martes, 21 de junio de 2011

Deleuze's Spinoza

                       - SPINOZA'S DELEUZE -
                  Formal Distinction, Univocity and the Attributes
   In this paper I would like to consider some of the ramifications which follow for Spinoza’s so-called ‘ontological argument’, which advances the thesis of ontological univocity, i.e. that there is one being, and that being is said in one and the same sense. For this I follow some considerations set forth by Gilles Deleuze’s (1992) reading in Expressionism in Philosophy: Spinoza. I begin by outlining how Deleuze historically situates Spinoza’s famous triadic distinction in the Ethics between substance, attributes and modes along the axes of three classical Cartesian distinctions: real, modal, and of reason. I then propose a tentative reconstruction for Spinoza’s argument for the thesis that ontological univocity is to be understood from the impossibility of real distinctions being numerical, and vice versa.
          I consider some difficulties which emerge for this position, given the Deleuzean consideration that attributes stand for formal distinctions, which are real and multiple qualitatively, but quantitatively singular. I show the motivations behind taking the alternative approach of making attributes mere distinctions of reason, and finally suggest why Deleuze’s reading encourages us to reject this option. I conclude that understanding attributions of essence as pertaining to formal distinctions rather than of reason is crucial in order to understand Spinoza’s perplexing idea about there being ‘degrees of reality’ in substance. Attributes can therefore me understood as adding to the reality of the same without dividing it.

I – Descartes Three Distinctions and Duns Scotus’ Formal Distinction
          Deleuze’s reading of Spinoza begins by situating the latter within the frame of a set of classical Cartesian aporias. These concern the distinction, already made explicit in Descartes, between substances and modes: the former referring to what “is in itself”, and the latter to what “is in something else”[1]. This seems readily continuous with Spinoza’s own definition at the outset of the Ethics which claims that substance is what exists and is conceived by itself, and modes are what exist and is conceived through something else (E, P1D3, P1D5). As Deleuze notes, Spinoza elsewhere explicitly associates his own distinction between substance-modes to Descartes’ own distinction when reshuffling the latter’s three kinds of distinctions: “We need to recall what Descartes has taught, viz. that there is nothing in nature but substances and their modes. From this a threefold distinction of things is deduced (1.60-62), viz. real, modal, and of reason.” (M, 353)
        Yet Spinoza will also underline later that modes cannot exist / not be conceived without substance, even if the two make up the ‘total of existence’ (E, P1P15). This means that the modes, which must exist in something else, do so in substance (God) and can only be conceived through substance (E, P1P23).  In turn, substance is known through the attributes which express its essence (E, P1DIV). Since existence pertains to the essence of substance, it clearly follows that the modes that exist in substance which they presuppose are also known through the attributes they imply (E; P1P23, P1P14). Or as Deleuze puts it: “If modes always presuppose a substance, and are sufficient to give knowledge of it, they do so through a primary attribute which they imply, and which constitutes the essence of the substance itself” (Deleuze 1992: Pg. 29) Whereas for Descartes the distinction between attributes and essences distinguishes between different substances, Spinoza will advocate the thesis of a single substance corresponding to all attributions of essence. We shall return to this crucial distinction below; but for now let us unpack how Descartes distinguishes through the tripartite distinction between substances, attributes and modes, three different kinds of distinction. We formulate these abstractly, in continuity with Deleuze’s presentation as follows (Ibid):
1)      Real distinction – For any x, x is really distinct from y if and only if x is ontologically different from y, i.e. if x and y are non-identical substances. The corresponding principle here is that of exclusion (between distinct substances).
2)      Modal distinction – For any x, x is a modal distinction of y, if and only if x ontologically implies y, while y does not ontologically imply x. That is, x is a mode of y just in case the existence of x implies the existence of y, but not the converse. The corresponding principle here is that of unilateral determination (between modes and substances).
3)      Distinctions of reason – For any x, x is a distinction of reason of y, if and only if y epistemologically implies x. That is, x is an attribute of y, if and only if y cannot be known except by way of x, i.e. if x is a (primary) attribute of y. The corresponding principle here is that of abstraction (between attributes and substances).
     As we shall see, Spinoza’s early ontological reorganization of the relation between substances, attributes and modes in the Ethics attempts precisely a corrective to how Descartes conceives and articulates these three kinds of distinction. We may just note provisionally that the thesis of ontological univocity must imply that for Spinoza knowledge of the attributes could only entail a singular substance, i.e. that the distinction of reason couldn’t ever suffice to be a real distinction in the Cartesian sense. 
           To show how this anticipated asymmetry in particular leads into a different ontological picture in Spinoza’s account, Deleuze proceeds to articulate briefly some of the central ‘ambiguities’ in Descartes view. First, with respect to the role that attributes come to play, both as qualifying, insofar as they discern the quality of the substance, and as diversifying, insofar as they distinguish varieties of modes in substances. Thus Descartes claims with respect to the attributes of thought and extension that they “…can be considered as constituting the natures of intelligent and bodily substance respectively”, but also that they “…can be taken as modes of a substance, that is insofar as one and the same mind can have many different thoughts; and one and the same body can be extended in many different ways.”[2] If there is more than one mind and one body, that is, if there exist more than one substance with the same attributes, and if modes are identified with attributes, then it plainly follows that different substances may share the same attributes and the same modes.
          In this way, while attributes in one sense specify or qualify the nature of the specific substances they apply to, in another sense they describe diverse modes which may obtain for different substances, i.e. bodies and mind bearing shapes and thoughts[3]. Different substances may thereby bear the same attribute by discerning appropriate modal essences, provided they are taken to be dependent on substance; different bodies may share the same shapes, and different minds the same thoughts. The conclusion Deleuze wishes to underline from this is finally that for Descartes “there exist substances sharing the same attribute. In other words, there are numerical distinctions that are at the same time real or substantial.” (Deleuze 1992: Pg. 30) The notion of numerical distinction is introduced here by Deleuze, but remains underexplained in the text. By way of clarification, let us expand our original ‘Cartesian set’ to include this latter variation:
4) Numerical DistinctionFor any x, x is numerically distinct from y if and only if both x and y are conceptually identical occasions of z; that is, if both x and y are mere instances of the same concept. The corresponding principle here is diversification (between attributes-modes sharing the same concept).
       That numerical distinction can amount to real distinction entails thus that the diversification of modes-attributes as many instances of under the same concept index likewise many different substances to which these instances belong.  Let us for the moment obviate if this reconstruction is a good one, or whether the ‘major difficulties’ Deleuze attributes to Descartes’ system from these considerations actually obtain.
    The second important feature in the Cartesian set of distinctions for our purposes concerns real distinction alone. It must be possible to specify how one knows of real distinction, that is, how an ontological division between things can be epistemologically adequate to our conception of different things. In other words, how conception can amount to more than mere abstract ‘distinctions of reason’ between attributes of substances; but be capable of distinguishing really distinct substances as such. For this, Descartes, as well as Spinoza, will testify to the completeness of the Idea in real distinction: x is known to be ontologically different from y, if and only if, the Idea-concept of x excludes everything in content of the Idea-concept of y, and vice versa. Thus Spinoza claims, in continuity with Descartes, that “By substance, I mean that which is in itself, and is conceived through itself; in other words, that of which a conception can be formed independently of any other conception.” (E, P1D3)
          This obviously already presents a problem for Descartes in conjunction to the earlier possibility that numerical distinctions may be real. This is the case since if more than one substance shares the same concept, but given that conceiving of real distinction requires conceptual exclusion, it becomes impossible for real distinction between such substances to be conceived without confusion. But let us leave this aside for the moment[4]. The second conclusion Deleuze draws is that “Real distinction brings with it a distinction of things… that is, a corresponding numerical distinction.” (Deleuze 1992; Pg. 31)
        Spinoza, says Deleuze, begins his enterprise in the Ethics around these two conclusions. First, he rejects the thesis that more than one substance could share the same attribute. In order to show this, Deleuze argues, Spinoza offers two distinct arguments. For the purposes of this paper it should suffice to briefly focus on the first, which involves a reduction ad absurdum, and finds its locus around the first eight propositions of part I of the Ethics. The reconstruction Deleuze offers is the following:
(1)   Two or more substances cannot share the same attribute for they would have to be distinguished by their modes, which is inconsistent with the thesis that the substance is self-distinct and thus that it precedes modes (E, P1P5).
(2)   Substance cannot have an external cause, since this entails it would have to share the same nature-attribute with its cause. (E, P1P6)
(3)   Thus there can be no numerical distinction in any substance, of whatever attribute; meaning that substance is infinite. (E, P1P7-P1P8)

  Now, Spinoza’s argument for (3) might seem baffling at first; concerning the nature of infinity in substance. Deleuze’s exposition does not clarify how infinity arrives in the third step, yet it remains crucial for the development of the argument below, as we shall see. The key to the argument, however, is given in the proof for Proposition 8, which we can attempt to reconstruct coarsely as providing the following argument:
A)    Existence follows from the essence of substance (P1P7).
B)    There can only be one substance per attribute [NSA principle] (P1P5).
C)    What exists finitely is limited by other finite existence (P1D2, P1P8)[5].
D)    But if another substance exists, then it must also exist by necessity (P1P7)
E)     Therefore, two substances cannot exist as finite, for then they would share the attribute of necessary existence by (A), and this contradicts (B).
F)     Therefore, substance must be infinite.

        In any case, Deleuze’s reconstruction shows finally that numerical distinctions cannot apply to real substances, since substance has unique attributes, and as infinite is necessarily unlimited by another of its kind, i.e. sharing the same essence. This already seems to argue for the uniqueness of substance; but Deleuze doesn’t seem to acknowledge that such an argument is implicit here; and claims instead that for Spinoza the uniqueness of substance is rather secured from Proposition 9 onwards. Numerical distinctions, for their part, are allotted to being proper to modes alone. But of course these have no properly ontological valence outside the modes, i.e. they are not real. So finally to make of real distinction a numerical distinction would be to confuse real distinctions with merely modal distinctions.
       For Deleuze, these considerations open up a crucial question concerning the attributes, and their qualifying role for substance. Since it has been admitted that modal distinctions are not real, then it follows that the diversifying role where attributes are tethered to modes is disavowed. Yet it remains it remains unsettled what the status of attributes becomes in what concerns their qualifying role; as specifying the nature or essence of the one substance. Are such qualifications real, or are they merely distinctions of reason which fall short of being real? On the one hand, Spinoza clearly seems to testify in favor of the reality of attributes, since they are “conceived to be really distinct”. (E, P1P10) But this division among the attributes cannot result in a division between real substances, since this would make real distinction modal or numerical, which we have seen is part of what Spinoza seeks to overcome against Descartes (Deleuze: Pg, 35) The question becomes thus how is it possible to reconcile Spinoza’s claim that a) there is only one substance for all the attributes (ontological univocity), with b) the reality of a diversity of attributes of that same substance which do not divide it.
      On the one hand, the uniqueness of substance follows from its indivisibility or its being-unlimited, since several infinite substances would violate the principle according to which an attribute cannot apply to one more substance (P1P5, P1P12-13). Additionally, since substance is absolutely infinite, all attributes must pertain to it. It follows from the hypothesis of there being two or more substances, that they would share necessary existence and therefore once again violate the NSA-principle (P1P13-14)[6]. It will follow that thus all attributes must be said of one and the same substance, or which amounts to the same, ontologically in one and the same sense. But then in what sense are we to understand the “reality” according to which the more attributes a substance has the more real it is? (P1P12, P1D4) Put differently, how can attributes amount ever to more than mere distinctions of reason, proper to ‘abstraction’?
        At this juncture, Deleuze proposes the argument that in spite of Spinoza’s disavowal of peripatetic distinctions, one must understand the status of attributes as being formal distinctions of substance[7]. For Deleuze this implies to narrow down the difference between a purely qualitative sense in which each attribute coins one essence respectively, and another sense in which all of these pertain to one and the same substance. Thus, even though Spinoza says that attributes are ways in which substance is conceived, this does not imply that there ought to be as many substances as attributes. As Deleuze puts it: “in positing as many substances as there are different attributes we make of real distinction a numerical distinction, confusing real distinction not only with numerical distinction, but with distinctions of reason as well.” (Deleuze 1992; Pg. 37) To unpack the notion of formal distinction let us propose the following formulation:
4)      Formal Distinction – For any x and for any y, x and y are formal distinctions of z, if and only if x and y qualify z completely independently without diversifying it, i.e. x and y are formal distinctions of z just in case x and y are said qualitatively distinctly of z but quantitatively univocally of z. The corresponding principle here is that of pure qualification (between attributes and a corresponding single substance of which these attributes are said).
      We should first note that this distinction is strong enough to preserve the intellect’s seizure of substance; even if in this case we are no longer dealing with mere abstractions through which substance might be known clearly and distinctly or not. The qualification of substance is not representational insofar as the attributes are said to follow from substance as the latter makes its essence transparently known. The subtlety here concerns the preservation of the criterion for real distinction, which involves the completeness of the Idea, wed to the claim that real distinctions may apply to one and the same substance. As a result, attributes would not jeopardize the thesis of ontological univocity advanced by Spinoza’s monistic ontology. Yet the argument apparently ‘bridges the gap’ between attributes and substance in a way that seems to rearticulate the status of distinctions of reason altogether, unyoking the latter from the diversification of modes and thus of numerical distinction. The argument Spinoza would follow under this reading can be roughly reconstructed as follows[8]:
1) There are infinite attributes, which function as ‘points of view’ on a substance (P1D4).
2) Because of the NSA principle, each attribute is a point of view on only one substance.
3) Numerical distinction is never real distinction; conversely, real distinction is never numerical distinction.
4) Thus, the real distinction between the infinite attributes is not a numerical distinction.
5) Thus, the substance which each attribute is a point of view on is necessarily the same substance. The attributes are qualitatively distinct, but quantitatively identical.
6) There is thus only one necessarily existing substance which has all attributes.

        This reading allows Spinoza to avoid two dangers which follow from reducing attributes to being mere ‘points of view’ or distinctions of reason, and which the intellect perceives or ‘abstracts’ from substance to know of it, but fall short of being real determinations of the same. First, this latter option might seem a palatable ‘subjectivist’ possibility, according to which substance has only one essence, but can be conceived in different ways by finite intellection. But as has been pointed out (notoriously by Gueroult (1992)) subjectivism runs into a problem when explaining the variety of attributes as mere abstractions. This can be gauged from the implication that attributive distinctions are on that account taken to be merely abstract conceptualizations on substance, in conjunction with the claim that even God qua infinite intellection thinks attributes. In a nutshell, since Spinoza claims that even the infinite intellect conceives of the plurivocity of attributes, it would follow that not even the infinite intellect understood itself correctly[9]. Thus either subjectivism is false or Spinoza is forced into contradiction[10]
        Against this reading, taking attributes as real but purely formal distinctions allows us to make sense of the claim according to which “The more reality or being a thing has the greater the number of its attributes.”  (P1P9). Taking attributes to as formal distinctions which are real, the points of view on substance constitute not just external viewpoints by finite intellection, but actually are ways in which singular substance comes to conceive of or express itself in a variety of forms. Thus, the manifold attributions are mere occasions of the infinite intellect coming to know itself in its qualitative plurivocity, while retaining its ontological univocity. On any account, one might object that it is still unclear why having more powers or attributes amounts to an increase in the reality of substance. Yet Deleuze’s reading as reconstructed has the additional virtue of not making reference to this notion; even if as we have been arguing it poses a way to render it compatible with the rest of Spinoza’s argument.
     Furthermore, this reading seems to sit well with Spinoza’s claim that God’s intellect must be fully actual and not merely potential, so that the attributes which express its essence would be mere potentialities which may or may not apply to his substance. Since the latter option would divide the intellection of attributes, it would thereby occasion a split in which the essences expressed in the attributes would be said to be merely ‘potencies’, contradicting the continuity between God’s intellect and his essence as fully actual. This point is clearly expressed by Spinoza in the second note to Proposition 23, which states that: “ Further, all the philosophers whom I have read admit that God's intellect is entirely actual, and not at all potential; as they also admit that God's intellect, and God's will, and God's essence are identical, it follows that, if God had had a different actual intellect and a different will, his essence would also have been different; and thus, as I concluded at first, if things had been brought into being by God in a different way from that which has obtained, God's intellect and--will, that is (as is admitted) his essence would perforce have been different, which is absurd.” (P1P23).
           If from this it follows that God cannot will to create those intellects which grasp his essence in any less perfect a manner than his own infinite intellection, since “…we can have no sound reason for persuading ourselves to believe that God did not wish to create all the things which were in his intellect, and to create them in the same perfection as he had understood them.” (Ibid) And since it is in the function of the finite intellect to conceive of the essential and infinite attributes or modalities of God and nothing else, it plainly follows such a comprehension is adequate to God’s essence (P1P11, P1P30).
        Second, this reading allows to dodge the problems that arise from an ‘objectivist’ reading according to which attributes express a variety of essence of the same substance, but which are distinct things in themselves[11]. Although this preserves the reality of distinct attributes, it seems at odds with univocity in the way of postulating these relatively independent essences which give distinct entities. Specifically, this reading seems at odds with Spinoza’s clear commitment to the thesis that it is of the nature of substance to conceive of itself by itself, without any of them having determined the other (P1P10). Therefore, if we made of the attributes multiply a variety of things and of substance a complex set of such things, we would have to admit of an internal division within substance, which runs against the clear statement according to which: “It is thus evident that, though two attributes are, in fact, conceived as distinct--that is, one without the help of the other-yet we cannot, therefore, conclude that they constitute two entities, or two different substances.” (Ibid) The real distinctions pertaining to the attribute must thus at once preserve the requirements of conceptual independence (real distinction) and ontological univocity (the distinction of reason). This strange hybrid structure assigned to substance is precisely that given by formal distinction, which isolates attributions of substance merely qualitatively. Ontological monism is left thereby unthreatened.
       A possible objection, raised by Jason Waller (2009), stipulates that it is problematic to identify the qualitative variety of attributes by way of different essential attributions, since this seems to require that each attribute be determined in terms of unique definitions which individuate essences[12]. According to Waller, this reading seems in friction with Spinoza’s account, according to which, following T. C Mark, “each attribute is not something that can be described or understood at all. … [T]he attributes are objects of perception.”[13] Thus, one might claim, Deleuze still subordinates the attributes to distinctions of reason by rendering them descriptively determinable, rather than ‘perceptually’, in order to qualify substance.
          However, separating attributes and essences from descriptive determinations seems problematic considering that the attributes must be conceived of as really distinct, that is without the help of one another. (P1P10). If we follow Deleuze in claiming that Spinoza here follows Descartes’ notion of real distinction, so that conceptual independence is necessary to individuate substance as causa sui through the attributes, then Waller’s objection seems difficult to endorse. If we add to this Spinoza’s claim that the essence of substance has no existence outside that of the attributes which express its infinite being, then dislodging attributes from possible clear and distinct conceptual expressions does not sit well either. In this regard, Deleuze concludes that “Attributes are Words in Spinoza, with expressive value: they are dynamic no longer attributed to varying substances, but attributing something to a unique substance.” (Deleuze: Pg 45).  This way, Deleuze can claim that attributes are forms common to both finite creatures and infinite substance; insofar as they forms of substance and modes univocally. This seems preferable to the alternative of attributes being abstracted from instances of modal, finite creatures, and transferred to God by analogy, since this again seems to downgrade attributes to the status of mere distinctions of reason (Ibid; Pg. 46-47).
          Having said this, even if God and creatures differ in essence and existence, the same forms are said of both, insofar as they constitute the essence of the former and imply the (finite) essence of the latter within themselves. This allows Deleuze to reconcile two claims which seem apparently at odds in Spinoza: a) that things which have nothing in common cannot be the cause of another (P1P3), and that if a thing is the cause of the essence and ground of another it must nevertheless differ both in the ground of its essence, and in that of its existence (P1P17). Or as Spinoza puts it: “Now the intellect of God is the cause of both the essence and the existence of our intellect; therefore the intellect of God in so far as it is conceived to constitute the divine essence, differs from our intellect both in respect to essence and in respect to existence, nor can it in anywise agree therewith save in name, as we said before. The reasoning would be identical, in the case of the will, as any one can easily see.” (P1P17N) The attributes are thus finally targeting the infinite forms of substance which are common to God, whose essence they conform, as well as modes-creatures, whose finite essence they imply. With these rejoinders in mind, we can sum up our brief and tentative overview of Deleuze’s reading by proposing the ‘Spinozist’ reworked criterion for real distinction as formal[14]:
S1)  Real/Formal distinction – For any x and for any y, x and y are formal-real distinctions of z, if and only if x and y qualify z completely independently without diversifying it, i.e. x and y are formal distinctions of z just in case x and y are said qualitatively and distinctly of z, but quantitatively univocally of z. The corresponding principle here is that of pure qualification (between attributes and a unique substance of which they are said).
             An additional set of problems might follow from Deleuze’s more provocative twofold characterization of formal distinction as implying that qualitatively “there is one substance per attribute” (Deleuze 1992: Pg, 32). We shall leave this issue undecided for the present paper, but note nevertheless that this seems to squarely reintroduce numerical distinction into real distinction. For how are we to understand that in a qualitative sense there is one substance per attribute if not as the claim that there exists in a sense a one-to-one correspondence between essence and attribute? From what we have considered thus far, it is clear Spinoza’s argument for univocity allows for no such ambiguity: the variety of essence attributed to substance must apply to it as instances of the same substance in all cases, and without ambiguity of sense. If not, this reading of Spinoza runs the risk of having smuggled a dubious duplicity of sense which threatens ontological univocity[15].
        Although these tentative approximations to Spinoza’s text are doubtlessly naïve and require a stronger consideration of alternatives, they at least flesh out some of the basic issues surrounding the status of attributes, and perhaps at some of the motivations guiding Deleuze’s ascription of reality to them under the unspoken aid of the ‘formal distinction’.
  • Benedict Spinoza, Ethics, translated by R.H.M Elwes, Digireads, 2008.
  • Thomas C.  Mark, The Spinozistic Attributes, Philosophia 77.7, 1977.
  • Jason Waller, Spinoza’s Attributes and the Intermediate Distinctions of Henry Ghent and Duns Scotus, Florida Philosophical Review, Volume IX, Issue 1, Summer 2009.
  •  Edwin Curley, Behind the Geometrical Method: A Reading of Spinoza’s Ethics, Princeton, NJ: Princeton, 1988.
  • Martial Gueroult, Spinoza, Vol. 1: Deiu Paris: Aubier, 1992.
  • Jonathan Bennett, A Study of Spinoza’s Ethics, Indianapolis: Hackett, 1984.
  • Michael Della Rocca, Representation and the Mind Body Problem in Spinoza, New York: Oxford UP, 1996.
  • Gilles Deleuze, Expressionism in Philosophy: Spinoza, translated by Martin Joughin, Zone Books, Brooklyn, 1990 (fourth printing, 2005).
  • Rene Descartes, Oeuvres de Descartes, ed C. Adam and P. Tannery, 11 vols, Paris 1897-1909.

[1] Descartes, Principles 1.48, 1.49
[2] Descartes, Principles, I.63-I.64
[3] Descartes, Principles, I.56
[4] A more pressing question becomes for Descartes how he can guarantee that conceiving of things as really distinct can target really distinct things; that is how thought can know the distinctions between substances it draws are in fact real. His answer in the Meditations will be that the non-deceptiveness of God guarantees this transitivity between the clear and distinct Ideas which make up our knowledge, and the real which furnishes the world. Thus to our conception of possible different substances there corresponds actual differences among substances. But since substance is by definition that which can exist by itself, and since I may think of the reality of a given substance isolated by a particular Idea in distinct occasions, these would have to correspond to real instances in every case. Spinoza, however, will have no truck with the thought that substance could designate possibility.
[5] We note in passing that Spinoza’ seems to shift slightly from Definition II, which claims that what is finite is what may be limited by something of the same nature, to Proposition 8 which claims that everything which exists as finite is necessarily limited by another finite existence.
[6] In this regard, I disagree with Deleuze’s claim that Spinoza’s argument presents a baffling split from the first eight propositions where the aim he claims is purportedly to show that there cannot exist more than one substance per attribute [NSA], and the latter which determines the uniqueness of substance as such.
[7] Deleuze explicitly tethers this notion of formal distinction to Duns Scotus; who, according to Deleuze, finds no problem in the thesis that one and the same substance may possess an infinity of attributes.
[8] I owe this possible reconstruction to Peter Wolfendale.
[9] Martial Gueroult, Spinoza, Vol. 1: Deiu (Paris: Aubier, 1992).
[10] As Jason Waller argues, this kind of problem seems also to extend to other reading’s such as Bennett’s where extension and thought are ‘really’ distinct but not really ‘fundamental’, even if they are perceived as such by finite intellection. This leaves Bennett open to the charge that the intellect ‘misperceives’ the nature of substance in the plurivocity of attributes. See Jonathan Bennett, A Study of Spinoza’s Ethics (Indianapolis: Hackett, 1984), 64-65. For a critique of Bennett along these lines see Michael Della Rocca, Representation and the Mind Body Problem in Spinoza (New York: Oxford UP, 1996), 164. For Waller’s own reading, which overlaps in identifying attributes with Scotus’ formal distinction (albeit in a slightly different way than Deleuze), see Jason Waller, Spinoza’s Attributes and the Intermediate Distinctions of Henry Ghent and Duns Scotus, Florida Philosophical Review, Volume IX, Issue 1, Summer 2009. In it, Waller demotes formal distinction to an ambiguous status somewhere in-between real distinction and a distinction of reason (which he calls ‘conceptual distinction’). I believe this intermediary status is, however, problematic for reasons which must be left for another occasion.
[11] Edwin Curley, Behind the Geometrical Method: A Reading of Spinoza’s Ethics, Princeton, NJ: Princeton
UP, 1988, 29-30
[12] Jason Waller, Spinoza’s Attributes and the Intermediate Distinctions of Henry Ghent and Duns Scotus, Florida Philosophical Review, Volume IX, Issue 1, Summer 2009, Pg. 97.
[13] Thomas C.  Mark,  The Spinozistic Attributes, Philosophia 77.7 (1977), Pg. 76.
[14] As Deleuze notes, this fundamental reworking of real distinction entails at the same time a reshuffling of the status of modal and distinctions of reason. The former become unyoked from the contingency of unilateral implication, since in the Spinozist monism everything is necessary. Thus we would have to propose something like a principle of ‘bilateral implication’ which wouldn’t compromise the methodological anteriority of substance, to run along Spinoza’s necessitarian argument. Or as Deleuze puts it “Everything is necessary, either from its essence or from its cause: necessity is the only affection of being” (Deleuze 1992; Pg 38).
        Similarly , given the power of finite intellect to discern the communitarian forms in the attributes, which are proper to both substance and modes, the mere ‘abstraction’ which articulates the Cartesian ‘distinction of reason’ falls off; instead wedding it to real distinction in a reworked conception of formal distinction.

[15] Something like this seems to have underwritten Descartes and Suarez in their rejection of the notion of formal distinction in Duns Scotus.