Schopenhauer, Metaphysics, Humility and Panpsychism Alistair Welchman To be given at the North Texas Philosophical Association, March 31st-April 1st 2017 Please contact author to cite 1 Introduction Schopenhauer develops a transcendent metaphysics, based on a claim that goes beyond the possibility of experience. I shall argue however that this claim does not depend closely on transcendental idealism, following a recent similar interpretation of Kant due to Rae Langton.1 This makes for some surprising connections to contemporary metaphysics. 2 Transcendent Metaphysics Schopenhauer’s famous metaphysical view is that Kant’s thing in itself should be identified with the will: after ‘the world as representation, or appearance’ is subtracted, ‘all that remains is the purely metaphysical, the thing in itself which we will recognize in the second book as the will.’ (WWR2 SW3:22). a) Schopenhauer’s Argument The first book of the World as Will and Representation is devoted to the world as representation, but is constantly interrupted by references to the world as will. In the very first section of the first volume of The World as Will and Representation, Schopenhauer describes an ‘inner reluctance’ (WWR1 SW2:5) he expects the reader to feel at the idea that the world is (only) representation. The root cause of this reluctance is that the world as representation ‘exists only relatively’ (WWR1 SW2:8-9), that is objects (representations) ‘can be exhaustively traced back to the necessary relation of objects to each other, so that the being of objects consists in nothing but this relation.’ (WWR1 SW2:7) Inter-object relations come in several forms, in particular spatio-temporal relations and causal relations. In his account of matter, Schopenhauer argues that ‘matter is, in its entirety, nothing other than causality’ (WWR1 SW2:10). So all the material, spatio-temporal properties of objects are relational. At the start of WWR Book II, Schopenhauer outlines two ways in which non-relational properties might enter into the world. First, he argues that all scientific explanations eventually postulate inexplicable fundamental forces. The argument depends on Schopenhauer’s view of causal explanation in general. He thinks that the behavior of any empirical object is exhaustively predictable from two factors: the situation the object is in and its ‘character’ or inner propensities. Sometimes we can explain the causal powers of an object on the basis of the causal powers of its constituent parts. But such explanations have to come to an end at some point in unanalyzed powers or fundamental forces. These Naturkräfte are what is missing from the world (considered only) as representation or object. Second, Schopenhauer distinguishes between the what things are like ‘from the outside [von außen]’ (WWR1 SW2:118) and their ‘inner essence [inneres Wesen]’ (WWR1 SW2:116). This is, of course, more 1 Rae Langton, Kantian Humility: Our Ignorance of Things in Themselves (Oxford: Clarendon Press, 1998). or less what one would expect from Schopenhauer’s Kantian heritage, which distinguishes how things appear (to us) from how they are in themselves. b) Humility Although Schopenhauer attempts to violate Kant’s epistemic strictures, I think it is important to see what the significance of those strictures is. Like Schopenhauer scholarship, Kant scholarship has a skeptical stance towards metaphysical entanglements. But Rae Langton has developed an interpretation of Kant according to which our ignorance about things in themselves, the thesis she calls Humility,2 is a substantive metaphysical claim. On this view, only the extrinsic properties of objects are cognitively available, their intrinsic properties are not. Importantly, Langton’s interpretation is neutral about transcendental idealism. In fact, the view is not just an interpretation of Kant, but a viable contemporary option in the philosophy of science (where it is known as epistemic structural realism).3 So first I am going to argue that Schopenhauer’s understanding of the role of the thing in itself in his system is close to the role that Langton’s Kant interpretation gives to the thing in itself in Kant himself. This is clearly an unstable situation, for if the thesis is correct, the properties of things in themselves are cognitively inaccessible, which is just what Schopenhauer denies. Here I am going to appeal to a contemporary defense of panpsychism to make Schopenhauer more plausible.4 This should not be surprising because Schopenhauer is regularly cited as an influential figure in the history of panpsychism.5 But in fact Schopenhauer is not a panpsychist: rather his argument is made plausible by the fact that it is structurally identical to these arguments for panpsychism. Langton frames her account of Kant in terms of Leibniz’s substance metaphysics. Thus, Langton phrases the difference between phenomena and things in themselves, which she calls ‘the Distinction,’ like this: ‘Things in themselves are substances that have intrinsic properties; phenomena are relational properties of substances.’6 The problem that her view of Kant is designed to solve is that Kant appears to think that the fact that our cognition is receptive (‘Receptivity’) entails that we have no cognitive access to things in themselves. She describes this conclusion as ‘Humility’ and defines it as follows, in accordance with the substance talk of the Distinction: ‘We have no knowledge of the intrinsic properties of substances.’7 But Receptivity does not entail Humility, at least not without a further premise. Langton supplies the wanted extra premise (‘Irreducibility’): ‘The relations and relational properties of substances are not reducible to the intrinsic properties of substances.’8 As I have already demonstrated, Schopenhauer’s anxiety about the world as representation is precisely that it gives us access only to the relational properties of things. 2 Langton, Kantian Humility, 21. James Ladyman "Structural Realism", The Stanford Encyclopedia of Philosophy (Winter 2016 Edition), Edward N. Zalta (ed.), URL = <https://plato.stanford.edu/archives/win2016/entries/structural-realism/>. 4 Galen Strawson ‘Mind and Being: The Primacy of Panpsychism’ in Panpsychism: Contemporary Perspectives, ed. Godehard Bruntrup and Ludwig Jaskolla (Oxford: Oxford University Press, 2017), 75-112. 5 David Skrbina, Panpsychism in the West (Cambridge, MA: MIT Press 2005), Chapter 5. 6 Langton, Kantian Humility, 20. 7 Langton, Kantian Humility, 21. 8 Langton, Kantian Humility, 124. 3 Schopenhauer is not famous for talking about substance, but in fact he does use the terminology fairly consistently in relation to the thing in itself, calling it, for instance, the ‘substance of nature’ (WWR1 SW2:168). Similarly, in WWR II, he identifies the ‘intrinsic essence [Wesen an sich] of appearances’ with their ‘intelligible substrate [Substrats]’ (SW3:13, tm), and later on, talking specifically about the body, he describes the will is the ‘metaphysical substrate, as the in-itself of the body’s appearance’ (WWR1, SW3:240, tm). But this does not matter too much as Langton’s claims can be reformulated without loss omitting the term ‘substance:’ Distinction, for instance, would be: ‘Things in themselves … have intrinsic properties; phenomena are relational properties of things in themselves.’ And so on for the others. The crucial issue concerns the distinction between intrinsic and extrinsic or relational properties. Langton uses the English terms ‘intrinsic’ and ‘extrinsic’ to translate Kant’s ‘etwas Inneres’ (A265/B321; A274/B330), literally ‘something inner’ and ‘äußere Verhältnisse,’ literally, ‘external relations.’9 Langton uses ‘intrinsic’ and ‘extrinsic’ rather than the more obvious ‘inner’ and ‘outer’ because she wants to argue that Kant’s endorsement of Humility does not depend on his transcendentally idealist account of space. ‘Outer’ or ‘extrinsic’ relations are therefore best understood in a more general sense, as extrinsic or non-intrinsic properties. Langton uses definitions from Lewis and Kim: x is an intrinsic property of y iff y’s possession of x does not entail the existence of anything except y.10 Both Kant and Schopenhauer argue that spatial properties are extrinsic. But Langton wants to resist the claim that extrinsic properties are restricted to spatial properties, in part because she wants her analysis of Kant to be consistent with a rejection of transcendental idealism’s account of space: as it stands Humility is not idealism, just, humility.11 Schopenhauer’s understanding of relational properties shares Kant’s ambiguity. He relies on an inner/outer distinction that is clearly narrowly concerned with the spatially external experience of a subject. For instance, he says ‘we can never reach the essence of things from the outside [von außen]’ (WWR1, SW2:118). Correlatively, it is through a kind of inner experience that we can, he thinks, gain ultimate access to the intrinsic properties of things i.e. thing in themselves. But this very use of inner experience to access intrinsic properties implies that the larger notion of intrinsic is also at play. For instance, when discussing the fact that materialists (incorrectly) posit matter as a thing in itself, he describes what they do (and hence the thing in itself) as something that ‘exist[s] intrinsically and absolutely [an sich und absolut existierend]’ (WWR1, SW2:33). And we have already seen that external experience is extrinsic not (just) because it is spatially external to the body but because matter is itself relational. This is a point that applies even if transcendental idealism is false. Receptivity is a potential obstacle to understanding Schopenhauer along the lines of Langton’s Kant. Kant is clearly committed to the idea that things must ‘affect’ us. Schopenhauer is critical of this view, arguing that in the doctrine of affection Kant ‘makes an inference to the thing in itself as the cause of appearance, applying the principle of sufficient reason in a way he himself forbids as transcendent.’ (WWR1 596). Thus, on the face of it, Schopenhauer seems to deny Receptivity. This obstacle can however be overcome. Consider the picture that Schopenhauer does endorse: (nonrepresentational) sensations are referred to an object as their cause. This object is a representation, an 9 Langton, Kantian Humility, 18. Langton, Kantian Humility, 18 and 18-9 note 6 where she traces some of the complications of these definitions. 11 Langton, Kantian Humility, Chapter 10. 10 element of the world as representation. But Schopenhauer’s worry about extrinsic properties is just that there appears to be something wrong with thinking that this object is just representation. It must also be something in itself. Thus, empirical receptivity is enough to generate Humility, at least as far as Schopenhauer takes Humility. Again, the same thing would be true even if objects were not representations: our external experience of them would still be of them as purely relational; what they are in themselves would still be inaccessible. Finally, we come to the Langton’s missing premise, Irreducibility (of extrinsic properties to intrinsic ones). This issue depends crucially on how one interprets causal powers, i.e. the dispositions that make it such that an object has the behavioral profile that it does. One intuitive suggestion would be that such powers or dispositions are intrinsic. Although the exercise of a causal power would be an extrinsic property (since it would imply the existence of what the power was exercised on), the mere disposition to have an extrinsic property might be thought to be intrinsic. But then intrinsic properties are perfectly cognitively accessible. But there is another interpretation of intrinsic properties that would make powers non-intrinsic: the guiding intuition is this: ‘things could be just as they are with respect to their intrinsic properties, yet different with respect to their causal powers,’ in particular if the laws of nature were different.12 In this case, intrinsic properties cannot be read off from (reduced to) extrinsic properties like causal powers. On the face of it, Schopenhauer seems not to share this intuition. If we identify intrinsic properties with intelligible character and causal powers with empirical character, then the former determines the latter so that it is not possible for latter to vary while the former remains constant. But this is not the end of the story. For the fact that intelligible and empirical character have the same content makes it implausible that one could have intrinsic properties and the other not. A better model might be to say that a set of causal powers comprising something’s character is always something extrinsic; it is redundant to duplicate this set of powers at the level of the thing in itself. Rather the thing in itself, the will, remains ‘free’ with respect to empirical character, which could always have been something else. Langton’s characterization of this view of intrinsic properties as being properties ‘compatible with loneliness and lawlessness’ is particularly Schopenhauerian: the freedom of the will just is its independence from all law, i.e. the principle of sufficient reason. c) The Will: Kantian Limitations. So, Schopenhauer’s conception of transcendental metaphysics can be understood along the same lines as Langton’s interpretation of Kant: we have cognitive access only to extrinsic properties of things; but things must also have an in-itself side or possess intrinsic properties that we have no cognitive access to. But, of course, Schopenhauer does think we have access to the thing in itself. How does he think this is possible? The crucial factor is the body. In intentional action, I am aware of my body in two distinct ways: as an object of representation among other objects, but also from the inside as willing. Schopenhauer has a pithy slogan for his basic metaphysical result: awareness of our own willing is ‘causality seen from within,’ an insight that he says is ‘cornerstone of my whole metaphysics.’ (FW, SW1:145). Since causality is matter, the character of anything, as it is in itself, is will. But to get to this result, he must overcome two formidable barriers. First, he must show how it is possible for us to have any knowledge about what 12 Langton, Kantian Humility, 118. the intrinsic properties of any things are, given the Kantian background we have discussed. Second, even if one concedes that our awareness of willing gives insight into our intrinsic properties, Schopenhauer must show (what to many seems the weakest link of his philosophy) that the same will is the in-itself of everything. There is little doubt that Schopenhauer has trouble making his metaphysical view about the nature of intrinsic properties consistent with his broadly Kantian principles. Even he admits that inner experience ‘does not provide a remotely exhaustive and adequate cognition of the thing in itself’ because ‘it is bound up with the form of representation, … breaks down into subject and object’ and ‘the form of time still remains.’ So Schopenhauer argues that inner perception ‘has thrown off the greater part of its veil’ but not all of it (WWR2, SW3:220, tm). However, there is another way of resolving this tension: simply allow the original exception to Kant to go through. Perhaps this can be done within a broadly Kantian framework of transcendental idealism. But this is not necessary, since the intrinsic/extrinsic distinction is independent of transcendental idealism. Then the claim would be that in inner experience we are aware of some(thing) that, while it does have transcendentally real temporal properties that are relational, also has intrinsic properties. This position then looks similar to, for instance, Russell, who writes ‘[a] piece of matter is a logical structure composed of events; the causal laws of the events concerned, and the abstract logical properties of the spatio-temporal relations, are more or less known, but their intrinsic character is not known.’13 Schopenhauer would quibble with the term ‘logical’ (since, at least in uncomplicated cases, we have intuitive not rational knowledge of causation) but would otherwise be in agreement. Furthermore, Russell thinks that we do have insight into the intrinsic properties of at least one kind of thing: ‘percepts, [which] are the only part of the physical world that we know otherwise that abstractly. As regards the world in general, both physical and mental, everything that we know of its intrinsic character is derived from the mental side.’14 Here there is an important similarity to Schopenhauer and an important difference. The similarity is that something ‘inner’ or broadly ‘subjective’ yields insight into intrinsic properties. The difference is that for Russell this is what would now be called the quale of a conscious experience, what it is like to have that experience, while Schopenhauer is concerned not with our awareness of the conscious features of awareness, but rather with the content, the accusative, of a particular kind of awareness, that directed inside to ourselves. c) The Will: Generalization. The second barrier Schopenhauer has to overcome lies between the view that the intrinsic nature of my body is will and the claim that will is the in-itself of everything. Schopenhauer appears to use an argument from ‘analogy:’ my body is a representation; I have special insight into its intrinsic properties (it is will); other things are also representations; so, they have the same intrinsic property as my body, ‘after all,’ Schopenhauer writes, ‘what other sort of existence of reality could we attribute to the rest of the corporeal world?’ (WWR1, SW2:125). On the face of it, this sounds absurd. Schopenhauer however forestalls one misunderstanding: intrinsically, a stone is indeed will; but ‘this should not be given the absurd meaning that the stone is moved by a motive in cognition just because that is how the will appears in human beings.’ (WWR1, 13 14 Analysis of Matter (Nottingham: Spokesman Books, 2007), 384. Analysis of Matter, 402. SW2:126) Motives are representations (that’s how Schopenhauer defines motives). So, they are extrinsic properties of willing, which can’t therefore belong to the intrinsic properties of the will. Consciousness is therefore not an intrinsic property of the will; in itself the will is ‘blind.’ (WWR1, SW2:135, and many other places) So Schopenhauer is not a panpsychist.15 Nevertheless, there are structural argumentative similarities to panpsychism that shed light on and make Schopenhauer’s ‘analogy’ much more plausible. Strawson agrees that ‘physics can’t characterize the intrinsic nonstructural nature of concrete reality in any respect at all.’16 And he endorses Russell’s claim that ‘we know nothing about the intrinsic quality of physical events except when these are mental events that we directly experience.’17 He goes on to define panpsychism as extending ‘the claim that everything is energy by saying “intrinsic nature of this energy is experientiality.”’18 Strawson instructively shifts the burden of proof for this generalization, arguing that we are reluctant to perform it only because we have a mistaken ‘picture’ that distorts our weighing of the evidence: ‘we tend to revert to a conviction that we have a basic grasp on things that allows us to be sure that the matter/energy whose spatiotemporal manifestations are all around us couldn’t literally be nothing but experientiality.’19 In other words, we tend not to have (appropriate) Humility,20 and think that science and outer experience in general in fact give the basic i.e. intrinsic nature of things. But outer experience (science) tells us exactly nothing about intrinsic properties, most especially not that there aren’t any; and so ‘we have … no idea of the intrinsic nonstructural stuff nature of the physical insofar as the physical is something other than the experiential.’21 Although Schopenhauer identifies a different intrinsic property, it is the same distorting picture that makes the generalization of that property seem implausible. 4 Conclusion Schopenhauer is anxious that our experience of objects leaves us alienated from their intrinsic properties because experience gives us access only to their extrinsic properties. Schopenhauer’s view that it is ultimately only external experience that is so limited is challenging, especially within a Kantian framework, but original in its analysis of the problem and description of inner experience of willing as an intrinsic property. His further generalization of this view can be illuminated by structural connections with panpsychism. 15 Skrbina, Panpsychism, argues the contrary, but his citations are misleading. For instance, on p. 118 he quotes (using an older translation) the following passage from Schopenhauer: ‘the force which attracts a stone to the ground is . . . in itself . . . will.’ But the context for this quotation is the passage at WWR1, SW2:126 where Schopenhauer denies precisely that this means that the stone is consciously willing! 16 Strawson, Mind and Being, 85. 17 ‘Mind and Matter’ in Russell Portraits from Memory (Nottingham. Spokesman 1956/1995), 153. Cited in Strawson, Mind and Being, 97. 18 Strawson, Mind and Being, 94. 19 Strawson, Mind and Being, 96. 20 Humility says was know nothing about intrinsic properties; appropriate (Schopenhauerian) Humility says we know nothing except in the case of inner experience of our own body. 21 Strawson, Mind and Being, 100. References A/B Immanuel Kant, Critique of Pure Reason. Cambridge. Translated and edited by Paul Guyer and Allen W. Wood. Cambridge: Cambridge University Press 1998. SW 1-7 Schopenhauers Sämtliche Werke, ed. Arthur Hübscher (Mannheim: Brockhaus, 1988), vols. 1-7. WWR1 The World as Will and Representation [Die Welt als Wille und Vorstellung], vol. I (1818/1844/1859). Trans. and ed. Christopher Janaway, Judith Norman and Alistair Welchman. Cambridge: Cambridge University Press, 2014. WWR2 The World as Will and Representation [Die Welt als Wille und Vorstellung], vol. II (1844/1859). Trans. E. F. J. Payne. New York: Dover, 1966. FW Prize Essay on the Freedom of the Will [Über die Freiheit des Willens] (1839). Trans. Christopher Janaway. In The Two Fundamental Problems of Ethics (1841/1860), 31-112. Cambridge: Cambridge University Press, 2009.