary conditions for each. I show that Spinoza does not rely on either of these to distinguish between essential attributes.
2.1. Reasoning Reason
Suárez believes that the distinction of reasoning reason has two necessary conditions. First, it must be constituted by the activity of the intellect and must have no foundation in reality; that is, this distinction does not depend upon the intellect reflecting upon some actual or possible distinction in nature, but is performed solely by the intellect (DM, 7, I.4). This condition is posited in the following paragraph: ‘[a mental distinction] which has no foundation in reality, is called distinction of reasoning reason (distinctio rationis ratiocinantis), because it arises exclusively from the reflection and activity of the intellect’ (DM, 7, I.4). Put more formally, this condition says that:
(i) Constituted by the intellect: if x is distinguished from y by reasoning reason, necessarily the distinction between x and y is constituted by the intellect and has no foundation in reality.
To illustrate, consider Suárez's example of Peter being distinguished from himself. The intellect can conceive Peter twice through the repetition of the concept ‘Peter’. Moreover, the intellect can relate both concepts. But this activity has no grounds in reality, for the distinction performed by the intellect neither exists nor pre-exists in nature. Thus, this activity does not have ontological import.
Suárez believes this kind of distinction of reason has a second necessary condition: it must be ‘ordered to one and the same adequate or simple concept of an object solely through mental repetition or comparison of the object. Hence, Peter is distinguished from himself either according to subject and predicate, when he is predicated from himself, or according to the term and subject of a relation, when he is said to be the same as himself’ (DM, 7, I.5). Suárez believes that a distinction of reasoning reason must involve a comparison between adequate concepts of one and the same thing. For Suárez an adequate concept is a complete representation of a thing.Footnote 4 Thus, a complete concept of Peter must wholly represent Peter. Additionally, Suárez holds that the adequate concepts involved in a distinction of reasoning reason must be able to be related in two ways: as subject and predicate or as term and subject of a relation. In his example, both instances of ‘Peter’ are concepts that wholly represent Peter. Moreover, one can relate these concepts as subject and predicate in the expression ‘Peter is Peter’ or as subject and term of the relation expressed by ‘Peter is identical to Peter’. Let us put this condition in a more formal way:
(ii) Adequate: if x and y are distinguished by reasoning reason, necessarily x and y are adequate concepts of one and the same thing that must be related either as subject and predicate or as term and subject.
Martin Lin has recently presented a rich and novel interpretation in support of the claim that when distinguishing between substance and attribute, Spinoza has in mind Suárez's distinction of reasoning reason. According to Lin, for Spinoza the concept of each attribute is a guise; a way under which God's essence is presented to an intellect. Crucially, Lin holds that this account is both consistent with Suárez's necessary conditions for a distinction of reasoning reason and applicable to Spinoza's essential attributes. But is this so? Since guises do not entail a metaphysical difference because these are not individuated by description but only syntactically, Lin believes that guises fulfil this condition (i) (Lin, Reference Lin2019, p. 76). According to Lin, guises have the same semantic properties because (a) the singular terms in both sentences have the same referent and (b) the general terms in both sentences have the same truth conditions. Hence, guises are only distinct in respect to their syntactic properties, and their distinction has no ontological import (Lin, Reference Lin2019, p. 101). Thus, Lin holds that to understand attributes as guises implies that ‘there is a world as-represented-by-the-language-of-extension and there is a world as-represented-by-the-language-of-thought. These are not really two metaphysically distinct worlds but merely artifacts of two ways of thinking about a single world’ (Lin, Reference Lin2019, p. 78). Lin also believes that that guises satisfy condition (ii), since each guise is an adequate concept that refers without the mediation of a description to one and the same thing. Under Lin's account, for Spinoza the whole essence of God is presented to the intellect under the guise or concept of Thought and the whole essence of God is presented to the intellect is presented to the intellect under the guise of Extension.
It can be contended against the claim that for Spinoza attributes are distinguished by reasoning reason arises when we consider Suárez's claim that the intellect constitutes this distinction by comparing one and the same concept purely through an act of mental repetition. One way of understanding this claim is as the repetition of one and the same term, expressing the same concept and referring to the same thing, both as the subject and predicate of the same expression – in Suárez's example, ‘Peter is Peter’ – or as subject and term – the relation expressed by ‘Peter is identical to Peter’.Footnote 5 But if this is so, Spinoza cannot be using reasoning reason to distinguish essential attributes because the relation of ‘Thought’ and ‘Extension’ is that of different terms expressing different concepts, and not that of one and the same repeated term. But to this objection it can be responded that what is relevant for Suárez is not the terms one uses but the semantic value of these terms. For example, in the expression ‘Peter is identical to Pierre’ we have two singular terms that refer – without the mediation of a description – to the same thing. Since ‘Peter’ and ‘Pierre’ have the same semantic value, both terms can be considered the same concept.Footnote 6
A stronger objection arises when we consider the second necessary condition that Suárez ascribes to a distinction of reasoning reason. According to this condition, for Spinoza each concept of an attribute adequately represents God's essence. But if this is so, the difference between God and a substance with only one attribute would be that the former has only one non-descriptive mode of presentation, whereas the latter has all. The problem is that Spinoza believes that the more attributes that belong to a thing, the greater the reality of that thing (E1p9d). Nor does Lin's account seem to satisfy Suárez's necessary condition (i) Constituted by the intellect of reasoning reason. For Spinoza. ‘the more reality belongs to the nature of a thing, the more powers it has, of itself, to exist’ (E1P11s). Thus, Spinoza believes that the reality of x is thus proportional to the existence of x. Since ‘reality’ has ontological import, then attributes must also have that import. But this is not possible if the distinction between attributes is solely constituted by the intellect. Consequently, if there is a distinction of reason between attributes this cannot be a distinction of reasoning reason, as Lin believes.
2.2. Reasoned Reason
According to Suárez, a distinction of reasoned reason ‘can be understood as pre-existing in reality, prior to the discriminating operation of the mind, so as to be imposing itself, as it were, on the intellect, and to require the intellect to recognize it, but not to constitute it’ (DM, 7, I.4). Thus, whereas a distinction of reasoning reason is merely constituted by the intellect, a distinction of reasoned reason can be recognized by the intellect as pre-existing in a thing. For Suárez, a distinction of reasoned reason:
arises not entirely from the sheer operation of the intellect, but from the occasion offered by the thing itself on which the mind is reflecting. Hence the foundation that is held to exist in nature for this distinction is not a true and actual distinction between the things regarded as distinct; for then not the foundation of the distinction but the distinction itself would precede the mental operation (DM, 7, I.4).
The first part of this passage deals with the nature of the recognition involved in this type of distinction. Suárez posits that to perform a distinction of reasoned reason requires an occasion offered by a thing existing in nature. Thus, for Suárez a necessary condition of a distinction of reasoned reason is that is must be somehow discovered by the intellect, rather than being fabricated by it. Put more formally,
(i) Recognised: if x is distinguished from y by reasoned reason, necessarily the distinction between x and y is recognised by the intellect.
In the passage quoted above Suárez posits that the thing upon which the intellect reflects provides the foundation for the distinction. In the second part of this passage we learn that this distinction is not actual in that thing – it does not exist in the thing, but only pre-exists in it. Thus, for Suárez:
(ii) Pre-exists: if x is distinguished from y by reasoned reason, what distinguishes x and y is not actual but pre-exists in nature.
Suárez posits a further condition that is seldom acknowledged. For him, a distinction of reasoned reason ‘has a foundation in fact, even though formally it will be said to spring from inadequate concepts of the same thing’ (DM, 7, I.5). This distinction ‘arises from inadequate concepts of one and the same thing. Although the same object is apprehended in each concept, the whole reality contained in the object is not adequately represented, nor is its entire essence and objective notion exhausted, by either of them’ (DM 7, I.5). Suárez is saying here that a distinction of reasoned reason must involve inadequate concepts; that is, concepts that do not capture the whole reality or essence of the object that they represent. Crucially, Suárez believes that the inadequacy of the concepts involved in a distinction of reasoned reason is determined by the origin of those concepts. To illustrate how this works, he posits the distinction holding between God's justice and God's mercy:
in God we distinguish His justice from His mercy, because we do not conceive the sublimely simple virtue of God as it is in itself and according to the full range of its energy. We partition it into concepts in line with the various effects of which that eminent virtue is the principle, or by analogy [proportionem] with the various virtues which we find distinct in man, but which in an ineffably eminent manner are found united in the absolutely simple virtue of God (DM, 7, I.5).
Suárez believes that our inadequate representations of God involved in this sort of distinction can arise in two distinct ways: (a) when the intellect represents God trough numerically distinct concepts of the effects of God and (b) when the intellect represents God's essence through concepts of other things that arise through an analogy [proportionem] with virtues that are distinct in human beings. Thus, both (a) and in (b) involve an extrinsic denomination: the denomination of one thing through something that does not belong to that thing but that belongs to a further thing.Footnote 7 Put formally,
-
(iii) Inadequate: if x and y are distinguished by reasoned reason, then the concept of x and the concept of y are inadequate concepts of one and the same thing that involve extrinsic denominations.
Yitzhak Melamed has recently argued that for Spinoza the distinction between substance and attribute must be one of reasoned reason. Melamed believes that what is distinguished by reasoned reason are different aspects of one and the same thing. Unfortunately, Melamed does not offer a definition of aspect nor does he provide an extended characterization of it. However, he does indicate that his understanding of aspect includes the following necessary conditions:
(a) if x and y are aspects of z, x and y must be recognized by the intellect
(b) if x and y are aspects of z, the distinction between x and y must have a foundation pre-existing in z.
(c) if x and y are aspects of z, then z can be adequately conceived under each x and y.
Note that (a) and (b) mirror conditions (i) and (ii) of Suárez distinction of reasoned reason.
There are two principal problems with Melamed's interpretation. First, (ii) and (b) cannot apply to the attributes of God as Spinoza understands them, for then essential attributes such as Thought or Extension would be in God's essence solely in principle. To put it differently, under this scenario, no attribute would be actual, but Spinoza holds that in God all is actual. It can be responded to this objection that for Spinoza each attribute is identical to the essence of God, and that it is the distinction between all the attributes that is in principle in that essence. However, this line of response would be unsatisfactory, for under this scenario for the essence of God to be actual at least one attribute must be actual. But then it would not be possible to distinguish that attribute from all the other attributes that pre-exist in that essence. Neither does it help to argue that the sum of all attributes is the essence of God and that it is the distinction between the attributes in this sum that is in principle in that essence, for then all attributes would be actual, and we would not have a distinction of reasoned reason after all.
Second, Melamed's condition (c), on the other hand, seems to be in direct conflict with Suárez's condition (iii). The problem for Melamed is that if he posits that Spinoza accepts (c) and rejects (iii) he would need to accept that Spinoza is not using Suárez's distinction of reasoned reason. But if he posits that Spinoza accepts (iii) he would need to concede that attributes are inadequate representations of God's essence. We have seen that for Suárez both the inadequate concepts we form through the effects of God or via analogy necessarily involve extrinsic denominations. But neither Thought nor Extension can be predicated of God in this way. When we denominate a thing through an extrinsic denomination, we posit nothing in the thing that we are denominating (Suarez, Reference Suarez and Doyle2017, p. II.9). Since extrinsic denominations do not pick out anything in the essence of the thing denominated, when we use these denominations, we only improperly denominate something. Consequently, Spinoza cannot accept that the distinction of reason in (Distinction of Reason Between Essential Attribute and Substance) is a distinction of reasoned reason.
3. Spinoza's Use of the Distinction of Reason
In the previous section we saw that Spinoza cannot accept that the distinction of reason in (Distinction of Reason Between Essential Attribute and Substance) is one of reasoning reason or a distinction of reasoned reason. Since these are the only two forms that a distinction of reason can take, it follows that God and his attribute cannot be rationally distinct and, consequently, that essential attributes and substance cannot be distinguished by reason. But, if this is so, how can we explain Spinoza's claims in the CM that ‘that distinction is said to be of reason which exists between substance and its attribute’? (CM, II.5). In this section, I answer to this question.
3.1. ‘Attributes’, not Attributes
That Spinoza is not applying Suárez's distinction of reason to attributes such as Thought or Extension becomes clear when we consider that in the CM Spinoza adopts a Cartesian framework in which to each substance only corresponds one principal or essential attribute. Thus, Spinoza posits that: ‘We have divided substances into two chief kinds, extension and thought; thought we have divided into created, or the human Mind, and uncreated, or God’ (CM, II.1). In this passage Spinoza is saying that substances can be divided into two types based on their essential attributes: Extension and Thought. In turn, substances belonging to the latter type – but not substances belonging to the former – can be further divided based on a genetic criterion: some are created, some are uncreated. Spinoza goes on to posit that an instance of a created thinking substance is a human mind. Since there is more than one human mind, created thinking substances are a plurality. However, there is only one uncreated substance: God. Hence, in the CM God is portrayed as the one and only uncreated substance which must also be a thinking substance.
Now, Spinoza does hold in the CM that from the fact that a substance and an attribute are distinguished by a distinction of reason it follows, by the principle of transitivity of identity, that two attributes of that substance must also be distinguished by reason. So, Spinoza accepts in the CM that a substance has only one principal attribute, but he also says that there is a distinction of reason between two attributes of the same substance. This apparent contradiction is dissipated when we acknowledge that Spinoza holds in the CM that to whatever substance, thinking or extended, created or uncreated, a plurality of non-principal attributes belongs. These are the propria or non-essential but necessary properties of that substance.Footnote 8 Propria ‘belong to a thing, but never explain what it is’ (KV, I.7.6). Thus, Spinoza believes that the uncreated thinking substance – that is, God – must be infinite and eternal, whereas a created substance must have a duration. However, these necessary properties do not constitute the essence of those substances and thus we cannot understand what those substances are by understanding these properties.Footnote 9 If Spinoza holds in the CM that two attributes of God are distinguished by a distinction of reason, he must be referring to a distinction holding either between Thought and a proprium of God – like infinity – or to a distinction between two propria of God – for example, eternity and infinity. Since Spinoza believes that the distinction of reason between the attributes of a substance follows from the distinction of reason between each attribute and its substance, this would mean that Spinoza believes that there is a distinction of reason between a substance and each of its propria. Consequently, from Spinoza's use of the distinction of reason in the CM it does not follow that he accepts (Distinction of Reason Between Essential Attribute and Substance). Rather, it follows that he accepts the following principle:
(Distinction of Reason Between proprium and Substance): the distinction between any single proprium of a substance and that substance must be a distinction of reasoned reason,
Further evidence supporting the claim that Spinoza accepts this principle emerges when we consider the example that according to Spinoza: ‘that distinction [a distinction of reason] is said to be of reason which exists between substance and its attribute, as when duration is distinguished from extension’ (CM, II.5). To see how this example supports that Spinoza accepts (Distinction of Reason Between proprium and Substance), recall first that for Spinoza in the CM Extension is not an essential attribute of God; the only principal attribute that Spinoza ascribes there to the uncreated substance is Thought. Hence, in this passage Spinoza must be dealing with a created or finite substance. This is consistent with the fact that it is not eternity but duration or determinate existence that is predicated of it. Now, note that it cannot be the case that duration is here taken as an essential attribute, for we have seen that for Spinoza in the CM there are only two kinds of such attributes: Thought and Extension. Since Spinoza claims that this passage deals with the distinction between a substance and its attribute, by ‘substance’ he must be referring here to the principal attribute of a created substance – in this case, Extension.
Given the Cartesian framework of the CM, Spinoza does not accept there (Distinction of Reason Between Essential Attribute and Substance). Rather, Spinoza accepts in the CM (Distinction of Reason Between proprium and Substance). This step represents some progress, for (Distinction of Reason Between proprium and Substance) does not conflict with Spinoza's understanding of God's essential attributes, as (Distinction of Reason Between Essential Attribute and Substance) does. However, we still need to disambiguate Spinoza's use of ‘distinction of reason’ in this principle: does (Distinction of Reason Between proprium and Substance) involve distinction of reasoning reason or does it rather involve a distinction of reasoned reason?
3.2. Which Distinction of Reason?
Let us begin with the distinction of reasoning reason. Recall that for Suárez, this sort of distinction has two necessary conditions: (i) Constituted by the intellect and (ii) Adequate. Can (Distinction of Reason Between proprium and Substance) satisfy all these conditions? Condition (ii) of reasoning reason states that this distinction must involve the comparison of concepts that adequately represent one and the same thing. But the propria of God are not his essence, and hence the concept of a proprium does not represent God's essence. It follows that every concept of a proprium, or even a concept involving the concepts of all propria, will be, at best, an inadequate representation of God. But nor is Suárez's condition (i) satisfied. This condition says that a distinction of reasoning reason must be constituted by the intellect.Footnote 10 But Spinoza believes that infinite things in infinite modes follow from the necessity of God's essence (E1p16), God acts from an essential necessity (E1p17), and God is the cause of all things (E1p18). Consequently, for him the propria of substance, such as being the cause of everything or necessity, have real metaphysical weight. Thus, the distinction of reason involved in Distinction of Reason Between proprium and Substance cannot be one of reasoning reason.
Can this be a distinction of reasoned reason? Evidence that Spinoza believes that this is so must come from the three necessary conditions of this distinction being fulfilled. Are the concepts of propria (i) recognized as (ii) pre-existing in God? This seems to be the case, for Spinoza believe that propria are necessary properties of God that are not in God's essence but necessarily follow from that essence. In this sense, the intellect can reflect upon God's attributes and recognize that from these properties such as infinity or eternity follow.
What about condition (iii)? If the distinction of reason in (Distinction of Reason Between proprium and Substance) is a distinction of reasoned reason, then it necessarily must involve inadequate concepts that arise through extrinsic denominations. Is this consistent with Spinoza's use of the distinction of reason? Evidence that this is so can be found both in the TEI and in the KVFootnote 11. In a footnote in the TEI Spinoza says that when the essence of a thing is inadequately inferred from another thing: ‘we understand nothing about the cause except what we consider in the effect…something clearly conceived is attributed to the cause on account of the effect, as we shall show in an example; but nothing is attributed to it except propria, not the essence of a particular thing’ (TEI, 13). Spinoza is saying here that the concept of propria are inadequate concepts that arise when we consider a thing in terms of its effects, which as we have seen necessarily involve considering a thing through an extrinsic denomination. This becomes clear when we consider the following passage of the KV:
Everything which men ascribe to God besides these two attributes [Thought and Extension], must, if it does otherwise belong to him, either be an extrinsic denomination, such as existing through himself, being eternal, one, immutable, etc., or be in respect to his actions, such as that he is a cause, a predeterminer, and ruler of all things. These are all propria of God, but they do not give us any knowledge of what he is.Footnote 12
Spinoza is saying here that there are two ways in which propria can be predicated of God: either a proprium is an extrinsic denomination of God or else that proprium is predicated of God in respect to God's actions or effects. Thus, Spinoza seems to be dividing the concepts of propria in the very same categories under which the inadequate concepts involved in Suárez's distinction of reasoned reason can arise. Consequently, the distinction involved in (Distinction of Reason Between proprium and Substance) must be a distinction of reasoned reason.
4. Descartes’ Shadow?
In the introduction of this paper, I mentioned that (Distinction of Reason Between Essential Attribute and Substance) is supported by three claims:
(1) The Metaphysical Thoughts. Spinoza seems to use the distinction of reason in this treatise to distinguish between substance and essential attribute.
(2) Descartes’ Shadow. If Spinoza does not distinguish substance and attribute through a distinction of reason, he would be going against Descartes without alerting his reader.
(3) The Intellect in the Definition of Attribute. The fact that Spinoza includes the intellect in the definition of attribute indicates that for him a distinction between substance and attribute is a distinction of reason.
So far, I have focused on (1). In this section, I turn to (2).
Descartes's Shadow involves two distinct but closely related worries. First, it is widely accepted that for Descartes substance and attribute are distinguished by reason. Let us call this:
(i) Cartesian Distinction Between Substance and Essential Attribute: if x is a substance and y is an essential attribute of x, necessarily the distinction between x and y is a distinction of reason
Second, it can also be contended that, in rejecting Cartesian Distinction Between Substance and Essential Attribute to distinguish between substance and attribute Spinoza would be introducing a distinction that is not part of Descartes' catalogue of ways in which things can be distinguished. According to this catalogue, there are three ways in which things can be distinguished: by reason, modally, or really. This catalogue seems to be endorsed by Spinoza in the CM. But to distinguish substance and attribute modal distinction is off the table, for it involves a mode, and real distinction must also be discarded, for it holds only between substances.
(ii) Cartesian Catalogue of Distinctions: x and y are distinct if and only if x and y are either rationally, really, or modally distinguished.
In what follows I argue that (i) and (ii) are false and that, consequently, we should not worry about Descartes's Shadow.
4.1. Descartes’ Distinction of Reason
Let us begin with (i) Cartesian Distinction Between Substance and Essential Attribute. Although this principle is widely accepted, two facts indicate this is not true. First, the paradigmatic passage that is used to support (i) comes from the Principles of Philosophy (PP), where according to Cottingham's translation from the Latin Descartes says:
Indeed, it is much easier for us to have an understanding of extended substance or thinking substance than it is for us to understand substance on its own, leaving out the fact that it thinks or is extended. For we have some difficulty in abstracting the notion of substance from the notions of thought and extension, since the distinction between these notions and the notion of substance itself is merely a conceptual distinction [Nonnulla enim est difficultas, in abstrahenda notione substantuae, a notionibus cogitations vel extensions, quae scilicet ab ipsa ratione tantùm diversae sunt].Footnote 13
Note that in contrast with Cottingham's translation, in this passage Descartes does not explicitly use the technical term ‘distinction of reason’ to qualify the distinction between the notion of substance and the notion of attribute. All he says is that they are distinguished in terms of reason. Now, that Descartes does not have in mind the technical term ‘distinction of reason’ in this passage is supported by the French translation of the PP, both revised and accepted by Descartes:
For there is some difficulty in separating the notion of substance from the notions of thought or extension, which of course differ from substance, because we sometimes consider thought or extension without reflecting on the very thing which thinks or is extended.Footnote 14
In this passage, Descartes does not make any reference to the term ‘reason’. He explains that the distinction between substance and attribute arises because we can consider the essence of a substance without reflecting upon its existence; that is, when we distinguish an essential attribute from that to which it is attributed. For example, the attribute of Thought and a mental substance – a mind – are one and the same thing, but we can separate the notion of Thought from the notion of a mind.
A further thing to note is that Descartes' terminology is consistent with what seems to be a clear rejection of Cartesian Distinction Between Substance and Essential Attribute. That in these passages Descartes cannot have in mind a distinction of reason becomes clear when, in a letter to an anonymous correspondent, he clarifies his use of this distinction of reason in the article of the Principles of Philosophy where he explains the distinction between substance and attribute:
I do not remember where I spoke of the distinction between essence and existence. However, I make the distinction between modes, strictly so called, and attributes, whiteout which the things whose attributes they are cannot be; or between the modes of things themselves and the modes of thinking…Thus shape and motion are modes, in the strict sense, of corporeal substance; because the same body can exist at one time with one shape and at one time with another, now in motion and now at rest; whereas, conversely, neither this shape nor this motion can exist without this body…But existence, duration, size, number and all universals are not, it seems to be (me?), modes in the strict sense… they are referred to a broader term and called attributes, or modes of thinking, because we do indeed understand the essence of a thing in one way when we consider it in abstraction to whether it exists or not, and when we consider it as existing; but the thing itself cannot be outside our thought without its existence, or without its duration or size, and so on. Accordingly, I say that shape and other similar modes are strictly speaking modally distinct from the substance whose modes they are; but there is a lesser distinction between the other attributes. This latter can be called modal -as I did at the end of my Replies to the first Objections -but only in a broad sense of the term, and it is perhaps better called formal. But to avoid confusion, in article 60 of Part One of my Principles of Philosophy, I call it a conceptual distinction -that is, a distinction made by reason ratiocinatae (CSM, III, p. 280).
Descartes makes clear that he is not referring to the distinction between the essence and the existence of a thing. That is, this is not the distinction between an attribute – an essence – and a substance. Moreover, when Descartes next explains what he has in mind by ‘formal distinction’ he bundles together the distinction between a mode and its substance – for example, shape and body – and the distinction between a proprium and its substance – for example, duration and body. Descartes holds that it is only the latter which should be called a distinction of reason. Finally, Descartes says the term ‘distinction of reason’ should always be understood as a distinction of reasoned reason, hence he does not accept any distinction of reasoning reason. But then, either Descartes (a) accepts that the distinction between substance and attribute is recognised by the intellect, is not actual but pre-exists in nature, and involves an inadequate concept, or else (b) he is understanding this distinction in a way that is alien to Suárez and to Descartes's contemporaries.
Textual support for (a) is found in Descartes’ account of formal distinction in the First Set of Replies:
As to the ‘formal’ distinction….this kind of distinction does not differ from a modal distinction; moreover, it applies only to incomplete entities, which I have carefully distinguished from complete entities. It is sufficient for this kind of distinction that one thing be conceived distinctly and separately from another by an abstraction of the intellect which conceives the thing inadequately… (CSM, II, p. 85)
Descartes recognizes that any ‘formal distinction’ – that is, any modal distinction and any distinction of reasoned reason – involves an inadequate concept of the thing thus distinguished. But Descartes also believes that a substance is adequately conceived through its principal or essential attribute. Thus, it is hard to see how Descartes would accept the claim that when I distinguish the attribute of Thought from my mind conceived as a substance, I am inadequately conceiving that substance. But then the distinction between an attribute and a substance cannot be a distinction of reasoned reason. And since Descartes does not accept distinctions of reasoning reason, he cannot accept (i) Cartesian Distinction Between Substance and Essential Attribute.
4.2. Descartes’ Catalogue of Distinctions
In the Principles of Philosophy Descartes presents a catalogue of distinctions that seems to be endorsed by Spinoza in the CM. This can lead to accepting the following claim:
(ii) Cartesian Catalogue of Distinctions: x and y are distinct if and only if x and y are either rationally, really, or modally distinguished.
Descartes believes that if x and y are modally distinct, then either x or y is a mode. For example, my idea of Greta is modally distinct from my mind, a thinking substance. Thus, since a modal distinction involves necessarily a modification of a substance, Descartes cannot accept that substance and attribute are modally distinct. But neither can Descartes accept that the distinction of attribute and substance is a real distinction, for he believes that if x and y are really distinct then each x and y are numerically distinct substances. For example, if my mind and Thought are really distinct, then my mind is one substance and Thought is a further – numerically distinct – substance. But Descartes cannot accept this, for he believes that Thought constitutes the essence of my mind, and a thing is numerically identical to its essence. Since we have seen that Descartes also rejects that substance and attribute are distinguished by reason, it follows that he must reject (ii) Cartesian Catalogue of Distinctions. Evidence that Spinoza accepts that these catalogue does not apply to essences appears when we consider that for him a real distinction involves two substances, a modal distinction involves either two modes or a mode and a substance, and a distinction of reason a substance and a proprium or two proprium. Thus, Spinoza seems to believe that this catalogue involves only distinctions between substances, their modes, or their non-essential properties.
I have argued for the negative claim that Descartes and Spinoza do not believe that Descartes’ Catalogue of Distinction applies to the distinction between substance and attribute. Thus, for them the distinction between substance and attribute is not a distinction of reason. But Descartes and Spinoza do believe that substance and attribute can be distinguished: what is for them the distinction between substance and attribute? Although Descartes does not say explicitly, he does posit that the essence (attribute) and existence (substance) of one and the same thing are ‘in no way distinct’ outside of thought. To illustrate this distinction, he uses as an example ‘Peter is a man’, the relation between Peter and man. Thus, what we have here is a type of distinction that Descartes acknowledges but does not name, that is performed by the intellect, and that does not belong to his catalogue: the distinction between one and the same thing understood as substance and attribute, existence and essence, or subject and essential predicate.
Spinoza seems to be following Descartes when he explains in the KV that essence of God and God's existence are the same thing (KV, I.1.2). Spinoza discusses the relation between God and all attributes in the second chapter of the first part of that book. First, Spinoza holds that to demonstrate that God is is to demonstrate that God exists. Now, according to Spinoza whatever exists must either exist as a substance or as a mode.Footnote 15 Spinoza believes that if x exists as a substance, existence must belong to the essence of x. And if this is so, x must necessarily exist (E1p7). On the other hand, if x exists as a mode, existence does not belong. Thus, to demonstrate that God is to demonstrate that God exists as a substance. Next, Spinoza holds: ‘Now that we have demonstrated that God is, it is time to show what he is. He is, we say, a being of which all, or infinite, attributes are predicated, each of which is infinitely perfect in its own kind’ (KV, I.1.2). So, the distinction between God's essence and God's existence is the distinction between what God is – God understood as Thought, Extension, and all other attributes – on the one hand, and the understanding that God exists as a substance, on the other (ibid.).
The distinction between substance and attribute is a distinction between the existence and the essence of a thing. But what is, for Spinoza, the distinction holding between two attributes of the same substance? Note that Descartes does not have to deal with this question because for him each attribute belongs a numerically distinct substance. Thus, in an indirect way, tokens of attributes are really distinct. But Spinoza does not believe that each attribute corresponds to numerically distinct substance. Hence, since the Cartesian Catalogue only applies to the realm of existing things, this catalogue cannot be used, even if it is only indirectly. But that does not mean that Spinoza cannot make sense of this distinction or that for him attributes are or indiscernible or numerically identical. An example of the distinction between essential attributes available to Spinoza comes from the Aristotelian tradition represented by Suárez. According to this tradition, one human being, say Peter, exists as one and the same thing can have several essential constituents or predicates – say, rational and mortal – which can be recognized as numerically distinct by an intellect. Crucially, this distinction must be performed by the intellect but is grounded in reality. Similarly, under the Cartesian framework if a substance had more than one attribute, that intellect can recognize these attributes as the numerically distinct essential constituents without resorting to the catalogue of ways in which existing things can be distinguished.
I have argued that this distinction between attributes does not hold at the level of existence, but only at the level of essence. Thus, although performed by the intellect, this is not a distinction of reason. I acknowledge that it can be objected that if Descartes or Spinoza believe that there is a distinction in the realm of essences that is not captured by the Cartesian Catalogue, they would have named this distinction. Here, I speculate that they might have not felt the need to do so for two reasons. First, as we have seen, this is not a distinction amongst things, and this is the type of distinction that Descartes and Spinoza seem to be interested in singling out. Second, for those familiar with the Aristotelian tradition, the distinction between existence and essence, or between two or more essential attributes of one and the same thing, can be straightforwardly understood and does not seem to require a catalogue.
5. The Definition of Attribute
We have seen that claims that (1) The Metaphysical Thoughts and (2) Descartes’ Shadow are mistaken. But we still need to deal with
(3) The Intellect in the Definition of Attribute. The fact that Spinoza includes the intellect in the definition of attribute indicates that for him a distinction between substance and attribute is a distinction of reason.Footnote 16
In this section, I show that (3) is also false.
Spinoza defines attribute as follows: ‘By attribute I understand what the intellect perceives of a substance, as constituting its essence’ (E1pd4). As we have seen, the intellect is involved in both the distinction of reasoning reason and the distinction of reasoned reason. On the other hand, the intellect does not seem to be involved in the real or modal distinctions, the other two distinctions in the Cartesian Catalogue of Distinctions. Thus, if we accept that the Cartesian Catalogue of Distinctions is exhaustive, the following principle follows:
Intellect: if the distinction between x and y is performed by the intellect, then x and y are distinguished by reason
In the previous section, I showed that the Cartesian Catalogue of Distinctions is not exhaustive. Both Descartes and Spinoza recognize a further distinction: the distinction between attribute and substance is that between one and the same thing conceived as existence and essence or as subject and essential predicate. Now, crucially for the purposes of this section both Descartes and Spinoza hold that this distinction must be performed by the intellect. Moreover, amongst the both in Arnauld and Nicole's Logic or the Art of Thinking and in Johannes Clauberg's Old and New Logic – the most influential logic textbooks of Spinoza's time and both written by Cartesian philosophers – the distinction between existence and essence and between subject and essential predicate are distinctions performed by the intellect.Footnote 17 But we have seen that the distinction between existence and essence cannot be a distinction of reason. Consequently, the claim that if the distinction between x and y is performed by the intellect, then x and y are distinguished by reason, is false.
The interpretation that I have put forward here accommodates the presence of the intellect in Spinoza's definition of attribute. Moreover, it also does so in a more satisfactory way than interpretations that contend that the presence of the intellect in that definition indicates that for Spinoza substance and attribute are distinguished by a distinction of reason. Recall that the essence of a thing is what that thing is and this is understood by the intellect and that, by abstraction, the same thing can be perceived as some thing that exists. Thus, although we have two numerically distinct thoughts of the same thing, and although these have a foundation in reality, they do not track a distinction pre-existing in nature. On the other hand, as we have seen, a distinction of reasoned reason, the only sort of distinction of reason recognized by Descartes, necessarily involves inadequate conceptions that depend upon a distinction pre-existing in the thing thus distinguished.
Let me conclude by noting that although the distinction between substance and attribute is performed by the intellect, from this it does not follow that each attribute is numerically identical to the essence of its related substance. Recall that for Descartes each attribute is numerically identical to the substance whose essence it constitutes. And since a thing is numerically identical to its essence, then each attribute is numerically identical to its related substance.Footnote 18 But Spinoza believes that there is only one substance and that this consists of all attributes. What should we do in this case? Although Spinoza does not answer to this question directly, an explicit answer can be found in Arnauld and Nicole's Logic or the Art of Thinking. For these logicians, when a thing has an essence with more than one attribute, it is not the case that each attribute is numerically identical to that essence. Rather, only the sum of all essential attributes is.Footnote 19 This is so because an essence and the sum of its attributes must be convertible. In the case of an essence with only one attribute, it is clearly the case that this attribute is convertible with that essence. But if an essence has more than one attribute, only the totality of those attributes is convertible with that essence. By way of example, Arnauld and Nicole refer to a case in which the essence of one individual – call him Socrates – is ‘human’ and ‘human’ involves the attributes ‘rational’ and ‘mortal’. In this case, neither ‘rational’ nor ‘mortal’ alone would be convertible with ‘human’ and hence cannot be numerically identical to this essence. Crucially, this scheme is consistent with Spinoza's claim that the essence of God is to consist of all attributes: only the totality of the attributes is convertible with that essence. Thus, when the intellect distinguishes between existence and essence, it is distinguishing between a substance and the totality of its attributes.
6. Conclusion
Spinoza cannot accept (Distinction of Reason Between Essential Attribute and Substance) if he introduces the distinction of reasoned reason in the CM only to distinguish between substance and proprium. Crucially, Both Spinoza and Descartes accept that the Cartesian catalogue of distinctions only applies to the realm of things – and not to the realm of essences – and that the distinction between a substance and its essence must be performed by the intellect. The presence of the intellect in this definition indicates that Spinoza acknowledges that the distinction between substance and attribute is that between one and the same thing understood as existence and essence or as a subject and its essential predicate. But it does not indicate that for him each attribute must be numerically identical to its related substance.Footnote 20
