Between Realism and Anti-realism
I want to begin situating Deleuze’s thought relative to realism and anti-realism by addressing two questions Deleuze received after he gave his “Method of Dramatization” talk to the Société française de Philosophie on 28 January, 1967. The first is from Ferdinand Alquié, who expressed the following concern upon hearing Deleuze use examples from science and psychology to make philosophical points:
I understand that Mr. Deleuze criticizes philosophy for making the Idea a conception that is not adaptable, as he would like, to scientific, psychological, and historical problems. But I think that alongside these problems there remain classical philosophical problems, namely problems having to do with essence. In any event, I don't believe, as Deleuze does, that the great philosophers have never posed such questions.1
Coming from the man who was overseeing Deleuze’s work on Spinoza at the time, this criticism got Deleuze’s attention. The second question is from Alexis Philonenko, a Kant and Fichte scholar, who sought clarification of Deleuze’s argument concerning the relationship between the representational and the subrepresentational. Philonenko compared these arguments to Maïmon’s, noting that the differential elements for Maïmon compare to Deleuze’s subrepresentational elements, and the representational compares to the integration of these differentials. A consequence of this approach for Philonenko is skepticism, for we are without a criterion whereby we can discern ‘what we produce and what the object produces...[and] what is produced logically and what is not…’ This leads to his question:
So this is what I want to know: what part does illusion (or the illusory) have in the movement of differential elements?2
Before addressing Deleuze’s response to these two questions I want to set forth how these questions give rise to issues concerning realism. We will then be better able to place Deleuze’s thought into the constellation of debates that surround realism, anti-realism, and speculative realism.
Alquié’s concern with Deleuze’s talk was that the distinctiveness of philosophy was being supplanted by science. Is philosophy merely a midwife for the sciences? For Deleuze the answer is clearly no. When asked whether the topological model Deleuze and Guattari put forth in the conclusion to A Thousand Plateaus is ‘transposable into mathematics [and] biology,’ Deleuze says, ‘it is the other way around,’ and to clarify this point he adds: ‘I feel that I am Bergsonian – when Bergson says that modern science has not found its metaphysics, the metaphysics it needs. It is that metaphysics that interests me.’3 This metaphysics should not, Deleuze stresses, be ‘in the style of Kant,’ but rather ‘in the style of Whitehead.’4 What this style involves will be discussed below, but it is important first to appreciate the fact that Deleuze is interested in the metaphysics modern science needs, rather than the science philosophy needs.
With this turn to metaphysics we also return to the choice that has come to characterize much of post-Kantian philosophy—one can follow Humean empiricism and its positivist and nominalist variants or one can follow Kantian critique and its idealist and rationalist variants, but one cannot follow both. Alexis Philonenko’s question regarding illusion comes from the Kantian perspective. For Kant there is the transcendental illusion whereby one extends the concepts of the pure understanding to form judgments regarding things that are beyond the limits of possible experience (e.g., the existence of God, the claim that the world must have a beginning in time, etc.). The point of Kant’s critical project is to determine the limits that draw the line between legitimate and illusory claims to know what is real and what is not. But even with this project in hand, Kant admits that the transcendental illusion ‘does not cease even after it has been detected and its invalidity clearly revealed by transcendental criticism.’ (A297) The mind naturally and inevitably slips into illusion. On Philonenko’s reading of Maïmon, we cannot even be certain about the line we draw between legitimate and illusory judgments since it is uncertain whether the integration of differentials from which our judgments arise are the products of our own making or a real consequence of a real object. Similarly for Hume, when we are led through the association of ideas and habit to impute causality and necessity to objective relationships, or to infer a substantial self that accompanies our thoughts and perceptions, Hume too would leave us with no way to differentiate between reality and illusion other than the strength and vivacity of the beliefs themselves, which is certainly no guarantee. As Deleuze says of Hume, ‘From the point of view of philosophy, the mind is no longer anything but delirium and madness. There is no complete system, synthesis, or cosmology that is not imaginary.’5 Deleuze’s response to Philonenko’s question may thus seem surprising, but it is unambiguous—for him there is no place for illusion with respect to the subrepresentational field – this field is ‘perfectly determinable’ as Deleuze puts it – and ‘the illusion only comes afterward, from the direction of constituted extension and the qualities that fill out these extensions.’6 For Deleuze, therefore, the real is to be associated with processes that constitute the givenness of objects rather than with the constituted, identifiable objects and categories themselves. It is for this reason that Deleuze identifies the style of metaphysics he is interested in with Whitehead instead of Kant; rather than base an understanding of reality upon identifiable categories and forms of judgment, Deleuze argues that our scientific, representational understanding of reality presupposes subrepresentational processes that are not to be confused or identified with that which is identifiable as a result of these processes.
In the choice between Kant and Hume, therefore, it appears Deleuze’s choice is clear—Hume. But where does this leave Deleuze relative to realism, and metaphysics in particular? At first it might seem that Deleuze should be placed solidly within the anti-realist camp. After all, if what can be said of reality may, in the end, be illusory, and if much like Hume Deleuze is willing to say that ‘there is no complete system, synthesis, or cosmology that is not imaginary,’ then it would seem to be difficult to place Deleuze within the realist camp. But is he an anti-realist?
To answer this question I will sketch out the difference between realism and anti-realism by turning to Quentin Meillassoux’s critique of correlationism, which ultimately is a radical critique, as Meillasoux puts it, of any ‘contemporary opponent of any realism.’7 Put briefly, correlationism is the position which holds that we cannot know reality as it is in-itself but only as it is for-us, as a correlate of consciousness, language, culture, conceptual scheme, etc. Meillassoux notes that Fichte’s Principles of Scientific Knowledge is the ‘chef d’oeuvre of such a correlationism’ in that it shows how any attempt to posit a reality as independent of any positing is still a reality that is posited as such. Correlationism, however, is not committed to a subject-object dualism but more importantly rejects any attempt to hypostatize a reality that would be an autonomous and independent reality. This is why Meillassoux will understand correlationism not as ‘an anti-realism but [as] an anti-absolutism,’ for it is invoked ‘to curb every hypostatization, every substantialization of an object of knowledge which would turn the latter into a being existing in and of itself.’8 Whether these autonomous beings are Ideas or objects, correlationism, for our purposes, is anti-realist in that it holds that any reality in-itself is always from the start an in-itself correlated with something else.
Let us turn now to more traditional understandings of anti-realism—namely, those of the analytic tradition. The term itself was first used by Michael Dummett to characterize those positions that argue that reality is what is necessary in order to state meaningful, true sentences. Dummett will refer to this as semantic realism, and others will call it deflationary realism, but it is clear that it is a form of correlationism or anti-realism in that the real is real only insofar as it is necessary to the utterance of meaningful, true sentences – or, as Quine famously put it, “To be assumed as an entity is, purely and simply, to be reckoned as the value of a variable. In terms of categories of traditional grammar, this amounts roughly to saying that to be is to be in the range of reference of a pronoun.”9 Donald Davidson will later criticize Quine for continuing to adhere to what he refers to as the third dogma of empiricism, by which Davidson means the continuing adherence to the scheme-content dualism whereby a pure content (‘sensory firings’ in Quine’s case) is forged and translated by a conceptual scheme into the content of our knowledge of reality. In rejecting the third dogma Davidson may appear to be a realist – he does, after all, claim to uphold a form of monism which asserts that there is nothing but objects and events – and yet when he argues for the reality of these objects and events his argument in the end is that they (especially events) are real only because we could not utter the meaningful sentences we do without them. And finally, though we could go on, in David Lewis’ modal realism, a realism that goes well beyond common sense naïve realism in that Lewis accepts the reality of possible worlds, Lewis argues for the reality of these possible worlds by showing that they are necessary if our everyday counterfactual claims are to be meaningful and true. In all these cases and in others we could list, the real is real only by virtue of the necessary role it plays in some other process (namely, a linguistic and semantic process), and thus correlationism and anti-realism are alive and well within the analytic tradition.
To move beyond correlationism, Meillassoux draws support from an important ontological claim that he derives from it, even though he admits that correlationism ‘is not an ontology, strictly speaking.’ Meillassoux argues that ‘[a]ccording to the correlationist, if I remove myself from the world, I can’t know the residue. But this reasoning supposes that we have access to an absolute possibility: the possibility that the in-itself could be different from the for-us. And this absolute possibility is grounded in turn on the absolute facticity of the correlation.’10 In other words, following from the epistemological claim that one cannot ‘know the residue’ of what would be independent of what is for-us, Meillassoux draws the ontological conclusion that the world could be other than it is for-us, which entails, in turn, the possibility of a world without correlation, a world without givnness. From here it is a simple step to the undermining of correlationism since if the real, X, can only be known as a posited X, then it follows from the ontological conclusion concerning the facticity of the correlation itself that there is the possibility of an X that is not posited, a world without givenness for-us.
One of the challenges that attends Meillassoux’s conclusion, and one that is central to what has been called speculative realism (but which has not been ignored within the analytic tradition as well), is to account, in a noncircular manner, for the facticity of thought itself—especially the normative patterns of thought—and hence for the relationship between thought and the structure of reality from which thought emerges. It would lead us too far astray to begin to detail these debates, but it should be noted that the Meillassoux’s own concerns regarding realism and the facticity of thought are widely shared and are being addressed from a number of different perspectives.
To begin to return this discussion back to the relationship between philosophy and science—and clearly science has a lot to say about the origins of thought—we will be aided by discussing necessary laws. We can begin with David Lewis’s Humean understanding of necessity, what he calls Humean Supervenience. On Lewis’s reading of Hume, any claims we make regarding the world that we take to be true, including claims concerning necessary laws, supervene upon a given distribution of particular facts. There cannot be a change in this distribution without a change in the claims or truths that supervene upon them. Given the laws of probability, the chance a single throw of the dice will give me a six are one in six. Three or four sixes may show up in a row, but given a large enough number of throws the number of times I throw a six approaches one in six. These laws of probability therefore supervene upon a given distribution of particular facts (rolls of the dice) in the world up to and including time1. If there is a non-zero chance, however, that after time1 sixes come up every time then that would effect the chance distribution at W at time1—it would be something higher than one in six, but this contradicts Humean supervenience. Lewis refers to this as an undermining future. In his analysis of Hume, however, Meillassoux argues that what makes Hume’s understanding of necessity possible is that there be a totality relative to which the particular facts are compared, and hence upon which the necessary laws supervene. It is the totality of throws at time1 combined with the throws after time1 that gives rise to undermining futures. Following Badiou and Cantor, however, Meillassoux argues for the not-All that cannot be totalized and which therefore undermines the necessary laws that would supervene upon a given totality. The notion of an undermining future would not even arise on this reading. This is not to say that there are no particular facts or regularities between facts. Within a large set of observations the odds of sixes appearing may be one in six, or there may be countless other regularities, but the ‘laws’ that supervene upon these regularities are, Meillassoux argues, ‘contingent. They are not necessary. As Hume said, we are unable to demonstrate any such necessity.’11 Turning now to Deleuze we have what you might call Deleuzian supervenience, whereby the axiomatic supervenes upon the problematic, but the problematic forever exceeds the axiomatic, it is the ‘power of the continuum, tied to the axiomatic but exceeding it.’ Axiomatics, or what Deleuze will also call major or royal science, draws from problematics the necessity of inventing and innovating, and problematics, or minor or nomad science, calls upon axiomatics to actualize solutions to the problems it lays out. Deleuze and Guattari are clear on this point: ‘Major science has a perpetual need for the inspiration of the minor; but the minor would be nothing if it did not confront and conform to the highest scientific requirements.’12 To begin clarifying this point we can turn to Spinoza, and to Deleuze’s review essay of Gueroult’s first volume on Spinoza’s Ethics. In this essay Deleuze argues that what is important about Gueroult’s approach is that it doesn’t begin with the idea of God (God enters the scene with the sixth definition and the ninth propostion). Does this mean that the first six definitions and eight propositions are inessential to Spinoza’s project – mere preliminary work Spinoza had to get out of the way before the real work began? For Gueroult and for Deleuze the answer is ‘no’. When the answer is yes, Deleuze argues, we get
…two misreadings of the attribute: 1) the Kantian illusion that makes attributes forms or concepts of the understanding, and 2) the neo-Platonic vertigo that makes attributes already degraded emanations or manifestations.13
With this claim we come to the central feature of Deleuze’s Spinozism—namely, the relentless tracking down of transcendence in all its forms. And here we have two misreadings of Spinoza that call upon transcendence. There is the Kantian illusion that turns attributes into concepts that transcend substance as that which is then applied to substance (this is also Badiou’s misreading of Spinoza); and then there is the neo-Platonic misreading of Spinoza that sees substance as a pure One and the attributes as degraded emanations of this One (Hegel and Bertrand Russell are guilty of this reading). Substance, however, is neither One nor multiple, but is rather a multiplicity, a substantive multiplicity. The attributes, therefore, are fully real and are really distinct from one another, but they are not numerically distinct and they do not transcend substance, nor are they the degraded emanations of substance—they constitute, to repeat, what Deleuze will call a substantive multiplicity:
The logic of real distinction is a logic of purely affirmative difference and without negation. Attributes indeed constitute an irreducible multiplicity, but the whole question is what type of multiplicity. The problem is erased if the substantive ‘multiplicity’ is transformed into two opposed adjectives (multiple attributes and one substance).14
For Deleuze, therefore, Spinoza does not begin with God but with the problematic, with substantive multiplicity. To understand Spinoza adequately, therefore, it is essential to clarify what is meant by substance, and in what way it is not to be confused with either God or the modes and attributes. By far the most common interpretation of substance is to understand it as a being, albeit an absolute, infinite being, that is in some way parceled by the modes and attributes. At the risk of slipping into an illegitimate anachronism, we could apply Heidegger’s ontological difference and say that substance is typically understood onto-theo-logically as being in contrast to Being. Although I will not, as some do, assert that substance is to be understood as Being, as no-thing in Heidegger’s sense of the term, I do think that substance is not to be conceived as a being either. This reading of substance naturally lends itself to the two misreadings of the attributes listed above. The reading I prefer is largely indebted to H.F. Hallett’s idiosyncratic reading of Spinoza. For Hallett substance is ‘absolutely indeterminate,’ or in-determinate as I prefer. Hallett’s interpretation is by no means the consensus view, but it does have important textual support. For example, since God is defined in the Ethics as absolutely infinite (1D6), God can in no way be limited or be in any way determinate, for to be determinate entails being related to another that it is not. As Spinoza argues in his letter to Jelles, ‘all determination is negation,’ and hence substance cannot be determinate if is to be absolutely infinite and affirmative, which is precisely how Spinoza understands it, as he makes clear in a letter to Lodewijk Meyer where he claims substance is the ‘infinite enjoyment of existing.’ This is also accounts for why God is absolutely infinite rather than infinite in its own kind, as the attributes are, since this would require being a determinate form of infinite and hence a form that could, when understood conceptually by way of the understanding – namely the infinite mode of understanding—be related to what it is not, to what is other than it. Yet another reason for adopting Hallett’s reading follows from a claim Spinoza makes in a letter to Jelles (letter 50) that anyone who ‘calls God one or single has no true idea of God’ because, as we’ve already noted, all determination is negation and God, as substance, is the ‘infinite enjoyment of existing,’ or infinite power as Spinoza will also put it.
To further clarify Spinoza’s position, and the Spinozist context within which Deleuze’s thought navigates, we can turn to Deleuze and Guattari’s claim, in A Thousand Plateaus, that God is a lobster, a double articulation. It is all too easy to underestimate the philosophical importance of this claim, but we see it at work in Deleuze’s essay on Gueroult, where the first eight propositions correspond to the first articulation; or, as Deleuze puts it, ‘the first eight propositions represent a first series through which we ascend to the differential constitutive elements’ – the attributes. As Deleuze had stressed earlier in the essay, there is ‘no ascension from attributes to substance…to absolutely infinite substance’; rather, there is an ascension through a ‘regressive analytic process’ to the ‘differential constitutive elements’ themselves, to the substantive multiplicity. Then there is the second articulation, the second series found in the 9th-11th propositions ‘through which,’ Deleuze argues, ‘the idea of God integrates these elements and makes clear it can be constituted only by all these elements together.’ The attributes, as a multiplicity of incommensurable and really distinct entities, come to be integrated by the power of causa sui whereby ‘essence is the cause of the existence of substance and the cause of the other things that derive from it.’ Understood in this way, God as causa sui is both the condition that enables the regressive analytic process that leads to a multiplicity of really distinct attributes – first articulation – and the conditioned that is the integration of this multiplicity – second articulation. God is self-caused, or God is a lobster, a double articulation.
We can now return to Alquié’s question and to Deleuze’s response, and in turn situate Deleuze relative to realism and anti-realism. As discussed earlier, Alquié was troubled by the use of examples from science in Deleuze’s talk, as if philosophy were for Deleuze a mere expository tool in the service of science. In response to Alquié’s question Deleuze stresses that he does ‘believe in the specificity of philosophy, and furthermore, this belief of mine derives from you yourself’; moreover, Deleuze immediately adds, ‘Even the concepts such as singular and regular, or remarkable and ordinary [i.e., philosophical concepts for Deleuze], are not exhausted by mathematics. I want to call on Lautman’s theses: a theory of systems must show how the movement of scientific concepts participates in a dialectic that surpasses them.’15 In other words, and to state this in terms used above, royal science and axiomatics develops concepts that participate in, or supervene upon, nomadic and problematic philosophical concepts, and the latter are not exhausted by the former. Deleuze’s seconding of Alquié’s belief in the specificity of philosophy should thus not be underestimated, especially in light of recent calls among philosophers to return to realism. Among the efforts to return to realism, there have been roughly two distinct approaches. In the first it is argued that the way to do justice to reality in-itself is through mathematics. For Badiou ‘ontology is mathematics,’ and Meillassoux tracks Badiou’s approach quite faithfully, arguing that ‘what is mathematizable cannot be reduced to a correlate of thought.’ The second approach accounts for the reality and nature of objects themselves by drawing from science, and dynamic systems theory in particular (DeLanda’s and Protevi’s work loom large here). There is tremendous work being done in each of these two approaches, but the concern, and this is precisely the concern that prompted Alquié’s question, is that these approaches will reduce philosophy to being simply an adjunct to mathematics and/or science. What is lost then is the distinctiveness of philosophy itself.
To clarify what it is that makes philosophy distinctive for Deleuze we can turn to how Deleuze’s understanding of the principle of sufficient reason, a good philosophical principle if ever there was one, differs from the positions of Lewis and Meillassoux. Since laws supervene upon the particular facts of the world for Lewis, and worlds, both actual and possible, that constitute a totality, then the claims of science can adequately represent the distributions and regularities of each world. The distribution of particular facts at each world thus provides a sufficient reason for the laws and claims that supervene upon these worlds. For Meillassoux, by contrast, the not-All undermines any necessity other than the necessity of contingency, and hence Meillassoux rejects the principle of sufficient reason. We cannot state why things are the way they are rather than another way. To justify this point Meillassoux relies upon the force of negation and contradiction (Hegel lurks behind the scenes here). Meillassoux explicitly argues, for example, ‘that to be contingent you must not be contradictory, because if you are contradictory you are everything and you can’t change.’16 A similar claim is made in After Finitude where it is argued that a philosophy of becoming will have to do away with contradictory entities for they cannot change and become what they are not—they already are what they are not. And in his critique of correlationism itself Meillassoux calls upon Fichte’s strategy, what Meillassoux calls ‘pragmatico-genetic contradiction’, to show that the content of a claim is contradicted by the very way in which this claim is put forward. The content of the claim ‘X is not posited’ is contradicted by the very fact that it is posited as not posited. Similarly, the content of correlationism’s claim that all reality in-itself is a correlate for-us is contradicted by the absolutization of contingency that goes along with this claim and which results in the facticity of the correlation itself. It is on the basis of these arguments, and the binary logic of contradiction, that Meillassoux rejects the principle of sufficient reason. ‘If you want to speak about what exists,’ Meillassoux argues, ‘you can only describe, as phenomenology does,’17 but the fact that it is something rather than nothing ‘cannot be described.’18 Despite the obvious differences between Lewis and Meillassoux—one accepts the principle of sufficient reason and the totalization of worlds, both actual and possible, while the other does not—they each share a commitment to a logic of representation—an ‘it is or it isn’t’ logic—that Deleuze identifies with the actualization of the actual world.
Deleuze, by contrast, and following from his Spinozism, argues for the development of philosophical concepts that are not exhausted by scientific or mathematical concepts. Philosophical concepts presuppose a logic of expression or sense rather than a logic of representation. Nearly thirty years after his “Method of Dramatization” talk, Deleuze will continue to make much the same point. As Deleuze and Guattari argue in What is Philosophy? science sets out to map functions that represent the actualization of the actual, or that accurately describe the regularities of the world. Philosophy, by contrast, creates concepts which counter-actualize the actual and involve the problematic upon which the actual supervenes. Understood in this way, Deleuze, unlike Meillassoux, will continue to adhere to the principle of sufficient reason. As he argues in his book on Leibniz, the principle of sufficient reason does not imply that every actuality—Adam’s eating of the apple for example—was determinately detailed and sketched in advance such that the actualization itself was a foregone conclusion to anyone with complete knowledge (such as God); to the contrary, for Deleuze the principle of sufficient reason is the substantive multiplicity, in the Spinozist sense sketched above, that is the sufficient reason for the determinate. As Deleuze will argue in numerous places, this substantive multiplicity is real but not actual—it is what he will also call the virtual. Deleuze’s logic of expression, however, does not entail a rejection of the actual world, an attempt to get out of this world and the logic of representation. It is, rather, an attempt to intensify the actual world, to problematize it, and to do so while remaining fully within the actual world. To argue otherwise would entail allowing for transcendence to sneak in through the back door, and Deleuze, like Spinoza, is keen to hunt down transcendence in all its forms.
From this perspective, therefore, Deleuze is neither a correlationist anti-realist, nor is he a straightforward realist. Deleuze is not a correlationist for there is no binary logic being followed, no for-us distinct from an in-itself; rather, Deleuze is a monist developing philosophical concepts that follow a logic of expression or sense rather than a binary logic of contradiction, a logic of representation. Each actual entity, or each mode for Spinoza, is thus to be understood as an expression of substance, a substance that is only identifiable as such when expressed in a mode. Substance, therefore, does indeed turn around its modes, as Deleuze had said was how he understood Spinoza. Each actual entity is hence a real, determinate expression of substance, of substance as substantive multiplicity, or it is the expression of the nondenumerable which Deleuze and Guattari define as being ‘neither the set nor its elements [the one or the multiple]; rather, it is the connection, the “and” produced between elements, between sets, and which belongs to neither, which eludes them and constitutes a line of flight.’19 Between the elements of the correlation, therefore, between the for-us and the in-itself, there is the “and,” and it is this “and” that ‘constitutes a line of flight’ that escapes and eludes every correlation, every attempt to reduce reality to a correlation. Similarly Deleuze is not an anti-realist in the analytic sense of the term for Deleuze accepts and argues for the claim that the real is not reducible to what is necessary for meaningful sentences, but asserts rather that the real as problematic, substantive multiplicities exceeds these representations and is irreducible to them, and thus the logic of representation presupposes the nonsense and nondenumerable upon which it supervenes. Deleuzian supervenience is thus not like Lewis’s Humean supervenience, since Lewis follows the logic of representation whereby the necessary laws supervene upon the particular facts of the world, and a change in the latter entails a change in the former. These laws therefore re-present the particlar facts of the world. For Meillassoux there is no supervenience since there is no correlation between a totalized world and that which represents this world, even though Meillassoux continues, as we saw, to adhere to a logic of representation. Deleuzian supervenience, by contrast, follows a logic of sense, and thus the actual, the axiomatic, supervenes upon the problematic which is not a correlate it represents, but rather is the expressiveness and power of the actual itself.
Deleuze is also not a straightforward realist either, if by realism one affirms an autonomous reality with properties and features that are independent of anyone’s beliefs, linguistic practices, conceptual schemes, etc. The very idea of an autonomous in-itself reality presupposes, as I’ve argued elsewhere (and draw heavily upon Latour when doing so), a stabilization of events that are indeed real but are neither autonomous or heteronomous. It is only as actualized, and hence in accord with the logic of representation, that one can then say of a substantive multiplicity that it consists of being a determinate duality between an autonomous in-itself and a heteronomous, correlationist in-itself-for-us, between facts and artifacts, between realism and anti-realism. Between realism and anti-realism, therefore, is the “and,” the substantive multiplicity that is tracked by the metaphysics Deleuze believes science needs. Deleuze does believe this metaphysics is possible and he adopts it as the distinctive task of philosophy.
1Desert Islands, p. 106.
2 Ibid., p. 114.
3 Collapse, p. 41.
5 Empiricism and Subjectivity, 83.
6 Method of Dramatization, p. 115.
7 Collapse, p. 408.
8 After Finitude, p. 11.
9 From a Logical Point of View, p. 13
11 Collapse, p. 441.
12 A Thousand Plateaus, p. 486.
13Desert Islands, “Gueroult’s General Method for Spinoza,” p. 149