This makes his behaviour so conspicuously different from that of an animal that one sees straight off how, as it were, fine, invisible threads (motives that consist of mere thoughts) steer his movements, while those of the animal are pulled by the coarse, visible ropes of what is present in intuition.
Arthur Schopenhauer, ‘Prize Essay on Freedom of the Will’
It is a commonplace that Schopenhauer holds a view of the relationship between will and rational intellect,1 between our volitional-desiderative and our rational-reflective natures, according to which the latter is the former’s obedient servant. ‘The will’, he tells us, ‘is first and primordial; cognition only comes in later, since it belongs to the appearance of the will as its instrument’ (W 1, 319). Famously, he cashes out this relationship between our volitional and intellectual natures metaphorically, as one ‘of the strong blind man [will] carrying the sighted lame man [rational intellect] on his shoulders’ (W 2, 209). In matters of human conduct, the following remark is perhaps the starkest statement of the will’s primacy over the intellect: ‘Actually the foundation and propaedeutic of all knowledge of mankind is the conviction that human behavior on the whole and in essence is not guided by his reason and its designs’ (P 2, 209).
It would appear, then, that no reading of Schopenhauer’s view of practical agency can be fully adequate without taking the ‘primacy of the will’ as some kind of exegetical constraint. For some commentators, abiding by this constraint makes it appropriate to read Schopenhauer along broadly ‘Humean’ lines, where that involves likening him to either the historical Hume or to contemporary Humeans about motivation and practical reason (or both).2 For example, Julian Young says that Schopenhauer’s view of practical reason ‘is just Hume’s: reason has no role in the determination of ends but is entirely concerned with the calculation of means’ (Young 1987: 17). We find comparable attributions when it comes to Schopenhauer’s theory of motivation. Patrick Hassan contends that Schopenhauer subscribes to ‘The Humean Theory of Motivation’, the claim that ‘belief is not sufficient for motivation, which always requires, in addition to belief, the presence of a desire or conative state’ (Hassan 2019: 5). Bernard Reginster (2006) says something similar in his treatment of Schopenhauer’s discussion of practical agency. On Reginster’s reading, Schopenhauer holds a view of motivation according to which motives ‘move us in accordance with the distinctive inclinations that form our “character”’ (Reginster 2006: 74). Reginster elaborates with an example: ‘The distress of others is a motive for me, for example, that is to say, it affects me in a particular way, because compassion is part of my character’ (Reginster 2006: 74). In other words, if a standing desire to relieve other people’s suffering were absent from my character, then I’d have no motivation to do so when confronted with their plight. But, if I am so motivated, then the natural explanation why is my standing antecedent desire (which is expressive of my character), together with some instrumental beliefs. And that’s the Humean view.
There is, however, something initially puzzling about saddling Schopenhauer with Humeanism. Schopenhauer takes motivation (in both humans and animals) to be centrally a cognitive capacity. In his earliest writings, he says that ‘the medium of motives is cognition: consequently the receptivity for motives requires an intellect’ (G, 49). Call this general thought the cognitive mediation claim.3 Additionally, consider the following passage from the Appendix to the main work where he discusses our capacity for ‘reflectiveness’, that is, Besonnenheit, something he calls a ‘gift of reason’:
It is this that allows us to look forward and backward and to get a comprehensive view of our lives and the ways of the world; it allows us to be independent of the present, to approach things deliberately, thoughtfully and systematically, to do evil as well as good. But what we do, we do with absolute self-consciousness. We know exactly what our will decides, what it will choose each time, and what other choice was possible under the circumstances; we come to know ourselves through this self-conscious willing and we reflect ourselves in our deed. (W 1, 548)
It should come as no surprise that Schopenhauer gives Besonnenheit (reflectiveness; ‘thoughtful awareness’) a prominent place in his discussions of rational agency, given that elsewhere in his philosophy he attaches great value to the kind of cognitive awareness that this kind of reflective distance seeds. Here, I have in mind his remarks on the nature and value of aesthetic and ascetic experience, each of which is a kind of experience that brings us important epistemic insights: the former into the Platonic Ideas, the latter into the alleged value of negation of the will. Naturally, the main difference between aesthetic experience and practical experience is that the latter is essentially will-involved whereas the former is essentially will-less. We can guess that it is precisely this structural difference that leads one to view practical experience, that is, human conduct, as exemplary of the will’s primacy. But the question remains how the primacy of the will, read along Humean lines, squares with the claim of cognitive mediation and the importance of reflectiveness.
Julian Young provides one answer. Noting passages where Schopenhauer appears to jettison the intellect’s motivational effects on action, Young claims that
all these passages amount to is the Humean truism that reason alone never moves us to act, plus the necessary corrective to Hume’s (Cartesian) assumption that all desires are transparent to their owners. They are completely compatible with the other Humean truism that reason does move us to act. (Young 1987: 60)
And so for Young, there is a Humean solution to the puzzle, which is just to say that reason does move us to act, and so it does have an important role to play in practical experience. However, it does not move us all by itself. Being its own source of motivation is not a part of reason’s practical significance.
While I agree with Young and others that the proper reading of Schopenhauer’s account of our agential psychology is one that takes this productive partnership between will and intellect seriously, I am not convinced that Schopenhauer endorses the ‘Humean truism’. Rather, as I will argue here, there is ample textual evidence suggesting that Schopenhauer allows for cases of motivation by reason alone (i.e., motivation by rational cognition). In particular, I will argue that he allows for cases of motivation by the essentially cognitive capacities of imagination and recognition. These are each kinds of cases where an agent’s motivation to action can be sufficiently explained without any appeal to antecedent desires, for the reason that no such desire lies at the source of the agent’s motivation. In other words, this is non-desiderative motivation. When read alongside the portions of the text where Schopenhauer does appear to allow for motivation by (what are often unconscious or uncognized) conative states, we are pushed in the direction of attributing to him a kind of motivational pluralism, or what I’ll sometimes call a multi-source view of motivation. One of my aims here is to argue that Schopenhauer is best read as subscribing to a view of this sort.
In this respect, Schopenhauer’s theory of motivation adheres to Stephen Darwall’s (1983) non-Humean idea that we ought to take a wider view of our motivational capacities than the Humean allows. ‘We may concede’, according to Darwall, ‘that whether a person can be moved by some consideration is determined by what we might call, in the broadest sense, his motivational capacities as well as by other factors’ (Darwall 1983: 39). But, Darwall continues,
[what] we should not concede […] is that the agent’s current desires function as a filter that determines which considerations can move him and which cannot, for a person’s motivational capacities, in the broadest sense, are not constituted simply by his desires but also by capacities of imagination, sensitivity, and so on. (Darwall 1983: 39)
For Darwall, we should ‘resist […] the [Humean] supposition that a person cannot be moved by some consideration without the existence of some prior desire’ (Darwall 1983: 41–42). It will be my contention here that the right way to read Schopenhauer is to see him as taking a wider view of our motivational capacities. A virtue of this approach is that it traces a strand through Schopenhauer’s account of practical agency that weaves together the remarks on Besonnenheit and the claim of cognitive mediation, thus bringing a certain unity to his view that is not often found in extant interpretations.
If I am right that we should attribute a multi-source view to Schopenhauer, then that is enough to show that he is not a Humean.4 For pluralism about the source of motivation is incompatible with the strong Humean claim that desire must always figure somewhere in the explanatory story about how an agent’s motivation is generated. However, I take showing Schopenhauer to be a kind of motivational pluralist to be a weaker argument against the Humean view. There is, I think, a stronger argument in the neighborhood as well, one that gains some support directly from the cognitive mediation claim, but also from Schopenhauer’s non-causalist metaphysics of action.5 That argument claims that when it comes to understanding rational agency, motivational states and desiderative states must be understood as distinct in kind, meaning that desires are not even of the right psychological kind to motivate or cause rational action. Whether this revisionary picture of motivation is fully tenable is not a matter I claim to settle here. Rather, my aim is simply to call attention to it as a way of garnering interest in an intriguing possibility in philosophical psychology that Schopenhauer’s account may open up.
In the next section I introduce the basics of Schopenhauer’s account of motivation and discuss various possibilities for understanding what a Schopenhauerian motive is and what motivation is like for Schopenhauer. In sections 3 and 4 I dig deeper into motivation’s cognitively mediated quality by examining Schopenhauer’s metaphor of motives as ‘fine, invisible threads’, making use of it to argue that Schopenhauer takes rational motivation to have a constitutively diaphanous character. This leads into a discussion of motivation-by-imagination. Section 5 considers another case of cognitive motivation, what I call motivation-by-recognition. In section 6 I present the stronger argument against Humeanism by way of considering an important objection. Section 7 explores possible connections between the ‘revisionary’ reading of Schopenhauer’s account of motivation sketched in section 6 and his so-called ‘expressivist’ theory of action. Section 8 concludes.
Let’s start with some of the basics of Schopenhauer’s account of motivation, focusing first on the role he assigns to cognition and then on his notion of a motive. Motivation, he says repeatedly, is ‘causality that goes through cognition’ (E, 54).6 Only creatures with an intellect are capable of motivation, since they are the only creatures capable of cognition. Thus, if we find a creature lacking in intellect, as is the case with plants and rocks, then they are not capable of motivation (though they are still causally affected in that they react to natural forces and external stimuli). To have an intellect is, at a minimum, to have the capacity to form representations of one’s immediate environment. Minimally, then, to be capable of motivation is to be capable of being caused to act by representations. But human beings are not the only creatures with the capacity to form representations of the world. Therefore, we are not the only creatures capable of acting on motives. Animals do this too.
At one point, Schopenhauer defines ‘animal’ as ‘that body whose peculiar external movements and alterations in accordance with its nature always follow upon motives, that is, upon certain representations that are present to its already presupposed consciousness’ (E, 54). He then adds: ‘all the movements that the animal completes as an animal, and which precisely for that reason depend on what physiology calls animal functions, happen in consequence of an object cognized and so upon motives’ (E, 54). The species of representational cognition of which animals partake and in which we locate their motivations is what Schopenhauer calls ‘intuitive’ cognition.
The following passage from the main work illustrates well how Schopenhauer thinks about intuitive cognition:
But just as the visible world is there as soon as the sun rises, so too the understanding with its one simple function transforms dull, meaningless sensation into intuition (Anschauung) in one fell swoop. What the eye, the ear, the hand senses is not an intuition: it is merely data. Only when the understanding proceeds from the effect back to the cause is the world present in intuition, spread out in space, its form capable of change, its matter persisting throughout all time […] (W 1, 33)
Departing from Kant’s epistemology, Schopenhauer views the understanding’s main structural role as that of applying the principle of causality to raw sensory data so that the agent may form an intuitive cognition of her worldly environment. And so whereas Schopenhauer agrees with Kant that raw sensory impressions—the deliverances of mere sensibility—are not yet cognitions, he parts company with him in claiming that the epistemic subject can have a genuine cognition of the world without having brought that intuition under a concept. Intuitions without concepts are not blind; rather, assuming conditions are good (no fake barns in view), they are direct empirical perceptions of the world.
Returning to animal motivation, the idea is that animals are motivated by intuitive representations, that is, by empirical perceptions. Importantly, this is the only way that animals can be motivated to act: ‘they have none other than intuitive representations and consequently know only what is directly present, live in the present alone’ (E, 56). Schopenhauer continues: ‘The motives by which their will is moved must, therefore, always be intuitive and present’ (E, 56). Shortly after, he will say that an animal is always subject to ‘the immediate compulsion of objects present in intuition affecting his will as motives’ (E, 57). Taking a very simple example, the owl perched in the tree perceives the vole and is subsequently and immediately compelled to dive in an attempt to strike it. In this case, its motive for acting is its empirical perception of the vole.
While it is sometimes the case that human beings act like owls, it is far more common for our motivation to take a different form. It does so because, unlike animals, we have in addition to the capacity for intuitive cognition the capacity, born of our rational nature, for non-intuitive or abstract cognition. In a word, we are rational. Reason in Schopenhauer is ‘the faculty of concepts’ (W 1, 548). A concept is a ‘non-intuitive, abstract, universal representations’ (E, 56). Hence, one of reason’s primary activities is abstracting concepts from the world of intuitive cognition, designating them with ‘words’, combining them in various ways, and thus generating the human powers of thought and language. The following passage explains the effects rational thought has on motivation:
The human being […] by virtue of his capacity for non-intuitive representations, by means of which he thinks and reflects, has an infinitely wider field of vision [than animals], which embraces what is absent, past, and future: because of this he has a sphere for the influence of motives, and consequently for choice as well, very much greater than that of an animal that is restricted to the narrow present. It is not as a rule what lies before his sensible intuition, what is present in space and time, that determines his doing: rather it is mere thoughts [Gedanken], which he carries around with him everywhere in his head and which make him independent of the impression of the present. (E, 56–57)
And so whereas animal motives are empirical perceptions, human motives are ‘mere thoughts’, that is, abstract representations. Elsewhere, Schopenhauer will make the point by contending that ‘people are generally in control of their reason, which is to say they are thoughtful [besonnen], that is, they make decisions according to well-considered, abstract motives’ (W 1, 327).
Walking by the corner bakery, I perceive the bread being baked. But no matter how enticed I am by the sight of the warm loaves, I don’t go crashing through the storefront, snatching all the bread in sight. At any rate, that is not what I do if I am acting rationally.7 For Schopenhauer defines a rational action as one that is ‘completed exclusively after well deliberated thoughts and so quite independently of the impression of the intuitive present’ (E, 57). Assuming, then, that I am acting rationally in this case, my motive will be something like the thought: ‘some bread would be nice’. Entertaining this ‘mere thought’, I make my way inside to wait in line. Since my concern in this paper is with Schopenhauer’s picture of rational agency, the cases of motivation I’ll have in mind will all be cases of motivation to a rational action in his sense of the term. Taking Schopenhauer literally, what we should say about the present example is that my motive is a thought with some bread-directed content. Again, something like: ‘some bread would be nice’. Call this the literal reading. It is this reading whose details I plan to work out here, as it strikes me that we lack a complete understanding of the full import of this claim about motivation’s cognitively inflected structure.
On its face, the literal reading is unsatisfying because it is badly opaque. Saying that motives are thoughts doesn’t tell us much about what they are. Sure, for something to be a motive to intentional action, the agent has to be thinking about it or ‘cognitively entertaining’ it in some sense. But in what sense of ‘thinking about’ or ‘cognitively entertaining’ it? Part of the sense of thinking about something that Schopenhauer must have in mind is the sense evoked by the idea of mentally representing to oneself some objective state of affairs. Our thoughts represent states of affairs abstractly; whereas our perceptions do this intuitively. So, to say that motives are mere thoughts is to say that they are abstract representations of states of affairs. However, in being abstract, motives have license to not only represent the states of affairs that we directly encounter, but merely possible states of affairs as well; that is, states of affairs that we fabricate in our minds. For instance, I might decide not to enter the bakery, and my motive could be the possible imagined state of affairs where I buy some bread, woof it down, and feel lethargic afterwards. Entertaining that thought, I see that I don’t want that for myself this afternoon, so I don’t go inside.
At this point, then, we can say that a motive is an abstract representation whose content is some actual or possible state of affairs.8 Christopher Janaway, however, citing a remark from W 2, points out that Schopenhauer also takes motives to be purposes (Janaway 1989: 214). In ‘On the Basis of Morals’, Schopenhauer explicitly associates talk of ends or purposes with talk of motives, going as far as to define an end as ‘the direct motive of an act of will’ (E, 160).9 What is it to entertain a purpose? Janaway gives a commonsense reading of this capacity in terms of an agent’s ‘having wants or desires’ (Janaway 1989: 214). Thinking of motives as purposes means we must add another ingredient to the mix of Schopenhauerian motives, one that we might’ve expected to be included all along. For now we see that a Schopenhauerian motive will involve something conative, like a want or desire.
Taking Schopenhauer’s talk of abstract representations and purposes equally seriously, Janaway concludes that ‘the state occupying the place of a Schopenhauerian motive has to be a complex state, in which I (a) have a representation of an object, (b) have a desire concerning some change in the state of that object, and (c) am disposed to try to bring about that change in that object’ (Janaway 1989: 214). At this point, seeing Schopenhauer as a kind of Humean about motivation tempts us. For looking at Janaway’s treatment, we have all the necessary ingredients of the Humean theory of motivation, though somewhat differently arranged.
Consider Michael Smith’s classic statement of the Humean theory: ‘motivation has its source in the presence of a relevant desire and means-end belief’ (Smith 1987: 36). Neil Sinhababu, another defender of the Humean theory, puts one of its central claims—what he calls ‘The Desire-Belief Theory of Action’—this way: ‘Desire is necessary for action, and no mental states other than a desire and a means-end belief are necessary for action’ (Sinhababu 2009: 465). For the Humean, then, motivation involves something conative and something cognitive: a desire and an instrumental belief. Janaway, we saw, argues that Schopenhauer provides us with ingredients of both sorts. One difference, though, is that Schopenhauer rarely talks about the cognitive aspect of motivation in terms of belief; he prefers instead to talk in terms of thoughts. Still, it could be true that in order for a thought, like ‘some bread would be nice’, to become motivating for me, it must be related to a nest of background beliefs. I wouldn’t think about eating a loaf of bread if I didn’t believe, say, that my mouth was not wired shut. We could also say that my being motivated to eat the bread assumes I am going to find a means to do it, that is, a plan involving instrumental beliefs. In that way, I can never really be motivated to act without having some beliefs that are relevant to the possible action.10
Maybe, then, all Schopenhauer is getting at with the ‘mere thoughts’ claim is the Humean idea that motivation requires instrumental beliefs to team up with the agent’s desires. In a bit, I will argue that Schopenhauer means to be doing a lot more with the claim that motives are mere thoughts than saying that motivation requires instrumental belief. Still, for now it is enough that he does want to include some cognitive component in motivation. So that box of the Humean theory is checked off. What about the conative dimension?
One problem that arises here is, in Janaway’s words, Schopenhauer’s ‘strenuous divorcing’ of the subject as willing from the subject as representing (Janaway 1989: 216). As we’ve seen, Schopenhauer insists that motives, as causes that ‘pass through’ cognition, fall squarely on the side of representation: they are cognitive states. Whereas the conative states commonly understood to be relevant to action, things like inclinations, passions, and emotions—what he elsewhere calls ‘the manifestations of willing’—fall on the side of willing (E, 38).11 And so the puzzle is trying to figure out what role conation has, if any, in his view of motivation.
Consider some remarks from W 2 about the various kinds of conative states that are apparently relevant to motivation: ‘Inclination is any strong susceptibility of the will to motives of a certain kind;’ ‘Passion is an inclination so strong, that the motives that excite it exercise a power over the will which is stronger than that of any possible motive acting against them’ (W 2, 592); ‘The emotion is a stirring of the will, just as irresistible yet only temporary, by a motive that does not obtain its power through a deep-rooted inclination’ (W 2, 593). Notice how with each of these descriptions Schopenhauer is careful to keep the conative aspect of our agential psychology that he is describing separate from the idea of a motive. To take one example, he says that motives excite passion, and in saying so seems to rule out that motives are passions. This is initial evidence against saddling Schopenhauer with either a Hume-like or Humean view of motivation; it is textual support that pushes him more in the direction of at least a multi-source view of motivation, since here we see him he driving a wedge, as Janaway says, between conative states and motivational states, that is, between states of willing or desiring and states of representing.
We will come to see that part of what explains this wedge is the overarching non-causal theory of action in which Schopenhauer embeds his account of motivation. In particular, such a theory forces Schopenhauer to adopt a certain (unintuitive) picture of the causal antecedents of action, and of what is going on with agents in the run up to what they do that is at odds with the Humean view. Indeed, this would seem to be what is responsible for the claims we just saw him making about the relationship between things like inclination, passion, and emotion and an agent’s motivation. For those conative states seem to not figure in his picture as motives, and yet they seem important in the explanation of many things that we do. But I want to hold off confronting these issues until after we’ve more fully grasped motivation’s cognitively mediated structure.
I want to make further sense of what I referred to above as the literal reading of Schopenhauerian motives, the one that says they are ‘mere thoughts’. We can start by connecting Schopenhauer’s discussion of motivation with some things he says about another one of our agential capacities: reflective deliberation. Schopenhauer contends that deliberation involves a ‘full-fledged battle between several motives […]’ (W 1, 324). He also believes that the battle of motives, which is a hallmark of human deliberation, tells us something about human motivation itself. For only human beings experience the effects of motives this way, since in order for motives to do battle with each other, it has to be the case that several of them can be entertained simultaneously, and that is only available to beings with a capacity for abstraction. Describing the battle model, Schopenhauer explains that
motives are almost always abstract representations that the spectator does not have access to, and the necessity of their operation is hidden from even the agent himself behind their conflict. This is because it is only abstractly that several representations can lie side by side in consciousness, as judgements and chains of inferences, and operate against each other, free from all temporal determination, until the stronger among them overpowers the rest and determines the will. (W 1, 324)
Schopenhauer labels this ability our ‘deliberative capacity’ (Wahlentscheidung) (W 1, 324). Although both humans and animals have the capacity for motivation, only the latter has the capacity for deliberation. From this we might be tempted to conclude that motivation and deliberation are distinct capacities. And yet, when rational agency is Schopenhauer’s target, he often draws a tight connection between them to the point where deliberation just is rational motivation, and where to be rationally motivated just is to deliberate before acting.
This connection comes out in his claim that all rational actions bear ‘the character of the deliberate and intentional’ (E, 57). Crucially, this need not be read as the claim that agents only act rationally when their actions are the result of conscious, step-by-step deliberation; we do not have to read Schopenhauer as saying that everything we do is something we were conscientiously deliberating about doing. His aim instead is to draw attention to a perceptible quality of rational conduct, one that is explainable in terms of the distinctive motivational capacities human beings bring to bear on their actions. It is on the basis of observing the qualitative difference between human and animal behavior that we arrive at an understanding of what is special about human motivation, namely, its deliberative or reflective quality.
Turning one’s gaze to the qualitative dimension of an agent’s conduct ‘one sees straight off how, as it were, fine, invisible threads (motives that consist of mere thoughts) steer his movements’ (E, 57). How should we understand the metaphorical language employed here? One possibility is that, in saying that we are steered by ‘fine, invisible threads’, Schopenhauer is calling attention to the fact that we are not empty-headed actors. That is, we are thinking beings always wrapped up in thought, meaning that whatever is going to get motivational purchase on us is something that will first have to be transformed into the mode of thought. A central idea here is that this aspect of our rational nature—that we are wrapped up in thought—effects a certain detachment in us from our immediate environment. This is something Schopenhauer is at pains to make clear in his insistence that human motives are ‘not bound to a certain proximity in space and time like merely intuitive motives’ (E, 58); again, he says they are ‘free from all temporal determination’ (W 1, 324). The result is that far from desire being the filter through which motives operate, it is an agent’s rational cognition which does the filtering. The motivation of rational agents occurs ‘through a mediation of concepts and thoughts in a long chain’ (E, 58).
How does this help us refine the literal reading introduced above? In the following way: to say that in a case of rational conduct motives are ‘mere thoughts’ is to say that they are the rational, rather than intuitive, influences on action. On this elaboration of what I am calling the literal reading, Schopenhauerian motives are the rational influences on action, where ‘rational’ here means ‘mediated by a reflective intellect’, rather than, say, ‘responsive to good reasons’. It is this mediated or thoughtful (‘besonnen’) quality of rational motivation that gives our actions ‘the character of the deliberate and intentional’.
Despite arguing for the reflectively mediated quality of human motivation, Schopenhauer remains adamant that ‘this does not in the least remove its causal nature and the necessity posited with it’ (E, 58). And so the question arises: If, by insisting on the mediated quality of human motivation Schopenhauer is not aiming to upend the causal view of motivation, what is he trying to do? As alluded to already, Schopenhauer wants to draw our attention to a special qualitative feature of rational motivation. I was trying to get at this above with the idea that human beings are perpetually ‘wrapped up in thought’, which is a claim that Schopenhauer’s account of motivation presupposes and takes seriously. In this section, I am going to refer to this feature as the diaphanous character of motivation.12 With motivation’s diaphanousness in view, we start to see why Schopenhauer supports taking a wider view of our motivational capacities than the Humean allows.13
Return to the metaphor of motives as ‘fine, invisible threads’. As I see it, it is open to an external and an internal reading. On the external reading, Schopenhauer makes this observation in order to distinguish the ‘look’ of human motivation—its formal observable qualities—from animal motivation, the former being so ‘conspicuously different’ from the latter (E, 57). There is nothing fine or delicate about the external appearance of animal motivation, and certainly nothing invisible. Witnessing the owl dive towards its prey, we not only see the way it behaves, but we quite literally see the motivation for its behavior: namely, its prey. Animal motivation is not fine or delicate, but coarse.
While there are certainly cases where this can happen for human actions, Schopenhauer often says things which suggest it is not so easy to detect externally, from the look of what someone is doing, what their motives are. At least part of the reason why our motives are externally invisible in this way has to do with the kinds of considerations that are capable of motivating us. On this score, notice what Schopenhauer says about the ‘genesis’ of rational motives:
For now they are mere thoughts that the human being carries around inside his head but whose genesis lies outside it, often at a great distance indeed, sometimes in his own experience from past years, sometimes in extraneous transmission through words and writing, even from the most distant times, yet in such a way that their origin is always real and objective […] (E, 61)
We can trace the origins of my motivational proclivities to things like past experience, transmissions from writings (perhaps one’s favorite novels), and even to ‘the most distant times’, an idea Schopenhauer sadly leaves undescribed.
Returning to the example of buying bread at the bakery, we can say that past experiences tell me that this bakery bakes great bread; thus, the origin of my motivation in this case lies in these past experiences. More importantly, my particular motive to go inside today need not make any reference whatsoever to my occurrent desires. Past experience tells me that I enjoy this bakery, and so it has become my go-to spot—I patronize it out of habit. Let’s say it becomes my habit to buy bread on Mondays. I now have a habitual motive for buying bread. In the right context (on Mondays), my motive for going into the bakery thus need not be anything more than the thought that ‘it is Monday, and on Monday I get my bread for the week’. Or just: ‘it’s Monday’. Indeed, given my habit, we can say that I would be going here to get my bread even if I was in no particular mood to do so, for my mood has nothing to do with my motivation to get bread on Mondays. Rather, my motivation originates in past experience and habit. This, then, is motivation-by-past-experience-informed-habit. Maybe a clunky expression, but a familiar motivational phenomenon.
Let’s now consider a different case of cognitive motivation: motivation by imagination. Schopenhauer is clear that we can be motivated by imagination. Sometimes, he says, ‘completely imaginary circumstances can operate like real ones, not only in the case of a single illusion, but even overall and permanently’ (W 1, 322). This fits with a claim we already saw him make, namely, that in matters of motivation rational agents have ‘an infinitely wider field of vision’ than animals, one ‘which embraces what is absent, past, and future’, and because of which their ‘sphere’ of motivational influence is much greater (E, 57). When taken together, these remarks lend plausibility to the idea that human beings can be motivated by imaginings.
Even an everyday action like going to the bakery to buy some bread might be motivated by an act of imagining. This is crucial because it implies that even these kinds of perfectly ordinary actions are such that their motive may not be externally perceivable, meaning they too are the kinds of actions that can be steered by a fine, invisible thread and so expressive of motivation’s diaphanous character. Here, we see that the driving idea behind Schopenhauer’s attributing a diaphanous or fine-grained character to rational motivation is the fact that my motives need not have any strong connection to my occurrent desire or to my present perceptions. This is what he is getting at with his repeated use of metaphors of spatial and temporal distance in his treatment of the issue. And this is where the internal reading of the metaphor of motives as ‘fine, invisible threads’ becomes important.
In saying that imaginary circumstances can operate as my motives, Schopenhauer is saying that as imaginative agents we are in principle capable of being motivated by just about anything we can think of. Although it is my habit to go to the bakery on Mondays, here I find myself motivated to go on a Thursday. Why? Last week, my grandmother died. The bakery is the last place her and I were together. And right now I find myself thinking about her, imagining what it would be like if I were on my way to meet her there now. Playing this game of imagination with myself, I head out the door, walking towards the bakery as if I am going to meet her, even though I know she is dead. In doing so, I am aware that what it motivating me is not one of the ‘real’ motives in my environment, but something I imagine. In that way, I am aware of the diaphanousness of my own motivation.
Plenty of philosophers have made the case that imagination is a motivational force.14 While I don’t plan to explore the ins and outs of the contemporary debate here, there is one aspect of this line of inquiry worth mentioning. In a recent paper, Neil Van Leeuwen discusses how ‘different sorts of imagining […] play critical roles in causing and even (for some sorts of action) constituting actions’ (Van Leeuwen 2016: 287). One of the sorts of imagination that Van Leeuwen touches on is what he calls ‘constructive imagination’. Here’s how he defines it:
The term [‘imagination’] can refer to the capacity to generate novel representations that are not arrived at simply by remembering or by perception of the nearby environment. To disambiguate, I call this capacity constructive imagination. (Van Leeuwen 2016: 295)15
As is clear from the definition, one of the features of constructive imagination is that through this capacity subjects are able to construct representations none of the inputs to which come from their immediate environment. This conception of constructive imagination, then, shares with Schopenhauer’s treatment of the cognitive dimensions of human motivation the idea that the representations that move us can be constructed free of the spatial and temporal prompts found in one’s immediate environment.16
Van Leeuwen goes on to describe two forms that constructive imagination can take. ‘First, constructive imagination generates representations of possible states of the world—ways the world might be—on the basis of which we might choose actions. Second, constructive imagination generates representations of possible actions to take’ (Van Leeuwen 2016: 295). The case of motivation-by-imagination described above falls under the first form, since in it I imaginatively construct a representation of a possible way the world might be were my grandmother to not be dead, namely, as the kind of place where she is waiting for me at the bakery, as always. On the basis of this powerful imagining, I am motivated to act.17 Since this kind of case would seemingly fall within the parameters of Schopenhauer’s discussion, we have some evidence that he supports motivation-by-constructive imagination. Moreover, if we look at the second form of constructive imagination Van Leeuwen talks about, we might think it gives a good characterization of how Schopenhauer conceives of rational motivation in general. For so often he makes it sound as though our motives just are imagined practical possibilities.
Of course it is fair to ask whether there is sufficient evidence that Schopenhauer takes the imagination to have a constructive dimension. For if not, then the points just made will seem to be coming badly out of left field. His discussion of artistic cognition, though, and in particular what he says about genius, can mitigate the force of this objection. Note what he says about the genius’ relationship to the actual objects in their environment:
[Actual] objects are almost always very deficient exemplars of the Idea presented in them: hence the genius needs imagination in order to see in things not what nature actually created, but rather what it was trying unsuccessfully to create […] (W 1, 210)
And shortly after: ‘imagination broadens, as much in quality as in quantity, the genius’s field of vision beyond the objects that are actually presented to him. This is why an uncommonly strong imagination is the companion—in fact the condition—of genius’ (W 1, 210). A genius is someone who sees the Ideas in things, even though nature did not put them there. Schopenhauer’s claim is that doing so requires an act of imagination. And since what imagination brings the genius to see is not anything in the objects themselves, it is plausible that the imagination’s function in this case is a constructive one. Schopenhauer is therefore open to the idea of imagination’s having constructive powers.
A lot of the reason why, from the inside, my motives appear as the diaphanous threads out of which I weave a possible course of action is precisely because of their constructed nature. For even if we allow Schopenhauer the claim that motives must always have their origin in something ‘real and objective’, it can still be true that once they’ve been taken up and mediated by cognition, their motivational power is transformed. In this section I want to explore some additional implications of the rationally constructed and cognitively inflected nature of motivation and its transformative effects.
Given the oversimplified equation Schopenhauer often puts forward to explain action—‘every individual action follows with strict necessity from the effect of a motive on the character’—he should be quite willing to allow that there are going to be certain objective circumstances which, when I find myself in them, will get a grip on me in a way that bypasses my rational cognition (W 1, 138). Quoting from the Latin expression which he introduces in both the main work and the freedom essay, Schopenhauer should admit that sometimes motives do operate according to their ‘real essence’ (esse reale) (W 1, 321). What makes an agent’s motivation in such a case ‘real’, rather than constructed, is simply that her motives will be the promptings of her immediate spatial and temporal environment; that is, she will not have the reflectively mediated motivation that we expect from rational behavior (in Schopenhauer’s sense of the term). In other words, this will be a case of non-rational motivation.18 Rational motivation, however, is different.
To see further reasons why, it is worth bringing the entire Latin expression into view:
To be effective, a motive does not just need to be present, it must be recognized [Erkanntwerden]: according to a very apt Scholastic expression that we have already mentioned once before, ‘the final cause operates not according to its real essence, but rather according to its known essence’. (W 1, 321; see also E, 72)
Here, we have a claim I’ll call Recognition.
Recognition – For a motive to be effective, it must be recognized by the agent.
Given what I argued above, I think we should read Recognition as applying, not to all cases of motivation, but only to cases of rational motivation. I can see this amendment as only being helpful for Schopenhauer, for otherwise there’s not a clear way to see how Recognition and his basic equation of action can co-exist. For clearly he wants to insist that in the case of rational action, motives do not operate according to their ‘real essence;’ and yet, when it comes to how he often describes the interaction between a person’s character and their circumstance, they do appear to operate this way. What I go on to say in this section and the next will therefore abide by this restriction of the scope of Recognition.
Schopenhauer spells out the details of Recognition with the following example:
For the relationship between (for instance) egoism and compassion to emerge in any given person, it is not enough for that person to possess wealth and see others in need; he must also know what wealth can do both for himself and for others; the suffering of others must not only present itself, he must also know what suffering is, as well as enjoyment. (W 1, 321)
Since we want to understand how this example is meant to elucidate Recognition, the first question to ask is, what are the motives in this example? The obvious motive in this case is another agent’s suffering. Thus, the motivational state up for discussion is one with the following content: ‘motivation to alleviate their suffering’. This is a case of compassionate motivation. How, according to what Schopenhauer says here, does an agent end up in that motivational state? How is compassionate motivation generated in someone?19
Schopenhauer says plainly that in order for someone to be so compassionately motivated it is not enough that they merely find themselves in the presence of someone else’s suffering: they must also ‘know what suffering is’. Now there is a way of reading this passage that sees it as consistent with the rough outlines of the Humean view. All Schopenhauer is getting at, the Humean says, is that in order for a compassionate person to be motivated to act from their compassionate disposition, they must be in the right epistemic states. In this case, they must believe their wealth is an effective means to alleviating suffering and they must be aware of—that is, they must recognize—the other agent’s suffering as suffering. On this reading, Recognition is just a claim about the necessity of instrumental beliefs in the generation of an agent’s motivation. What the Humean will add is that just as an agent cannot be properly motivated to act from her compassionate disposition without the right instrumental beliefs, she also cannot be motivated to act purely on the basis of recognizing another’s suffering. She also requires the standing desire (however generalized) that others do not suffer.
Though I’m willing to go along with the idea that the essence of Recognition is a claim to the effect that motivation requires an agent to be in the right sorts of epistemic states, I want to argue that Schopenhauer has something stronger in mind as well. For as I read the passage, it leaves open that there could be cases where an agent’s motivation to act compassionately is generated purely by a cognitive act of recognition, which might then in turn awaken what was up until that point a dormant and so causally inefficacious compassionate desire. Call this motivation-by-recognition, since the recognition does the motivating. To see how Schopenhauer’s example of a wealthy person fits the criteria of such a case, we need to add a couple of details.
Imagine that Schopenhauer’s wealthy person is a teenager who has been raised by selfish, pretentious, Wall Street types, the kind of people who bore us with stories about their hard work and long nights, and who avow that they see no reason to give their money away to those who do not ‘deserve’ it. Imagine further that Schopenhauer’s teenager has lived a fairly insulated life. She was raised to believe that the world is really a good place; she was told that what appears as an instance of unredeemed suffering is really the appearance of someone who hasn’t made the right choices. What I think we should say about such a person is that they’ve been habituated into adopting a certain practically-oriented epistemic outlook on things, one that has augmented what they recognize as motives and reasons for action. Where someone else recognizes the homeless veteran on the street as suffering, our wealthy teenager recognizes them as a victim of bad choices, or maybe bad luck. This outlook on things is practical and epistemic at once because it groups together, as we rightly should, what an agent perceives and believes and what she is motivated to do in light of those beliefs and perceptions.
Now imagine that our Wall Street teenager goes off to college and gets introduced to new friends with a different perspective on financial hardship and homelessness. If we are to say that they now have some motivation to alleviate the suffering of say, homeless veterans, then it becomes a question how that motivation was generated in them. On the stronger reading of Recognition that I am working out here, we can say that the agent recognized something about the world that they were not aware of before, and it is this ‘recognizing things aright’ that generates their motivation. Thus, we have a case of motivation-by-recognition. To be precise, we can call this a case of motivation by right recognition, that is, recognition of what ought to be done.20 I contend that this is a case of motivation which can be made perfectly intelligible without any reference to the antecedent desires of the agent.
It is also a case where adding a new instrumental belief as the key explanatory item for this new action doesn’t adequately capture what is going on. For this Humean move assumes that the agent is already latently motivationally—for the Humean, desideratively—inclined in the direction of alleviating others suffering, it is just that she has never known or never recognized the right way to go about it. So, we have a desire that’s gone unfulfilled. But even if this is all true of the agent, that doesn’t rule out the veracity of stronger reading of the passage that I’ve just offered according to which we can tell a different story about our agent, one where she is motivated by the essentially non-desiderative state of recognizing someone else’s suffering aright. For I do not take Schopenhauer to be saying that desire is not going to show up at all in her story of motivation. Rather, the idea is that according to the story, desire does not show up first.
Again, the moment our agent recognizes things aright might be accompanied by the awakening of a dormant, though now very strongly felt, desire to alleviate other people’s suffering. And this might make it hard to disentangle her cognitive act of recognition from the conative state of now desiring to help. From her perspective, they might seem one and the same. But for us, it is just a question whether the recognitional story gives a good explanation of the agent’s new motivation in this sort of case. If it does, then there are cases where a good explanation of the source of an agent’s motivation need make no reference to antecedent desires (how could it? She didn’t have any!). I’ve argued that this is one such case, and one which Schopenhauer thinks is often a reality.21
In developing his non-Humean account of motivation, Stephen Darwall wields an example that is strikingly similar to the kind of case Schopenhauer is considering in the passage up for discussion. And this likeness, I want to suggest, is further evidence that Schopenhauer allows for non-desiderative motivation. Darwall’s example is of a student, Roberta, who undergoes a sort of moral transformation upon arriving at university. Early in her time at university, Roberta ‘sees a film that vividly presents the plight of textile workers in the southern United States’. Darwall says she is ‘shocked and dismayed’ by what she sees, and shortly after decides with some other students to enact a boycott of one of the companies. Her role will be to distribute flyers at local stores that sell that company’s goods (Darwall 1983: 39–40).
Darwall’s point against the Humean theory of motivation is that if we look at Roberta’s personal biography, there is no prior desire that would explain her decision to join the boycott:
Not only does Roberta’s deciding to join the boycott for the reasons she does (considerations with which she is presented in the film) not require explanation by the presence of desire. In addition, she need not, and likely would not, view the cogency of her reasons as dependent on her actually acquiring the desire to aid in the effort. More likely she thinks, mistakenly or not, that the considerations adduced in the film itself provide reasons for others to join the boycott, whether they are in fact so moved or not. (Darwall 1983: 41)
Darwall contends that if what we are after is a good explanation of the source of Roberta’s motivation to relieve the workers’ suffering, we need look no further than the vivid awareness the film brought to her of this particular moral evil. Like Schopenhauer’s example, then, Darwall’s is an example of motivation-by-recognition.
But, the objection goes, isn’t it far more likely the case that the Wall Street teenager and Roberta, though they may have been lacking the particularized desires up for discussion—the desire to alleviate the suffering of homeless veterans and the desire to alleviate the suffering wrought by poor-working conditions and poverty—still each have the more general desire to help those in need? It’s just that that general desire has never been given the particular focus it gets in these two cases. But that itself does not mean that there were no other time where we could’ve observed Roberta behaving compassionately, for example, that time she came to the aid of her younger sibling when they fell off their bike. After all, it would seem implausible that a person could have no compassionate desires whatsoever, unless we’re talking about a sociopath.
I think there’s a good response to this objection, but I want to be careful in bringing out what that response is really about. As a reminder, we cause enough trouble for the Humean reading if we can show that Schopenhauer subscribes to a multi-source view of motivation, since such a view argues that some motivation is non-desiderative. What is crucial to pluralism, I want to suggest, is that there are (a) multiple sources of motivation that are (b) sufficient in themselves to explain different cases of motivation all the while (c) it’s also being true that quite often motivation is a mixed cognitive-conative bag. The multi-source view need not say something as foolish as that Roberta was for all intents and purposes a young sociopath before seeing this film. And neither does Schopenhauer have to say that our Wall Street teenager was utterly clueless about suffering before acquiring her knew knowledge. Instead, what the examples show is that motivation does not always go through, in the sense of originating in, desire. But that is already enough to show that the Humean view misses the mark, for it insists that essentially cognitive states are never sufficient for motivation. Plainly they are, for we’ve encountered examples of motivation where a good explanation of what is going on need make no appeal to the agent’s desires.
We can actually use a Humean conception of desire to strengthen this point that no such desires were present in the cases of motivation we’re considering. According to Sinhababu, one feature of desire is what he calls ‘The Attention-Direction Aspect’.
The Attention-Direction Aspect: Desiring that D will make agents more likely to focus their attention on things they associate with D than things they do not associate with D (Sinhababu 2009: 469).
For example, my standing desire to read literary biographies is evidenced in the fact that upon entering most bookstores, I quickly scan the aisle signs looking for the section containing them. Interestingly, Darwall himself takes Attention-Directedness as an important aspect of desire. Here’s what he says:
A desire consists not simply in the capacity to be moved by awareness of facts regarding its object. It is both more active and more focused than that. It includes dispositions to think about its object, to inquire into whether there are conditions that enable its realization. (Darwall 1983: 40)
But there were no such dispositions present in Roberta’s case—she simply had no general interest in the world’s suffering before going to college. And so if The Attention-Direction Aspect is a crucial feature of desire, then we can say that Roberta lacked the requisite antecedent desire in this case. Or, more modestly, we can say that she had the desire, but its dormancy made it so that it was motivationally inefficacious.
Sinhababu, responding to Darwall, says that the condition of Roberta’s upbringing made it so that ‘[stimuli] that would activate a latent preexisting desire that others not suffer’ were missing; hence, she was without anything that could bring ‘her desire that others not suffer to the forefront of her mind’ (Sinhababu 2009: 487–88).22 She is, then, very much like Schopenhauer’s wealthy teenager with the insensitive and greedy parents who was just never quite able to make the proper recognitions that would’ve activated her latent compassion. If that’s true, then upon achieving right recognition, Roberta would be speaking truly when she says: ‘I’ve always had a general concern for people, I just didn’t know the workers were suffering’. But not an ounce of her previous behavior would justify that claim. What’s true is that she had a very particular concern for those closest to her and that she was quite insulated. But those are different claims. And what becomes true of her after her episode of right recognition is that she now has a concern for the workers, a concern which may in fact be in keeping with her individual character, but not because she’s always been generally compassionate. No. If anything, it’s because she’s always been the kind of person in whom a certain highly individualized expression of compassion would come to the fore were she to undergo a change in her cognition.
With this in mind, we might take the Wall Street teenager’s and Roberta’s experiences as cases of transformative motivation.23 This is motivation to do something new which one’s previous self had never desired to do, but something the very doing of which will then create new desires that could then possibly serve as the source of future motivations. This is not something I am pulling out of thin air. Schopenhauer’s discussion of a notion he calls ‘the acquired character’ details how changes in an agent’s cognition can effect exactly these kinds of motivational transformations (W 1, 329).24 There would thus seem to be a strong unity between Recognition and the discussion of acquired character which he comes to some pages later in the main work, though it is something I won’t explore further here.
At this point I want to confront one of the more serious problems facing Schopenhauer’s view of motivation as I’ve interpreted it here. The worry is that by taking proper stock of motivation’s cognitively mediated structure we end up badly over-intellectualizing the phenomenon. As we saw Janaway state earlier, the issue here is that by making motivation into a highly cognitively mediated capacity, Schopenhauer has made it primarily all intellect, no desire or volition. ‘This may seem dubious’, Janaway tells us, ‘given that a motive has to include some specification of what the subject wants, and even, as I suggested, some specification of what the subject is disposed to try to do’ (Janaway 1989: 216). To put the worry slightly differently, even if it is true that Schopenhauerian motives are mere thoughts, we can still think that the process of being motivated to act on such thoughts surely requires an attitude that is best classified in the language of purpose, want, or desire, that is, a conative state. This is the worry I’ll be considering here.
D.W. Hamlyn notes that one particularly troublesome effect of this over-intellectualizing problem is that Schopenhauer ends up having to reject the very idea of unconscious motivation. Here’s how Hamlyn puts things:
I am inclined to think that despite what is often said about Schopenhauer’s anticipations of Freud, and despite what he himself said about our ignorance of our real nature and about our tendencies to self-deception, he cannot admit the possibility of unconscious motives. (Hamlyn 1982: 132)25
Hamlyn takes this result to follow straightforwardly from the claim that ‘motivation is causality seen from within’ (G, 137). For that implies, Hamlyn says, that an agent’s motives ‘exist as motives only qua known, and not merely as something knowable because belonging to a knowing subject’ (Hamlyn 1982: 132). Additionally, Hamlyn takes Schopenhauer to be committed to the claim that not only do motives explain action, but they ‘must also be seen by the actor as doing so’ (Hamlyn 1982: 132). It follows from all of this that whatever does not make its way into the agent’s cognitive perspective on her action—whatever she cannot ‘see’ as a motivating ground of her action—cannot be properly thought of as motivating her to act. And therein lies the impossibility of unconscious motives. (This would also seem to imply that desires don’t motivate without being deemed worthy by a reflective consciousness, an idea with Kantian undertones.)
Now if Schopenhauer were to allow a desiderative or volitional element on the side of motivation then he’d have no problem permitting unconscious motivation. He could just say that unconscious motivation is motivation by something essentially non-cognitive, something that did not make its way into the agent’s perspective on her action, and yet which moved her to act anyways. That he appears forced to close-off the possibility of unconscious motivation is especially problematic given his propensity to talk about cases of just that sort of thing, that is, cases where agents are moved to act by the ‘secret decisions’ of their will, decisions from which their ‘intellect remains so much excluded’ (W 2, 209).
For years we can have a desire without admitting it to ourselves or even letting it come to clear consciousness, because the intellect is not to know anything about it, since the good opinion we have of ourselves would inevitably suffer thereby. (W 2, 209–10)
Schopenhauer’s example here is someone who, at the death of a wealthy relative, becomes aware that all this time they were unconsciously desiring their death because they were desiring that some of their wealth would be left to them when they died (W 2, 210). Plausibly, this unconscious desire was the source of similarly unconscious motives as well. For example, the unconscious motive to maintain close ties with the relative in question. Because the story here is such a familiar one, if Schopenhauer has to deny that motivation ever works this way, then that is a mark against the adequacy of his theory. What options are open by way of response?
The first option is to throw up our hands in frustration at finding yet another inconsistency in Schopenhauer’s theory of practical agency.26 He both seems to permit and deny that things outside our conscious awareness move us to act. The best we can do in such a situation is just to determine which aspects of the text he truly puts his weight behind, or which seem truly indispensable to the central claims of his practical philosophy, and then to provide some reasons why the other bits of text don’t represent his considered view. But that itself will involve having to adjudicate among claims competing for the title of ‘central’. For as enticing as it is to say that ‘the primacy of the will’ is Schopenhauerian bedrock, we’ve now seen that there’s a strong case to be made that ‘cognition is the medium of motives’, with all that that entails, is also a cornerstone of his practical philosophy.
A second option was already discussed. Schopenhauer can insist that there can be no such thing as unconscious motivation to rational action, since ‘rational’, as opposed to ‘empirical’ or ‘perceptual’ (i.e., animal) action is defined as a movement of the body that is caused by reflectively cognized motives. Still, what he could allow for is unconscious motivation to non-rational action. Whenever we’re unconsciously motivated, we’re simply not acting rationally; thus, we’re acting more in our capacity as animals than rational agents. This response, however, might be taken as not really getting to the heart of the matter. For one could still wish to argue that rational motivation itself should not be overly-intellectualized. And one way of avoiding that is to allow for unconscious motivation that is not rationality-undermining. Perhaps here is where we press the point that Schopenhauer is best thought of as a motivational pluralist, since this allows us to make sense both of the ‘secret decision’ passage as indicating an instance of desiderative motivation and the forms of non-desiderative motivation previously discussed. Thus, Hamlyn’s worry is perhaps just a product of not seeing that this interpretation is available.
The third option is to stand pat, take Schopenhauer’s view of motivation at its letter (siding with Hamlyn on matter of unconscious motivation), and then try to defend it. On its face, this move looks highly revisionary, for it amounts to the claim that there really is no such thing as unconscious motivation to rational action. Indeed, the revisionary reading says that part of the merit of Schopenhauer’s account lies in debunking the philosophical-psychological myth that there is such a thing. In other words, of the many action-related or action-inclining psychological states, motivation is one for which consciousness is an indispensable part. This is not to deny that there are other states which tend towards action for which consciousness is not a necessary component. For example, there can be unconscious desires (just as Schopenhauer allows in the passage just quoted).27 But then this tells us something about how he conceives of desire, namely, as the kind of state that can be unconscious. And that is just what makes them ill-fit to be motivational states. Going revisionary, then, we see that Schopenhauer’s contribution in this area lies in the way he disentangles motivational states from desiderative states. Or, perhaps we should say the way he disentangles motivational states from some desiderative states: the unconscious ones. To see that there’s something to this, let’s think a little more carefully about the relationship between unconscious desire and motivation.
Consider what it means to say that an agent has an unconscious desire for X. It means that she has some pro-attitude towards X that is below the level of her conscious awareness. Our greedy agent unconsciously wants her close relative to die because she unconsciously wants her wealth. To see that she really has this unconscious desire, we can add in a version of Sinhababu’s claim about the Attention-Direction Aspect of desire. Assume attention can be unconsciously directed. Then, we can say that her attention is unconsciously directed by her unconscious desire at features of her relative’s condition and environment that are relevant to the desire’s potential satisfaction.28 For instance, during her visits, she might unconsciously attend with some avariciousness to her relative’s art collection. Additionally, her ears might perk up whenever she hears her mother speaking about her relative on the phone in hope that she might be receiving some bad (good) news.
All of this can be true of her. Is any of it evidence that she has any motivation whatsoever to see her relative die, or even to actively seek out their wealth whenever that day finally comes? If not, then her unconscious desires have no effect on what she will eventually be motivated to do in this context. Moreover, they don’t even gain that effect when paired with an instrumental belief. For all we know, our agent might consciously wonder about how she would go about acquiring her relative’s wealth when she dies. And yet, entertaining such ‘means-end’ thoughts need not activate this latent desire. Moreover, as was the case with Roberta, if she does become motivated to inquire into her relative’s fortune, we can explain that in terms of her having come to recognize an opportunity: her relative’s death. Or, it could be explained in terms of her cognitively entertaining such thoughts. But then these different sorts of cognitions are her motive, not the desire. Sure, entertaining such thoughts might awaken her unconscious desire, and it could be that very process that causes her ‘good opinion’ of herself to suffer a blow. Now that she’s motivated, she might realize she harbored this desire all along, and that can deal a blow to her self-image. But notice where desire figures in her story. It comes after her motivation to inquire into her relative’s estate has already been generated in her, just like Roberta’s desire to help the suffering workers comes after she’s already been motivated to help by seeing the film.
It is by virtue of motivation’s cognitively mediated structure that Schopenhauer sees every instance of my being rationally moved to act is one that is inflected by thoughtfulness. There might be other cases of ‘being moved to act’, but if that movement to action comes without cognitive inflection, then, according to Schopenhauer the revisionist, that is not motivation to a rational action, and so it is not informative about rational agency as such.
Having worked out the details of motivation’s cognitively mediated structure, I want to wrap things up with a brief examination of how Schopenhauer’s account of motivation fits within his larger theory of action. For, as I noted earlier, the wedge Schopenhauer appears to drive between willing and representing, and which becomes apparent once we work out the details of the claim that motives are ‘mere thoughts’, might best be explained in terms of commitments we find on display in his broader metaphysics of action.
As scholars have pointed out, Schopenhauer’s theory of action is itself fundamentally non-Humean or non-causalist.29 According to the causal theory of action, bodily movements are caused by some type of mental act, whether it be a volition, some other kind of purely mental willing, or perhaps a belief-desire pair. For the causal theorist, then, the key claim is that the relation between willing and acting is a causal one.30 Now we’ve already seen that Schopenhauer holds that motives are the causes of action, not willings. But if the will is not to be found in the causal antecedents of action, then where is it to be found?
Schopenhauer’s answer to this question is that the will is found in the identical place where the bodily actions is found, namely, in the publically observable bodily movement. In short, willing and acting are identical:
An act of the will and an act of the body are not two different states cognized objectively, linked together in a causal chain, they do not stand in a relation of cause and effect; they are one and the same thing, only given in two entirely different ways: in one case immediately and in the other case to the understanding in intuition. (W 1, 124–25)
The central idea is that the act of will and the subsequent bodily movement is the same phenomenon cognized in two different ways: once from within, once from without. And where there is identity there cannot be causality. Thus, we see that the basic non-causalist contention that the relation between will and action is non-causal finds a home in Schopenhauer.
Charles Taylor nicely summarizes the non-causalist’s central point: ‘Actions are in a sense inhabited by the purposes which direct them, so that action and purpose are ontologically inseparable’ (Taylor 1985: 78). What any proponent of the non-causal theory must do, then, is find a way to cash out the relation of ‘ontological inseparability’ in such a way that the non-causal connection between action and purpose is clearly delineated (for there has to be some connection, even if it is not a causal one). For Taylor, the connection or relation in question is ‘constitutive’; others call it an ‘expressive’ relation (Ridley 2018); and some use the language of embodiment, saying that actions are embodiments of the will.31
Turning back to Schopenhauer, Severin Schroeder offers a helpful characterization of the ontological inseparability claim as it appears in his account of action. According to Schroeder, Schopenhauer sees ‘a conceptual interdependence among a creature’s beliefs, desires, capacities, and actions’ (Schroeder 2020: 169). For Schroeder, the relation between will and action is a conceptual one.32 If you hold to this conceptual view, then you will argue that all explanations of action, whether in terms of beliefs, desires, capacities, emotions, and whatever else, are in a sense elliptical in that though you need only appeal to one item in the mental life of the agent to explain what she did, in appealing to just that one item you elliptically imply the rest, for that is just what conceptual interdependence amounts to.
How Schopenhauer’s non-causal metaphysics of action bears on his theory of motivation is not immediately clear. But one answer suggests itself. We can say that just as an agent’s will is only evidenced (expressed, constituted) in her action, so too is it the case that the only evidence we have for what an agent is motivated to do is whatever they actually do, that is, whatever publically observable bodily movement they execute. Taking an earlier example, I might think that I’d like some bread, and that ‘mere thought’ could be a possible motive for me to act. But it won’t be until I find myself at the counter ordering a loaf that I know that I was in fact motivated to get some bread. And here we see that what potentially unifies the identity theory of action and Schopenhauer’s view of motivation is a certain picture of the epistemic position of agents prior to action.
The main thought here is that our primary way of relating to ourselves, whether in the form of action or motivation, is expressive. That is, we only obtain accurate self-knowledge via the expressive medium that is our conduct.33 Until we act, we cannot be sure of the state of our will nor of the true nature of our motivation. In other words, the identity theory of action and the cognitively mediated structure of motivation share the idea that the agential states preceding action are not of the right kind to make it such that the agent can have what we might think of as final volitional foreknowledge, that is, knowledge of what she wills or desires to do (or who she fundamentally is) prior to actually acting.34
Part of the reason Schopenhauer makes this claim has to do with his peculiar conception of desire, as well his general conception of the antecedents of action. Here’s a representative passage:
For as long as it [action] is in process of becoming it is called a wish, when ready, a decision; but its being this is proven to self-consciousness only by the deed: for until that it is alterable. (E, 42)
The German for what Janaway translates here as ‘wish’ is Wunsch. In their translation of the main work, Norman et. al. translate Wunsch as ‘desire’, as in the following statement:
The desire is simply the necessary consequence of the present impression, whether it is the impression of an external stimulus or of a transient inner mood; for this reason it is just as immediately necessary and lacking in deliberation as the act of an animal: that is why desire only expresses the character of the species, as in animals, and not the individual, i.e. it merely signifies what human beings in general, not the individual who experiences this desire, would be able to do. (W 1, 326)
What both passages reveal is that Schopenhauer conceives of the earliest stage in the development of action as a state of wishing, where, if that is to be likened to desiring at all, it can only be a desiring in a thoroughly weakened sense.
One of the reasons we sometimes wish for X, rather than desire X, is that we tacitly acknowledge there’s not much we are going to be able to do to bring X about, and so we just wait to see whether X will happen for us. Schopenhauer is saying something similar about a certain stage in the motivational development of an action. The thought is that, early on, we may as well just be waiting to see whether what we want to do is really what we want to do.35 We’re fickle creatures, after all. The moment of action could come, and we might suddenly be utterly ambivalent about what we originally wished to do, and then do something else. But if this is how Schopenhauer conceives of the earliest stage of action, then of course he is going to part with the Humean, since what they want to locate at the earliest stage in the development of action is a state that is far more purposively directed. Having the desire for X, for the Humean, sets the whole process of motivation to X in motion, and even serves as evidence for what the agent wants or will do prior to acting. Schopenhauer rejects that picture. My Wunsch tells me very little about what I want or what I will do, and so counts as a rather flimsy source of evidence of my motivation.
This raises the question of how Schopenhauer is able to get desire back into the picture, that is, how he is able to place it in the antecedents of action in the way the multi-source view of motivation requires. Recall, however, that the argument for the multi-source view and the argument for Schopenhauer’s revisionary view are separate arguments, picking out distinct strands running through his philosophical psychology. I think there’s still more that needs to be worked out on the side of the revisionary view; in particular, we should like to know more about its ‘predictive’ models of motivation and deliberation and its concept of ‘wishing’. Still, since it is part of that view’s ambition to exclude desire from the antecedents of rational action, it could be that the answer to the question posed here of how desire gets back into the picture is simply that it does not, and that is what makes the view revisionary.
By taking a closer look at the intellectualist strand in Schopenhauer’s view of motivation and emphasizing the roles played by reflection, imagination, and recognition, we’ve positioned ourselves to see various aspects of his view that either get overlooked or are thought to be unremarkable. The resulting argument is that Schopenhauer takes a wider view of our motivational capacities than tends to be thought. In particular, he takes seriously the datum that we are thoughtful, imaginative agents, capable of constructing our own motivations quite literally out of thin air. I’ve argued that this pushes him in the direction of a kind of motivational pluralism, one that gives good explanations of commonsense examples (Roberta, the Wall Street teenager), and of the phenomenon I called transformative motivation. This interpretation rescues Schopenhauer from a certain misreading arising from viewing his theory of practical agency exclusively through the lens of the primacy of the will. This commits the error of missing how Schopenhauer centers Besonnenheit—‘reflectiveness’; ‘thoughtful awareness’—in his practical philosophy in much the same way that he does in his account of aesthetic and ascetic experience, though admittedly with a much different end in view. Finally, I argued that there is possibly a stronger, revisionary argument against the Humean view lurking here, one whose details I’ve only begun to sketch, but which I think merits a deeper probing.
1I add the qualifier ‘rational’ here because Schopenhauer believes that non-human animals also have an intellect as well as various rudimentary capacities for intellectual cognition, and so ‘reason’ and ‘intellect’ are not in fact interchangeable terms in his philosophy; nor is every form of cognition in Schopenhauer necessarily a form of rational cognition. Still, for the sake trying to keep things tidy, and because my concern here are the elements of his account of rational agency, when I use the terms ‘intellect’ or ‘cognition’ I mean their rational forms.
2For defenses of contemporary Humeanism see Smith (1987; 1998) and Sinhababu (2009). For good critical discussions see Wedgwood (2002) and Mason (2005). For objections to Humeanism, see Darwall (1983) and May (2013).
4There might be room here to interpret Schopenhauer’s discussion of non-desiderative forms of motivation as being in line with the historical Hume’s ‘calm passions’, and so in that way his view of motivation could turn out to be more Hume-like than Humean.
5For sustained treatments of Schopenhauer’s metaphysics of action that are at pains to work out the precise ways in which his view departs from Hume’s as well as other causal theories of actions, see Janaway (1989) and Schroeder (2020). For a more cursory treatment, one that is merely preparatory for a discussion of Nietzsche’s philosophy of action, see Ridley (2018). For a worked out explanation of the non-causal theory which mentions Schopenhauer as one of its historical proponents, see the chapter ‘Hegel’s Philosophy of Mind’ in Taylor (1985).
6Elsewhere in the prize essay he will say that ‘motives have their effect with necessity: but they have to pass through cognition, which is the medium of motives’ (E, 71); and in the revised dissertation, he says that ‘motivation is causality seen from within’ (G, 137).
9Janaway rightfully notes that we should be skeptical about an account of motivation that equates the purpose of an action with ‘the plain representation of an objective state of affairs’. After all, it is one thing for me to be thinking about the bread I see through the bakery window; it is yet another for me to make it my end to go inside and buy a loaf.
10Schroeder (2020) makes a point along these lines that I’ll discuss in more detail later.
11To be clear, when I say these conative states fall on the side of willing I just mean they refer to our volitional-desiderative nature. I don’t mean that they are equivalent with a volition that would be the cause of my action, since as noted earlier, Schopenhauer does not subscribe to such a causal-volitional theory of action.
13It is worth noting Schopenhauer’s non-reductive ambitions, since some Humeans will allow that there are cases where one appears to be motivated in a way other than the Humean describes, and, indeed, cases where the best explanation of an action would seem to bypass the Humean story, and yet, when we look more closely, we see that the Humean story is still in play. See, e.g., Smith (1998).
15Van Leeuwen (2013) cites a variety of studies in empirical psychology attesting to the legitimacy of this cognitive phenomenon and its effects on motivation and action.
16One difference between Van Leeuwen and Schopenhauer, though, is that we saw the latter insisting that the abstract representations that can be my motives must always originate from something ‘real and objective’. By contrast, Van Leeuwen argues that we need not accept this ‘empiricist’ view of constructive imagination. See Van Leeuwen (2013: 224–26).
17Van Leeuwen might say that this is an instance of what he calls ‘propositional imagining,’ rather than constructive imagining. In the former case of imagining, the agent still relies on factual beliefs about her environment, where these serve as her ‘cognitive map’. This is different from a case of constructive imagination where no such factual beliefs need be relied on. Van Leeuwen (2016) argues that both kinds of imagination can motivate action.
19One answer that is not appropriate here is: ‘they are either born with a disposition towards compassion or not’. For though that might seem to follow from certain remarks Schopenhauer makes about the unalterability of character—it’s so called ‘inbornness’—he also emphasizes that each of one of us is susceptible to the three basic types of incentives to action: egoistical, compassionate, and malicious. Thus, it is open to anyone to find themselves in this kind of moral motivational state, provided certain conditions are met. See Cartwright (1988).
20There’s space to argue that Recognition plays a central role in Schopenhauer’s account of moral motivation, for what motivates a compassionate person to act compassionately is recognition of another person’s suffering. And in cases of moral motivation of an even higher sort, what motivates is recognition of the non-distinctness of persons; that is, recognition of a deep metaphysical truth. See, e.g., E, 200–203, 249. One tension here, though, is that such moral recognition is meant to be intuitive, rather than rational or abstract.
21As an additional example, we might take Schopenhauer’s remark that the American penitentiary system reforms the agent’s head, not her heart (E, 71). For, if after being reintegrated into society, the prisoner ceases acting in the ways that got them sent to prison, then this will be some evidence of a change in motivation brought on by purely intellectual means. Though it could, admittedly, just be that these are old motivational energies that have been redirected. But notice what we’ll cite in explaining the redirection: recognition. So their new motivation still seems to be a case of cognitive motivation.
22For another style of objection to Darwall, one that worries that motivation without antecedent desire is altogether too passive, and so doesn’t give us a good account of the voluntariness of rational action, see Finlay (2007).
24A further elaboration of this point would draw even more heavily on things Schopenhauer says when discussing the acquired character, something I do not have time to get into here. Suffice it to say that proper placement of this element of his practical philosophy within the whole turns out to upend quite a bit of scholarly orthodoxy, especially as concerns his claims about the fixity of character and the alleged impossibility of a kind of motivational self-transformation. Though for discussion of the issues which do raise genuine questions see, e.g., Atwell (1990) and Cooper (1998).
25See also Hamlyn (1980: 88).
26For a treatment of other frustrating elements, including his doctrine of character and his discussion of remorse (Reue), see Atwell (1990).
28Schopenhauer’s discussion of the practical effects of drives seems to suggest that attention is the kind of thing that can be unconsciously directed. See “The Metaphysics of Sexual Love” in W 2 for various remarks on his concept of drive.
30Schroeder cites a remark Hume makes in the Appendix to the Treatise as evidence that Hume himself holds a volitional account of voluntary action. Whether or not this truly captures Hume’s view is not important for my purposes here, since the point is that Schopenhauer rejects views of this kind whether they are Hume’s or not.
31For example, Janaway discusses the relationship between will and action in terms of the latter’s embodying the former. See Janaway (1989: 226).
32Schroeder understands Schopenhauer as adhering to the following logical schema for action, which he says is ‘analytic’: ‘If a person A has an all-in desire to ϕ if p, and knows that he is able to ϕ and in no way prevented from ϕ-ing, then perceiving that p will make A ϕ’ (Schroeder 2020: 169). As far as I can tell, this is Schroder’s own spin on the so-called ‘Logical Connection Argument’. For an explication and defense of a revised version of that argument, see Otten (1977). For a refutation, see Lee (1981).
33This connects with Schroeder’s claim that there is a conceptual interdependence among an agent’s beliefs, desires, and actions. To say that action is ‘expressive’ of motivation is to say that what an agent does and how she is motivated to act cannot really be disentangled from each other. This is why those who favor some kind of ‘expressivist’ view of action are likewise fond of the so-called Logical Connection Argument.
34Something like this is also supported by the idea that Schopenhauer is a modal actualist, i.e., he believes that only the actual exists. See Atwell (1990).
35As Patrick Gardiner (1963) notes, one potential problem here is that Schopenhauer ends up making the whole process of motivation look much more like the rational process of self-prediction or self-forecasting than the desiderative one of trying to make the world become as one wants it to be. See Gardiner (1963: 164). One also wonders whether there is a fruitful overlap between Schopenhauer’s view and Velleman’s model of intentions as ‘self-predictions’. See Velleman (1989).
This is a list of abbreviations of works by Schopenhauer that I use in the paper. I use the abbreviations designated by the Schopenhauer-Gesellschaft. I lack access to the now standard German edition of the works of Schopenhauer, that is, Sämtliche Werke, Arthur Hübscher ed., 7 vols., 3rd ed. (Wiesbaden: F.A. Brockhaus, 1972); 4th ed., 1988. Because of this limitation, I have been forced to go far back in time and consult for the German, Sämmtliche Werke, Julius Frauenstädt, ed., 6 vols., 2nd ed. (Leipzig: F.A. Brockhaus, 1891).
|G||On the Fourfold Root of the Principle of Sufficient Reason and Other Writings. Translated by David E. Cartwright, Edward E. Erdmann, and Christopher Janaway. Cambridge: Cambridge University Press, 2012.|
|E||The Two Fundamental Problems of Ethics. Translated and edited by Christopher Janaway. Cambridge: Cambridge University Press, 2009.|
|P 2||Parerga and Paralipomena, vol. 2. Translated and edited by Adrian Del Caro and Christopher Janaway. Cambridge: Cambridge University Press, 2015.|
|W 1||The World as Will and Representation, vol. 1. Translated and edited by Judith Norman, Alistair Welchman, and Christopher Janaway. Cambridge: Cambridge University Press, 2010.|
|W 2||The World as Will and Representation, vol. 2. Translated by E.F.J. Payne. New York: Dover Publications, 1958.|
I want to thank Stephen Finlay, David Fisher, John Grey, Fred Schmitt, Sandy Shapshay, Ivan Verano, and Allen Wood for their willingness to discuss various aspects of this paper with me, regardless of the stage my ideas were at. I’d also like to thank two anonymous referees for providing plentiful and helpful feedback on an earlier version of this paper.
The author has no competing interests to declare.
Cooper, David E. 1998. “Self and Morality in Schopenhauer and Nietzsche”. In Willing and Nothingness: Schopenhauer as Nietzsche’s Educator, edited by Christopher Janaway, 196–216. Oxford: Oxford University Press.
Currie, Gregory. 2002. “Imagination as Motivation”. Proceedings of the Aristotelian Society 102: 201–16. DOI: https://doi.org/10.1111/j.0066-7372.2003.00050.x
Hamlyn, D. W. 1982. “Schopenhauer on Action and Will”. In Idealism Past and Present: Royal Institute of Philosophy Lecture Series, edited by Godfrey Vesey, 13: 127–40. Cambridge: Cambridge University Press. DOI: https://doi.org/10.1017/S0957042X00001577
Hassan, Patrick. 2019. “Schopenhauerian Virtue Ethics”. Inquiry: An Interdisciplinary Journal of Philosophy 65 1–33. DOI: https://doi.org/10.1080/0020174X.2019.1629337
Koßler, Matthias. 2009a. “Life is but a Mirror: On the Connection between Ethics, Metaphysics, and Character in Schopenhauer”. In Better Consciousness: Schopenhauer’s Philosophy of Value, edited by Alex Neill and Christopher Janaway, 77–97. Malden, MA: Wiley-Blackwell.
Lee, Jig-Chen. 1981. “An Objection to the Revision of the Logical Connection Argument”. Canadian Journal of Philosophy 11: 725–33. DOI: https://doi.org/10.1080/00455091.1981.10716333
Mason, Michelle. 2005. “Hume and Humeans on Practical Reason”. Hume Studies 31: 347–78. https://www.muse.jhu.edu/article/383291.
May, Joshua. 2013. “Because I Believe it’s the Right Thing to Do”. Ethical Theory and Moral Practice 16: 791–808. DOI: https://doi.org/10.1007/s10677-012-9394-z
Otten, James. 1977. “Reviving the Logical Connection Argument”. Canadian Journal of Philosophy 7: 725–43. DOI: https://doi.org/10.1080/00455091.1977.10716192
Reginster, Bernard. 2006. The Affirmation of Life: Nietzsche on Overcoming Nihilism. Cambridge, MA: Harvard University Press. DOI: https://doi.org/10.4159/9780674042643
Ridley, Aaron. 2018. The Deed is Everything: Nietzsche on Will and Action. Oxford: Oxford University Press. DOI: https://doi.org/10.1093/oso/9780198825449.001.0001
Schroeder, Severin. 2020. “Schopenhauer and Hume on Will and Causation”. In The Oxford Handbook of Schopenhauer, edited by Robert L. Wicks, 161–74. Oxford: Oxford University Press. DOI: https://doi.org/10.1093/oxfordhb/9780190660055.013.9
Shapshay, Sandra. 2019. Reconstructing Schopenhauer’s Ethics: Hope, Compassion, and Animal Welfare. Oxford: Oxford University Press. DOI: https://doi.org/10.1093/oso/9780190906801.001.0001
Sinhababu, Neil. 2009. “The Humean Theory of Motivation Reformulated and Defended”. Philosophical Review 118: 465–500. DOI: https://doi.org/10.1215/00318108-2009-015
Smith, Michael. 1987. “The Humean Theory of Motivation”. Mind 96: 36–61. DOI: https://doi.org/10.1093/mind/XCVI.381.36
Smith, Michael. 1998. “The Possibility of Philosophy of Action”. In Human Action, Deliberation, and Causation, edited by Jan Bransen and Stefaan E. Cuypers, 17–42. Dordrecht: Springer. DOI: https://doi.org/10.1007/978-94-011-5082-8_2
Taylor, Charles. 1985. Human Agency and Language: Philosophical Papers 1. Cambridge: Cambridge University Press. DOI: https://doi.org/10.1017/CBO9781139173483
Van Leeuwen, Neil. 2013. “The Meanings of ‘Imagine’ Part I: Constructive Imagination”. Philosophy Compass 8220–30. DOI: https://doi.org/10.1111/j.1747-9991.2012.00508.x
Wedgwood, Ralph. 2002. “Practical Reason and Desire”. Australasian Journal of Philosophy 80: 345–58. DOI: https://doi.org/10.1080/713659471
Young, Julian. 1987. Willing and Unwilling: A Study in the Philosophy of Arthur Schopenhauer. Dordrecht: Martinus Nijhoff. DOI: https://doi.org/10.1007/978-94-015-7756-4