publisher colophon

I argue that Hume's characterization of the passions as "original existences" is shaped by his preoccupation with Stoicism, and is not (as most commentators suppose) a ridiculous or trifling remark. My argument has three parts. First, I show that Hume's description of the passions as "original existences" is properly understood as part of his argument against the possibility of passions caused by reason alone (rational passions). Second, I establish that Hume was responding to the Stoics, who claimed that a rational passion is caused by a special type of impression that accurately represents its cause—the divine reason administrating the cosmos—in virtue of resembling it. Third, I argue that Hume rejects the Stoics' claim by appeal to the following feature of his general account of impressions and ideas: for Hume, ideas are exact "copies" or representations of the impressions that caused them, but simple impressions are "originals" that do not represent in virtue of resembling their external or physiological causes. On my reading of Hume, the 'original existence' passage therefore signals his rebuke of a key claim in the Stoics' theory of rational passions.


Hume, Stoicism, emotions, original existences, rational passions


hume's claim that the passions are "original existences" is widely read as targeting moral rationalists such as Samuel Clarke and William Wollaston (T 415).1 In opposition to Clarke's and Wollaston's claim that reason [End Page 609] alone can move us to act, Hume evidently claims that only the passions can motivate us, and he insists that the passions are brute, blind drives or 'original existences' that are immune from rational evaluation. However, that is the extent of the scholarly consensus. Many readers reject Hume's claim in the 'original existences' passage (hereafter, OE), seeing it as a ridiculous denial of the intentionality of our passions; other readers ignore OE, regarding it as a trifling remark that is inconsistent with what Hume says about the passions elsewhere.2

My aim in this paper is twofold.3 First, I argue for an historical claim, namely that Hume's characterization of the passions as "original existences" is shaped by his preoccupation with the Stoics, not with Clarke and Wollaston. Second, I argue for an interpretive claim, suggesting that certain puzzling features of Hume's claim are consistent with his general views on representation and make sense against the Stoic backdrop. A major issue that I do not take up is the ongoing debate about the so-called Humean theory of motivation that OE helped to inspire, which would take me too far afield from Hume's own views. However, my focus on Hume does yield a surprising finding of interest to contemporary discussions of Humeanism, for the outcome of my arguments is that Hume is not as skeptical of the role of reason in our motivating passions as he is usually thought to be.

Here is an overview of the paper. In section 1, I briefly review (what I call) the standard reading of OE, on which Hume is widely understood as claiming that our passions are fundamentally irrational. The high cost of this standard reading is that it makes Hume's view of the passions either obviously wrong or in conflict with his other pronouncements on the passions. In section 2, I argue that both of these interpretive conclusions are mistaken because there are historical aspects of OE to which the standard reading fails to attend. I present a new reading of OE as Hume's response to the Stoics, who held a once influential view of what the rationality of the passions consists in. On the Stoics' view, the passions are rational when the impressions causing them represent the point of view of the divine administrator of the cosmos in virtue of resembling it. In section 3, I draw on Hume's neglected distinction between passions and emotions to show that he designs OE to reject that particular Stoic view, on the grounds that the impressional or emotional causes of the passions cannot represent their causes in virtue of resembling them. In section 4, I conclude that OE does not commit Hume to a general ban on the passions' ability to represent. Hume's claim that reason is the "slave of the passions" initially appears to resist my interpretation, but I suggest that a deflationary reading of that notorious remark is itself unlocked by attending to his preoccupation with the Stoics (T 415). [End Page 610]

1. the standard reading of oe

In this section, I spell out the standard reading of OE and identify some of the philosophical virtues of this view of the passions widely attributed to Hume. I then explain the interpretive debate to which this standard reading gives rise.

1. 1. The Mutual Exclusivity of Reason and the Passions

Hume describes the passions as "original" at least twice. According to the standard reading, on each occasion he is claiming that desiring (or, having a passion) and thinking are mutually exclusive sorts of psychological activities.4 To see why Hume invites such a reading, we need only look at the following passage, in which he appears to claim that the passions, unlike reason, do not distinguish between truth and falsity:

Reason is the discovery of truth or falshood. Truth or falshood consists in an agreement or disagreement either to the real relations of ideas, or to real existence and matter of fact. Whatever, therefore, is not susceptible of this agreement or disagreement, is incapable of being true or false, and can never be an object of our reason. Now 'tis evident our passions, volitions, and actions, are not susceptible of any such agreement or disagreement; being original facts and realities, compleat in themselves, and implying no reference to other passions, volitions, and actions. 'Tis impossible, therefore, they can be pronounced either true or false, and be either contrary or conformable to reason.

(T 458)

Why might Hume have drawn this sharp dividing line between our rational and passionate mental states? The standard reading claims that Hume is identifying a basic fact about human psychology, which is that there are two fundamentally different components of human action. In John McDowell's image, the Humean model of action is quasi-hydraulic.5 Our passions are like the gas pedal in an automobile: they tap into the fuel that powers our thoughts and actions, but are themselves constitutionally indifferent to whether we are driving sensibly. Our reason is like our hands on the steering wheel: Reason aims to see how the world truly is, and steers our motivating passions accordingly.

It is this intuitively appealing picture of human psychology, according to the standard reading, that we find in OE.6 Behind that supposed Humean picture is an argument for the mutual exclusivity of reasoning and desiring. We can reconstruct that Mutual Exclusivity Argument as follows:

  1. (1). Reasoning is checking mental representations of various states of affairs for truth and falsity.7 [End Page 611]

  2. (2). The passions are perceptions ("facts or realities") that do not represent any states of affairs (the passions are "original").

  3. (3). Reasoning and having a passion, therefore, are mutually exclusive; the passions are not correct or incorrect (reasonable or unreasonable), but kindled or dampened.

For present purposes, (2) is the crucial premise. But is it "evident" that the passions are blind drives? Standardly, readers take Hume to provide a defense of that claim in OE:

A passion is an original existence, or, if you will, modification of existence, and contains not any representative quality, which renders it a copy of any other existence or modification. When I am angry, I am actually possest with the passion, and in that emotion have no more a reference to any other object, than when I am thirsty, or sick, or more than five foot high. 'Tis impossible, therefore, that this passion can be oppos'd by, or be contradictory to truth and reason; since this contradiction consists in the disagreement of ideas, consider'd as copies, with those objects, which they represent.

(T 415)

On the standard reading of this passage, Hume is denying, for example, that when I am angry, I am angry about something or at someone. That seems at odds with how we ordinarily speak about our feelings, but there are other cases in which we typically do acknowledge something like the view being attributed to Hume: (i) waking up on the proverbial wrong side of the bed and being grumpy for no reason; (ii) growing depressed and feeling a vague atmosphere of sadness or dread; or (iii) desiring sexual intimacy without having anyone or anything particular in mind. On the standard reading of Hume, he is claiming that all of our passions are like those three cases. In short, if the standard reading of Hume is correct, his claim is that the passions are "original existences" insofar as they are blind feelings. Moreover, if the passions are blind, the standard reading concludes, then Hume has shown that reason and the passions are mutually exclusive—for reason cannot judge the passions for truthfulness when they do not themselves represent how the world is.

1. 2. The Interpretive Debate over OE

Whatever the attractions of Hume's apparent view of human psychology, it comes at a high cost. To see why, consider this: if passions are "original existences" that have no reference to any other objects or states of affairs, then Hume is committed to a view on which our passions are no different from "purely affective physiological states, [such as] shudder[s], shiver[s],… [hot]-flashes," itches and tickles.8 Call this, the Tickles View.

A long-standing interpretive debate has sprung-up in response to the Tickles View of the passions. We have Anti-Ticklers on the one side and Tickler-Deniers on the other. Anti-Ticklers argue that the Tickles View is wildly implausible. If we take Hume at his word that, when we are angry, we are not angry at anyone or about anything, his view is bizarre and mistaken. Indeed, Hume offers no argument for his view, contenting himself with a flat rejection of the passions' intentionality. [End Page 612] However, given that our passions prima facie make some reference to the objects, people, or states of affairs that they are about, Hume's mere assertion to the contrary has little to recommend it. Anti-Ticklers like Robert Fogelin therefore conclude that OE is "perfectly awful" philosophy, and Terence Penelhum rates OE "one of Hume's worst arguments."9

Tickler-Deniers agree that OE seems to commit Hume to the Tickles View, but they deny that he really is committed to that view. If Hume really were committed to the Tickles View, Tickler-Deniers point out, then we should expect him consistently to hold that the passions have no "objects." But there are numerous passages in which Hume says that the passions do have objects and are about things. For example, "As the immediate object of pride and humility is self… so the object of love and hatred is some other person.… This is sufficiently evident from experience. Our love and hatred are always directed to some sensible being external to us" (T 329).10 Tickler-Deniers consequently argue that OE is an infelicitous remark that is "unrepresentative of Hume's claims about passions."11 Indeed, this camp emphasizes that OE disappears from Hume's later Enquiries and Dissertation on the Passions, which perhaps implies that Hume noticed the contradiction and removed the passage as "misrepresent[ing] his own views." As Tickler-Deniers like Annette Baier conclude, we must not allow a "very silly paragraph… [to] perversely dominat[e] the interpretation of [Hume's] moral psychology."12

1. 3. Summary of the Standard Reading of OE

On the standard reading, Hume's OE claim offers us a powerful picture of human psychology, but Hume's argument for that picture has a catch: OE commits us to the Tickles View of the passions, on which the passions are just brute sensations. The standard reading has consequently given rise to a debate that ends in a stalemate between Anti-Ticklers, who think Hume's claim in OE is ridiculous, and Tickler-Deniers, who agree that the passage seems ridiculous but deny that we should take it seriously. In the remainder of the paper, I offer a way to move beyond that debate.

2. a historically contextualized reading of oe

If we read OE as Hume intended, we retrieve a much more subtle and sophisticated view than the Tickles View. After establishing that OE is properly understood as an argument against a kind of rational passion, I develop three historical claims about Hume's target: (i) moral rationalists such as Clarke and Wollaston were not committed to the existence of rational passions, but (ii) the Stoics were, and (iii) Hume was skeptical of the Stoics' arguments. [End Page 613]

2. 1. Hume's Argumentative Strategy

Hume claims that OE supports his arguments about the motivational inertness of reason alone. He makes this claim both times he invokes the originality of the passions in connection with morals in the Treatise, namely at 2.3.3 and 3.1.1. In the former ("Of the influencing motives of the will"), Hume famously rejects two ways in which reason alone could motivate: (i) by demonstrative reasoning alone, (ii) by probabilistic reasoning alone. Schematically, his view is that neither demonstrative reasoning about logical or mathematical truth, nor probabilistic calculation of states of affairs, can motivate unless it serves some antecedent interest of the agent. The upshot of (what I shall call) Hume's Argument from Inertness is that if reason does not meet this antecedent interest requirement, it is motivationally inert.13

However, commentators seldom note that Hume proceeds to reject a third distinct way in which reason alone can motivate: (iii) by "giving an impulse in a contrary direction to our passion." I reproduce Hume's argument below, numbering and labeling its steps:

[1] Since reason alone can never produce any action, or give rise to volition [Argument from Inertness], [2] I infer, that the same faculty is as incapable of preventing volition, or of disputing the preference with any passion or emotion. [3] This consequence is necessary. 'Tis impossible reason cou'd have the latter effect of preventing volition, but by giving an impulse in a contrary direction to our passion; and that impulse, had it operated alone, wou'd have been able to produce volition. [4] Nothing can oppose or retard the impulse of passion, but a contrary impulse; [5] and if this contrary impulse ever arises from reason, that latter faculty must have an original influence on the will, and must be able to cause, as well as hinder any act of volition. [6] But if reason has no original influence, 'tis impossible it can withstand any principle, which has such an efficacy, or ever keep the mind in suspence a moment. [7] Thus it appears, that the principle, which opposes our passion, cannot be the same with reason, and is only call'd so in an improper sense.

(T 414–15)

In 1–4, Hume anticipates someone who concedes that reason alone cannot directly motivate the will through demonstrative or probabilistic reasoning, but who nevertheless maintains that reason alone can cause its own passions and so indirectly motivate the will. For example, when I am inclined to act on my brute passion for vengeance, reason "opposes" such a passion by causing a countervailing passion of its own for justice or clemency. Let us call these passions caused by reason 'rational passions.' In 5–7, Hume rejects the possibility of rational passions on the grounds that the ability of reason alone to cause passions presupposes its ability to cause motivating perceptions. But, according to Hume, we find no evidence that reason alone can cause motivating perceptions. Indeed, Hume thinks that his Argument from Inertness has already shown that reason alone is unable to generate the requisite affective force to influence us. Hume concludes that because reason alone cannot cause motivating perceptions, there is no support for the claim that reason alone can cause motivating perceptions such as passions. [End Page 614]

Significantly, however, Hume adds, "As this opinion may appear somewhat extraordinary, it may not be improper to confirm it by some other considerations," and he then offers OE. Hume therefore intends the passions' "original existence" to complete his argument against the rational passions. However, it is unclear how this maneuver works, and few readers take it seriously.14 I argue that Hume's strategy becomes intelligible if we retrieve the conception of rational passions to which OE is his answer.

2. 2. OE and the British Moral Rationalists

Scholars often frame OE as Hume's response to British moral rationalists such as Clarke and Wollaston.15 For example, Rachel Cohon suggests that OE "may be an attempt on [Hume's] part to parse the moral rationalists' position: perhaps when they say that immoral actions are contrary to reason, they mean that immoral actions are false—that they are inaccurate representations of reality."16 Cohon names Wollaston in this connection, and I agree that Hume is hostile to Clarke's and Wollaston's rationalist view about what moral values are: a kind of a priori truth or rational relation. As Hume summarizes the disagreement in his anonymously written Letter from a Gentleman, he "hath indeed denied the eternal Difference of Right and Wrong in the Sense in which Clarke and Woolaston maintained them, viz. that the Propositions of Morality were of the same Nature with the Truths of Mathematics and the abstract Sciences, the Objects merely of Reason" (Gentleman, 30). Nevertheless, Clarke's and Wollaston's cognitivist description of values is not at issue in OE, which Hume frames as a rebuke to rational passions.

The standard scholarship may object that Clarke and Wollaston subscribe to a second rationalist commitment that is a better match for OE's intended target: reason alone can motivate the will. According to Clarke and Wollaston, moral decision-making is like completing proofs in geometry. Once we recognize the moral relationships in which the pertinent facts stand, our course of action necessarily follows. Clarke, for instance, explains that just as

no one, who is instructed in mathematics, can forbear giving [her] assent to every geometrical demonstration, of which [she] understands the terms… so no [one], who either has patience and opportunities to examine and consider things [herself], or has the means of being taught and instructed in a tolerable manner by others, concerning the necessary relations and dependencies of things; can avoid giving [her] assent by the law or rule [of righteousness].

(DRN 201–2)17 [End Page 615]

The standard reading is again correct, of course, that Hume is hostile to the kind of motivational moral cognitivism defended by Clarke and Wollaston, and they are likely targets of Hume's rejection (in his Argument from Inertness) of the motivational efficacy of reason alone in its demonstrative and probabilistic employment. As I shall indicate, however, they are unlikely targets of Hume's OE argument against the existence of passions caused by reason alone.

Wollaston leaves the passions relatively un-thematized, but there are several places in which Clarke raises the possibility of unreasonable passions. Here are six passages in which Clarke speaks of unreasonable passions.

[1.] [T]he unreasonable desires of any depraved natures.

[2.] [W]hose wills are corrupted by particular interest or affection, or swayed by some unreasonable desire and prevailing passion.

[3.] [I]t is absolutely impossible in nature, that God should be deceived by any error, or influenced by any wrong affection.

[4.] [W]illfully and perversely allow themselves to be over-ruled by absurd passions, and corrupt or partial affections.

[5.] [S]uffer themselves to be swayed by unaccountable arbitrary humours, and rash passions, by lusts, vanity and pride; by private interest, or present sensual pleasures: these, setting up their own unreasonable self-will in opposition to the nature and reason of things, endeavour (as much as in them lies) to make things be what they are not, and cannot be.

[6.] This is indeed the great difficulty of life, to subdue and conquer our unreasonable appetites and passions.

Clarke is undoubtedly committed to the existence of unreasonable passions, the irrationality of which seems to consist in their being (or motivating us to be) heedless of moral truth.18

Yet, I find little evidence that Clarke thinks that there are reasonable passions, nor should we expect him to think that they exist, given his defense of the thesis that reason alone directly motivates the will. For example:

[The] reason of things… ought also constantly to determine the wills of all subordinate rational beings, to govern their actions by the same rules.… [For] God has made [intelligent creatures] so far like unto himself, as to endue them with those excellent faculties of reason and will, whereby they are enabled to distinguish good from evil, and to choose the one and refuse the other.… Which two things, viz. negligent misunderstanding and willful passions or lusts, are… the only causes which can make a reasonable creature act contrary to reason, that is, contrary to the eternal rules of justice, equity, righteousness and truth.

Clarke, whose view is that reason alone can and ought to motivate the will, spares little thought for the passions, which he instead regards as sources of moral error. Still, there is one passage in which Clarke could be read as embracing the possibility of a reasonable passion: "it is evident every [person] is bound by the law of [her] nature, as [she] is also prompted by the inclination of [her] uncorrupted [End Page 616] affections, to look upon [her] self as a part and member of that one universal body or community" (DNR 210). While the phrase "uncorrupted affections" is suggestive, the difficulty with reading this as Clarke's recognition of reasonable passions is that Clarke himself does not link 'reasonability' to the passions, either in the quote or in its context. Clarke is instead attempting to show that "the obligation to this great duty [of benevolence], may also otherwise be deduced" from human nature, specifically the "natural affection" for children, family, friends etc. (DNR 210). Given that Clarke provides no indication that he conceives of these uncorrupted natural affections (or passions) as rational per se, the most we can conclude is that Clarke allows that some passions are compatible with reason.

In sum, neither Clarke nor Wollaston match Hume's target in OE. While Clarke is committed to unreasonable passions, their "un-reasonability" turns on their being heedless of moral truths. Standard readers may be tempted to see here a connection to Hume. If we suppose that Clarke thinks such passions are unreasonable because they do not represent moral truths, then this looks quite similar to the view that standard readers take Hume to advance in OE. However, there is no evidence that Clarke has any theory about the passions' ability or inability to represent truth or anything else.

Moreover, Hume's quarry is different.19 Hume gives a two-part argument against the view that reason alone can cause passions, and there is no evidence that Clarke or Wollaston held that view either.20 Of course, it is possible that Hume is arguing against a position that no one ever held. However, if there is a real historical match, identifying it will place us at the interpretive vantage point from which to reconstruct Hume's intentions in OE. I shall argue that the Stoics did hold such a view and that Hume was skeptical of it.

2. 3. OE and the Stoics

The Stoics were committed to the existence of rational passions. In this section, I present the Stoics' theory of the rational passions (what I shall call their Rational [End Page 617] Passions Thesis), which comprises two interrelated but distinct claims. First, a causal claim: in general, reason causes the passions. Second, a representational claim: the rational passions in particular are true cognitions of value. I note at the outset that I do not intend my survey of Stoicism to be exhaustive, and there were doubtless various distinctions and disagreements among the Stoics that I shall not address.

The Causal Claim: Stoic cosmology holds that the universe is a kind of living organism, with passive and active aspects. The passive aspect is matter; the active aspect is divine mind or reason (logos). The Stoics compare the divine logos to a warm or fiery "wind" or "impulse" that animates the universe, just as breath animates the living body.21 On this view, then, in some sense everything is caused by these rational impulses permeating the cosmos. Here are three representative statements:

[1. Calcidius:] And so the universal body, according to the Stoics… is substance, because it is the prime matter of all bodies, and through it, they say, complete and universal reason passes, just like seed through the genital organs. This reason they take to be an actual craftsman.

[2. Alexander:] They [the Stoics] say that god is mixed with matter, pervading all of it and so shaping it, structuring it, and making it into the world.

[3. Seneca:] Our Stoic philosophers, as you know, say that there are two things in nature from which everything is produced—cause and matter. Matter lies inert, an entity ready for anything but destined to lie idle if no one moves it. Cause, on the other hand, being the same as reason, shapes matter and directs it wherever it wants, and from matter produces its manifold creations.22

Applying their theory of universal rational causation to the passions in particular, the Stoics conceive of even our most basic affects as "impulses" set in motion by the divine reason immanent in nature and our bodies.23

A caveat: although divine reason saturates everything in the cosmos, the Stoics claim that this reason is more highly concentrated in some places (namely, the human mind) than others. As Diogenes Laertius reports:

The world, in their view, is ordered by reason and providence: so says Chrysippus… inasmuch as reason pervades every part of it, just as does the soul in us. Only there is a difference in degree; in some parts there is more of it, in others less. For through some parts it passes as a "hold" or containing force, as is the case with our bones and sinews; while through others it passes as intelligence, as in the ruling part of the soul. Thus, then, the whole world is a living being, endowed with soul and reason, and having aether [i.e. the divine wind or logos] for its ruling principle.

The Stoics therefore make a distinction between two ways in which reason can cause passions. Reason can cause passions indirectly, through the divine logos immanent in bodily impulses. Reason can also cause passions directly, when the human mind [End Page 618] sets in motion affective impulses of its own. Thus Stobaeus and Philo speak of a "rational impulse" that is "formed by the tonic power of the mind" (LS 317). I shall refer to these latter passions, which are caused directly by human reason, as the Stoic rational passions.

The Representational Claim: Stoicism's rational passions are not only caused by reason, but are true representations of value.25 On the Stoic view, the passions are impulses of attraction and repulsion. These impulses arise from impressions of various objects or states of affairs as worthy of pursuit or avoidance.26 A passion is therefore a kind of moral judgment.27 When we experience a passion, we are in effect agreeing with its representation of something as valuable. This view raises two major issues: (i) what is Stoicism's theory of representation, and (ii) what, according to Stoicism, do the rational passions correctly represent? I shall say a few words about each.

First, the Stoics hold a kind of resemblance theory of representation, on which representation is an affordance of the similarity between objects and our conceptions of them. The nature of Stoic representation is underdeveloped in the scholarly literature, so rather than shadowboxing, I shall simply presuppose my interpretation for the sake of argument. In any event, a natural way of reading the following three testimonials is as attributing to the Stoics something like a resemblance theory of representation:

[1. Cicero:] Zeno… replied that the wise [person] would not opine since there was something cognitive. What then was this? Zeno, I suppose, said: an impression. What kind of impression? Zeno then defined it as an impression stamped and reproduced from something which is, exactly as it is.28

[2. Diogenes Laertius:] A presentation (or mental impression) is an imprint on the soul: the name having been appropriately borrowed from the imprint made by the seal upon the wax. There are two species of presentation, the one apprehending the real object, the other not. The former, which they take to be the test of reality, is defined as that which proceeds from a real object, agrees with that object itself, and has been imprinted seal-fashion and stamped upon the mind.

[3. Boethius:] On the philosophers' painted stoa the ancient thinkers instructed how it can happen that we perceive in the outside world images of those objects that then appear in our minds.… Is the mind like… a polished mirror that reflects what comes before it?

Second, the Stoics believe that the representations afforded by our rational passions provide insight into (what they characterize as) the divine administration of the cosmos. Many of the Stoics' examples of rational passions are variations of [End Page 619] the kind of tranquility or peace of mind that comes when we reflect from a cosmic or divine point of view. As Seneca puts it: "What is a happy life? Peacefulness and tranquility. Loftiness of mind will bestow this, and consistency which holds fast to good judgement. How are these things reached? If all of truth has been seen.… To put it in a nutshell for you, the wise [person's] mind should be such as befits god" (LS 396). The Stoic rational passions would therefore appear to be true moral judgments, insofar as they resemble what the divine administrator of the cosmos would feel when reflecting on a situation.29 The Stoics claim that all things harmonize or balance out in the good of the whole, and from that perspective the appropriate response to the ebbs and flows of our fortunes will be the equanimity characteristic of the rational passions. The Stoics call this a case in which "reason supervenes as the craftsman of impulse," for what results from our cognitive attunement to the divine mind or divine reason are rational impulses or passions (eupatheiai) that orient us to the pursuit of the right things (Lives, 2.193–95).30 Thus, Diogenes Laertius reports that the Stoics call a rational passion a "well-reasoned shrinking" from or "swelling" toward things, and Epictetus speaks (in the context of the passions) of our "correct use of impressions" (Lives, 2.221; Discourses, 1.1.7).31

In sum, the Stoics understand rational passions to correctly represent moral worth in virtue of resembling the perspective of the divine administrator of the cosmos.

2. 4. Hume on the Stoic Rational Passions

Hume was acquainted with, and increasingly skeptical of, the Stoic theory of rational passions. There are at least five pieces of evidence in Hume's writings leading up the publication of Book III of the Treatise in 1740, and two from the year immediately following, that indicate his familiarity with key aspects of Stoicism's Rational Passions Thesis.

(1) The earliest evidence of Hume's interest in Stoicism is his appreciation of Shaftesbury. Shaftesbury's Characteristics of Men, Manners, and Times was an early acquisition of Hume's personal library. The first of the three-volume set of Characteristics owned by Hume is signed and dated "David Home 1726," a year in which Hume was fourteen or fifteen years of age.32 The date is significant because in the Advertisement that Hume places at the front of his later Enquiries, he claims that the Treatise was "projected before he left College," which, if true, means that Hume began developing the ideas of the Treatise around 1726.33 Although I cannot discuss the issue here, there are some signs that Hume started reading his Shaftesbury soon after acquiring it.34 [End Page 620]

If we allow that Hume read Shaftesbury's Characteristics before drafting his Treatise, then it is likely that Hume came across Shaftesbury's discussion of the Stoics. In a pair of footnotes, Shaftesbury provides extensive quotations from Marcus Aurelius and Epictetus. Significantly, the passages that Shaftesbury reproduces are references to Stoicism's Rational Passions Thesis.35

Apprehension is everything, and this is up to you. Therefore, remove the apprehension when you wish, and there is a great calm as though you were rounding the headland, and all is still and the bay is still: Marcus Aurelius, Meditations 12.22. The soul is like the basin of water. Fancies are like the ray of light that strikes upon the water. Thus, when the water is disturbed, the ray seems too to be disturbed; but it is not. And so when anyone is agitated, it is not the arts and the virtues that are confounded but the spirit in which they exist. And when this steadies, they do as well: Epictetus, Discourses 3.3.20–22.

Calmness is one of the Stoic rational passions, and in their description of calmness as a certain movement of the soul caused by "apprehension," steady judgment, or harmonization with these (rational) rays or impulses streaming into us from without, Marcus Aurelius and Epictetus allude to the causal claim that reason somehow generates a rational passion.

You must give up desire completely, transfer your aversion exclusively to things that are the business of moral purpose: Epictetus, Discourses 3.22.13. This subdued or moderated admiration or zeal in the highest subjects of virtue and divinity, the philosopher calls commensurate and composed desire; the contrary disposition, the irrational and impetuous.… And hence the repeated injunction, refrain at some time altogether from desire, so that at some time you may actually desire with good reason. If you desire with good reason,… you will be desiring well.

Epictetus is distinguishing between improper and proper desires. The object of proper desires is sought "with good reason" and is a response of "admiration or zeal" before representations of "virtue and divinity." Here, Epictetus invokes the representational claim of Stoicism's Rational Passions Thesis: A rational passion is an impulse that correctly represents impressions of cosmic moral order. In sum, it is probable that Hume encountered these excerpts from the Stoics that defend both the causal and representational components of their theory of rational passions.

(2) In a 1727 letter to Michael Ramsay, Hume reveals that he had been reading Cicero and striving to emulate the Stoics' call to "peace of mind" (a paradigmatic Stoic rational passion). To achieve this mental state, Hume explains that he must be immune to "perturbation": "This Greatness & Elevation of Soul is to be found only in Study and contemplation.… Wherefore I wisely practice my Rules, w[hich] prescribe to check our Appetite" (Letters, 1.10). Not only does Hume's program of stilling the tempestuous movements of his mind through rules and study have a Stoic ring to it, but Hume himself also attributes the inspiration of his regimen to his reading of Cicero's popularization of Stoic philosophy in the Tusculan Disputations.

(3) The 1734 letter that Hume drafted to a physician confirms Stoicism's influence on his regimen. Hume states that his regimen had resulted in a psychosomatic breakdown, and he links the onset of his symptoms ("Scurvy Spots"; [End Page 621] "Watryness in the mouth"; inability "to follow out any Train of Thought"; "a Coldness & Desertion of the Spirit"; "Melancholy") with his reading of Stoic philosophy:

There was another particular, which contributed more than any thing, to waste my Spirits & bring on me this Distemper, which was, that having read many Books of Morality, such as Cicero, Seneca & Plutarch, & being smit with their beautiful Representations of Virtue & Philosophy, I undertook the Improvement of my Temper & Will, along with my Reason & Understanding. I was continually fortifying myself with Reflections against Death, & Poverty, & Shame, & Pain, & all the other Calamities of Life.

Hume's report on his self-improvement project is important in two respects. First, the authors to whom Hume attributes his project are either Stoics (Seneca) or sources of extensive report and commentary on the Stoics (Cicero, and Plutarch). Second, when taken in conjunction with his 1727 letter to Michael Ramsay, Hume appears to have been trying to improve his "Temper & Will" by means of the "Rules" issued by his "Reason & Understanding" (per the causal claim of Stoicism's Rational Passions Thesis) and to improve his "Reason & Understanding," in turn, with morally correct "Reflections" (per the representational claim of the Rational Passions Thesis).

(4) Hume's break from Stoicism is evident in a September 17, 1739, letter that he wrote to Francis Hutcheson, who had offered comments on what was likely a draft of the still unpublished Book III of the Treatise.36 Although Hutcheson's comments are not extant, Hume's reply to Hutcheson focuses, in part, on whether the mere cognition of virtue can give rise to a "Motive or impelling Passion":

You are a great Admirer of Cicero, as well as I am. Please to review the 4th Book, de finibus bonorum & malorum; where you find him prove against the Stoics, that if there be no other Goods but Virtue, 'tis impossible there can be any Virtue; because the Mind would then want all Motives to begin its Actions upon.… This proves, that to every virtuous Action there must be a Motive or impelling Passion distinct from the Virtue, & that Virtue can never be the sole Motive to any Action. You do not assent to this; tho' I think there is no Proposition more certain or important. I must own that my Proofs were not distinct enough, & must be altered.

Hume does not explain why he associates the view to which he is opposed with the "Stoics." We find a clue, however, in the fact that the language of his letter closely follows that of a passage from Book III's discussion of justice. There, having objected to the view that "a mere regard to the virtue of an action" can give rise to a motive for justice (T 478), he concludes that "no action can be laudable or blameable, without some motives or impelling passions, distinct from the sense of morals" (T 483). The language of "laudable or blameable" actions is significant because it, in turn, echoes passages in Treatise 3.1.1 that recast Treatise 2.3.3's argument involving the passions' "original existence." In Treatise 3.1.1, Hume:

(i) rejects the causal claim that reason alone can cause passions ("Morals excite passions, and produce or prevent actions. Reason is utterly impotent in this particular"); [End Page 622]

(ii) adds that he "shall endeavor to render [his argument] still more conclusive" with another observation; viz.

(iii) an invocation of the passions' existence qua "original facts and realities," allowing him to conclude to his satisfaction that while "actions may be laudable or blameable," there is no basis for the representational claim that the passions could somehow be "pronounced either true or false."

(T 457–58)

Two points emerge from this trail of textual breadcrumbs. First, Hume's letter to Hutcheson refers us to the "Proofs" of his argument in the Treatise. While wording of the letter admittedly resembles the proof that he provides in his discussion of justice, Hume speaks to Hutcheson of multiple proofs. Consistent with the wording of the letter, then, Hume's proof concerning justice calls upon the language he employed in his earlier proof rejecting the rational passions (on the grounds that the causal and representational claims made about the rational passions are untenable). Second, while Hume does not identify these claims about the rational passions as Stoic in either Treatise 2.3.3 or 3.1.1, a chain of references links those claims to a view that he does associate with Stoicism in the 1739 letter. In that letter, Hume presumably regards the view that the mere cognition of virtue (or justice) can generate a "Motive or impelling Passion" as a version of Stoicism's Rational Passions Thesis. In short, the letter to Hutcheson indicates that Hume himself considered key portions of the Treatise to contain anti-Stoic "Proofs."

(5) Hume's early memoranda also show signs of his familiarity with Stoicism. Ernest Mossner edited the memoranda into three categories: natural philosophy, philosophy, and miscellanea concerning economics, history, and politics. The memoranda are of uncertain dating, ranging from perhaps from 1739 to 1745, but there is some consensus that the earliest of them are "coterminous with the end of Hume's main work on the Treatise," which is to say from 1739 to 1740.37 Of the three sets of memoranda, those on philosophy are the closest to the Treatise in subject matter, and it is here that we find three references to the Stoics. (i) Hume lists the Stoics as one of "Three kinds of Atheists according to some" (Memoranda, 501, entry 12). (ii) Hume notes that "A Stratonician cou'd retort the Arguments of all Sects of Philosophy. Of the Stoics, who maintain'd their God to be fiery and compound" (Memoranda, 501, entry 15). (iii) Hume names the Stoics in his typology of "Four kinds of Atheists according to Cudworth" as "the Stoic or Cosmo-plastic" kind (Memoranda, 503, entry 40). Despite the fragmentary nature of these remarks, we can draw a pair of tentative conclusions. First, Hume seems to have been studying commentaries on Stoicism while he was fine-tuning the final arguments of the Treatise. Second, Hume has some knowledge of Stoic cosmology. In memorandum (ii), he recognizes that the Stoics conceived of their deity as "fiery." The Stoics used the image of fire to capitalize on their comparison of the cosmos to an organism animated by the deity's warm breath or pneuma; for example, Aetius reports that "The Stoics made god out to be intelligent, a designing fire… a breath pervading the whole world" (LS 274–75). In memorandum (iii), he notes that the Stoics viewed the universe as saturated with some kind of "cosmoplastic" [End Page 623] principle of organization. Taken jointly, these memoranda are significant because they reveal Hume's awareness of a key cosmological claim underpinning Stoicism's Rational Passions Thesis, viz. that there is a divine rational cosmic order.

(6) The two post-Treatise essays that I shall consider are "The Stoic" and "The Sceptic," published early in 1742 and presumably written the preceding year (i.e. just one year after Hume finished the Treatise). Consider this passage from "The Stoic":

But the social passions [of the sage] never afford such transporting pleasures, or make so glorious an appearance in the eyes of both GOD and [human], as when, shaking off every earthly mixture, they associate themselves with the sentiments of virtue, and prompt us to laudable and worthy actions.… [T]he human mind, which, being of celestial origin, swells with the divinest and most enlarged affections, and carrying its attention beyond kindred and acquaintance, extends its benevolent wishes to the most distant posterity.

Hume goes on to describe the sage's distinctive passions as "tuned to just harmony and concord" as the result of our contemplation, reflection, and mental industry (EMPL 153). Although Hume does not distinguish between the causal and representational claims of Stoicism's Rational Passions Thesis as such, we can detect his awareness of both in the distinction he does draw between the causes and objects of these Stoic passions. The cause is the mind and its rational activity (as opposed to the "earthly mixture" of the body). The objects are "sentiments of virtue" shared by humankind and God; on the Stoic view, this would mean that these passions are correct judgments of value because they approximate the impressions afforded by the point of view of the divine administrator of the cosmos.

(7) In "The Sceptic," Hume identifies some problems with both Stoic claims. In one passage, he argues that "to excite or moderate [a person's] passions, there are no direct arguments or reasons, which can be employed with any force or influence" (EMPL 171). He explains that it is as ridiculous to suppose we could excite a passion by reading the arguments of Seneca and Epictetus as to suppose we could cure heartsick by examining the skin of the beloved under a microscope with a view toward raising a countervailing passion of disgust (EMPL 172). Interestingly, Hume's example of curing love by exciting a countervailing passion through a kind of scientific contemplation is borrowed from Epictetus (Hume adds the microscope imagery).38 Therefore, Hume identifies the rational causation of passions—the topic addressed in OE—as a Stoic view.

In "The Sceptic," Hume also sketches concerns about Stoicism's claim that the rational passions correctly represent some cosmic moral order. Hume amasses, and subjects to ridicule, a series of quotes, including some from Stoic commentators such as Plutarch and Cicero, that urge us to reflect on how foolishly mistaken our pleasures and pains are when viewed from the point of view of "the order of the universe" (EMPL 173). In response, Hume says that a "supreme being" may well have such impressions of the order of the universe and "feel… their truth," but in contrast our "human capacity [to have passions]… [is] not… fortified by the experience of anything better" than the pursuits of common life (EMPL 175–76). [End Page 624] So, in these passages we find Hume skeptical that the passions of the Stoic sage represent moral truths beyond those afforded by human nature.

In short, I have identified five documents that evince Hume's preoccupation with Stoicism's Rational Passions Thesis during the period roughly contemporaneous with the writing of the Treatise. These documents also provide evidence that Hume recognized, and had access to texts that recognized, that the Stoics' Rational Passions Thesis encompasses a causal as well as a representational claim.

2. 5. Summary

Hume instructs us to read OE as part of a continuous line of argument against someone committed to the existence of rational passions. Clarke and Wollaston are not a good match for Hume's target, but the Stoics are, and I have presented textual evidence that Stoicism's Rational Passions Thesis was at the fore of Hume's concerns during the period surrounding his writing of the Treatise and OE. The implication of these findings for my historically contextualized interpretation of OE is the topic of the next section.

3. oe and stoicism's representational claim

On my view, the best way to read Hume's characterization of the passions as "original existences" is in terms of his rejection of the Stoics' Rational Passions Thesis. First, I argue that Hume's use of "original" refers to his Copy Principle's ban on simple impressions representing their causes in virtue of resemblance. Second, I argue that in OE Hume is rejecting Stoicism's representational claim about rational passions, which he considers to violate the epistemic strictures of his Copy Principle.

3. 1. Originality and the Copy Principle

Hume makes two distinctions at the outset of the Treatise. There, Hume first divides all the perceptions of the mind into two classes, impressions and ideas. Impressions are our most forceful or vivid perceptions, and ideas are somewhat fainter or less lively "images" of them (T 1). To develop the relationship between impressions and ideas further, Hume introduces a second distinction between simple and complex perceptions (T 2). If we grant Hume's distinctions, we arrive at four possible types of perceptions: simple impressions, complex impressions, simple ideas, and complex ideas.

Hume takes the impression/idea and simple/complex distinctions to license "one general proposition" that many commentators have labeled the Copy Principle.39 The Copy Principle flags a special feature of the relationship between simple ideas and simple impressions: "That all our simple ideas in their first appearance are deriv'd from simple impressions, which are correspondent to them, and which they exactly represent" (T 4). Hume's Copy Principle actually makes two distinct claims: (i) simple impressions cause simple ideas (T 5), and (ii) simple ideas are "exact copies" or resemblances of the simple impressions that caused them [End Page 625] (T 3). Provisionally, we can say that Hume's Copy Principle supports a kind of causal theory of representation, for what our simple ideas represent are the simple impressions that caused them.40 However, there is an asymmetry: Simple ideas represent the simple impressions that caused them in virtue of resembling or copying those causes, but simple impressions do not in turn represent their own physiological causes in virtue of resembling them. This is not to deny that simple impressions can represent, and there are passages in which Hume suggests they can. For example, speaking of the "simple and indivisible" images that appear to our senses, he writes of these simple impressions of sensation that "the only defect of our senses is, that they give us disproportion'd images of things, and represent as minute and uncompounded what is really compos'd of a vast number of parts" (T 28, emphasis added).41 But, however simple impressions of sensation represent, on Hume's view it is not in virtue of resembling their causes.42 Hume also thinks that in addition to impressions of sensation, there are impressions of reflection (what he later calls secondary impressions); namely, "the passions, and other emotions" (T 275).43 I shall argue that these impressions of reflection can represent as well, subject to the same caveat.

Hume's characterization of impressions as "original" typically serves as an in-house reference to that representational asymmetry of his Copy Principle. The following passage reflects this terminological practice:

Original impressions or impressions of sensation are such as without any antecedent perception arise in the soul, from the constitution of the body, from the animal spirits, or from the application of objects to the external organs. Secondary, or reflective impressions are such as proceed from some of these original ones, either immediately or by the interposition of its idea. Of the first kind are all the impressions of the senses, and all bodily pains and pleasures: Of the second are the passions, and other emotions resembling them.

(T 275)

While Hume does not here explicitly contrast original impressions with perceptions that copy or resemble other perceptions, a remark in the first Enquiry indicates that this is the intended contrast. There, Hume allows that, if they wish, philosophers can call our simple impressions innate, "understanding by innate, what is original or copied from no precedent perception" (EHU 2.9n./SBN 22n.). Following Locke and Berkeley's usage, Hume classifies an impression as "original" when it does not copy or resemble any precedent impression or cause.44 [End Page 626]

My interpretive claim is that Hume appeals to this definition of 'original' when he characterizes the passions as "original existences," but first I must address a worry about my strategy: Hume categorizes (what he calls indirect) passions, not as original impressions, but rather as secondary impressions produced by a "double association of impressions and ideas."45 To disarm this objection, I develop an interpretation of the Humean passions as compounds of impressions and ideas.46 Specifically, I argue that the passions are complex impressions of reflection, which contain simple impressions of reflection (emotions) as one of their component parts. Hereafter, my use of 'impressions' is shorthand for these simple impressions of reflection unless otherwise specified.

It is not obvious that Hume thinks the passions are compounds. Indeed, at one point he declares that "The passions of PRIDE and HUMILITY [are] simple and uniform impressions" (T 277). Elsewhere, however, Hume says that the passions are causally complex: "Since, therefore, these two particulars [of the cause] are easily separated, and there is a necessity for their conjunction, in order to produce the passion, we ought to consider them as component parts of the cause" (T 279). We might think that complexity in the cause is compatible with simplicity in the effect, but that is not Hume's view. The phrase "component parts of the cause" recalls language that Hume uses to formulate the seventh of his "rules by which to judge of causes and effects," in which he claims that complexity in causes tokens complexity in effects.47 He writes:

We may establish it as a certain maxim, that in all moral as well as natural phenomena, wherever any cause consists of a number of parts, and the effect encreases or diminishes, according to the variation of that number, the effect, properly speaking, is a compounded one, and arises from the union of the several effects, that proceed from each part of the cause.

(T 136)48 [End Page 627]

Given Hume's extension of this rule to moral phenomena, it likely informs his claim that the passions have "component parts," which means that the passions are compounded effects, and something that is compound is surely complex.49 However, if we remind ourselves that Hume also says that the passions are simple, his position looks contradictory.

Yet, Hume is not guilty of inconsistency. My explanation focuses on his claim that the passions are compounded out of an "association of ideas," which works on some object (e.g. a house) to give us the idea of its being "ally'd or related to us" or another, and an "association of impressions," which works on the quality (e.g. beauty) to "produce a separate pain or pleasure."50 Hume's analysis of the passions as causal products of a double association of impressions and ideas is significant because he therefore ought to regard the passions as complex or compounded effects, in keeping with his seventh rule for judging causes and effects, and Hume in fact does so in several passages.51 Given that Hume understands compounded effects to have component parts, it is unsurprising that he introduces some new terminology to refer to the impressional or hedonic component of the passion. He calls the impressional-component, an "emotion or original impression":

'Tis evident, then, that when the mind feels the passion either of pride or humility upon the appearance of a related object, there is, beside the relation or transition of thought [i.e. association of ideas], an emotion or original impression produc'd by some other principle [i.e. by the association of impressions].

(T 305)

Few commentators note Hume's distinction between the emotion qua component of a passion and the passion itself, and some assume that emotions and passions are equivalent.52 Hume is partly to blame for the neglect of his distinction, as he occasionally speaks of emotions and passions interchangeably.53 However, these passages' silence on a distinction must be put into perspective alongside the overwhelming number of passages in which Hume reserves the term 'emotion' to designate the product of an association of impressions (i.e. its typical physiological effect, a kind of sensible agitation).54 [End Page 628]

Hume's distinction between emotions and passions has two important exegetical consequences. First, it removes the apparent contradiction between his commitment to the complexity of the passions qua compounded effects, on the one hand, and his claim that the passions are "simple and uniform impressions," on the other.55 On my reading, when Hume mentions the passions' simplicity, he is referring to their emotional basis. In the following passage, for instance, Hume describes how we ordinarily identify the hedonic sensations or emotions (i.e. simple impressions) of a passion with the passion itself:

The second quality, which I discover in these passions, and which I likewise consider as an original quality, is their sensations, or the peculiar emotions they excite in the soul, and which constitute their very being and essence. Thus pride is a pleasant sensation, and humility a painful; and upon the removal of the pleasure and pain, there is in reality no pride nor humility. Of this our very feeling convinces us; and beyond our feeling 'tis here in vain to reason or dispute.

(T 286)

Hume is making a distinction between the causal complexities of a passion, as a passion appears to him in his capacity as a scientist of human nature, and the phenomenological simplicity of the passions. When we experience passions, we are usually not judging their causal ancestry, and, according to Hume, we therefore tend not to recognize the ideational component, the salience of which pales in comparison to the emotional or impressional-component. We count the "very feeling" as the passion itself, and we cannot be convinced by "reason or dispute" that we have or should have a passion (say, by someone showing us that the relevant causal conditions for pride are in place) if we do not actually feel the emotion. The phenomenological simplicity of the passions qua emotions that Hume describes, however, is consistent with the passions themselves being causally complex.

The second exegetical consequence of Hume's distinction between emotions and passions is that it recovers the background assumptions operative in OE. Recall my initial suggestion: Interpret "original," in accordance with Hume's epistemic use of the word elsewhere in the Treatise, as signifying that simple impressions do not represent their physiological causes in virtue of resembling them. The apparent barrier to that proposal was that Hume classifies passions as secondary, not original or simple. We can now see our way around that barrier. Hume conceives of the passions as compound or complex effects, one of the causal components of which [End Page 629] is a "simple and uniform" "emotion or original impression." So, when Hume speaks of "originality" in connection with the passions, he is invoking the non-resemblance claim with respect to the representational capacity of their emotional constituents.

In the locus classicus of OE, Hume invokes this claim in specific reference to the emotions:

A passion is an original existence, or, if you will, modification of existence, and contains not any representative quality, which renders it a copy of any other existence or modification. When I am angry, I am actually possest with the passion, and in that emotion have no more a reference to any other object, than when I am thirsty, or sick, or more than five foot high.

(T 415)

According to standard readings of this passage, Hume's claim that anger "has no more a reference to any other object" than thirst etc. is an implausible denial of the intentionality of anger. Moreover, according to standard readings, that claim may be inconsistent with his suggestion in the following remark that anger can nonetheless represent other states of affairs: "as anger or the appetite, which attends hatred, is a desire of the misery of the person hated, and an aversion to [her] happiness" (T 382; see also T 329–30). On my reading, the text warrants neither of these standard conclusions. Hume emphasizes being "actually possest with" anger or being "in that emotion[al state]," indicating that his focus is on the emotional constituents of the passion of anger. His denial that the emotional constituents of anger have "any representative quality, which renders it [anger] a copy of any other existence" reflects his denial that the emotional component parts of a passion do not represent their causes in virtue of resembling them. In this causal-resemblance sense of representation, anger qua emotion is a simple impression of reflection that, like the simple impressions of sensation that comprise our perceptions of thirst or sickness, represents certain states of affairs to us without serving as a mirror-like "reference to any other object" that is the physiological cause of that impression.56 In sum, in this first OE passage, Hume is neither making implausible claims nor contradicting himself, but merely ruling out the particular claim that the emotional constituents of passions such as anger represent their causes in virtue of resembling them.

Now consider the second OE passage:

Now 'tis evident our passions, volitions, and actions, are not susceptible of any such [rational] agreement or disagreement; being original facts and realities, compleat in themselves, and implying no reference to other passions, volitions, and actions.

(T 458)

One might object that Hume's failure to speak of emotions prevents me from replicating the interpretation I offered for the previous OE passage. However, Hume is likely adopting the terminology of the vulgar, who he says equate [End Page 630] the passions with their emotional constituents. On my reading of this passage, "passion" has a narrow sense that refers only to the emotional components of a passion (in contrast to a wider sense of "passion" that refers to the compound of hedonic impressions and ideas).57 Consequently, Hume is still labeling the passions as "original" because qua hedonic impressions they are not copies (of their physiological causes), the resemblance or agreement of which to the original reason could check.

Hume therefore conceives of the passions as complex mental states that are secondary in one respect (insofar as they are ideas or have ideas as their objects), but original in another respect (insofar as they are simple impressions or emotions that do not represent in virtue of resembling their physiological causes).

3. 2. OE: The Second Part of a Two-Part Argument against the Stoics

We are now in a position to see that OE is Hume's rejection of the Stoic Rational Passions Thesis. To review, Hume instructs us to read OE as completing his preceding argument against a particular causal claim; namely, the motivational efficacy of reason alone to generate "an impulse in a contrary direction to our passion." Hume says he will "confirm" his skepticism of that causal claim by invoking the "originality" of the passions, which is a word he uses when he wants to rule out a particular representational claim; namely, that the passions' impressional-components or emotions represent in virtue of resembling their causes. Hume therefore presents us with a two-part argument, the first part of which rejects the power of reason alone to cause passions, and the second part of which rejects the passions' ability to represent or copy any external objects that might have caused them.58 [End Page 631]

Hume's two-part argument makes sense if, as is likely, he is targeting Stoicism's two-part theory of the rational passions. Hume seems to have broken the Stoics' theory down into two claims: (a) rational passions are caused by impressions of some rational cosmic moral order, and (b) the rational passions represent that cosmic moral order in virtue of resembling the impressions that caused them. Hume believes that he can reject (a) with his arguments against the motivational inertness of reason alone. However, to complete his critique of the Stoics, Hume adds OE, which rejects (b) on the grounds that the impressional-component of a passion never resembles its causes.59

3. 3. Summary

In OE, Hume is attempting to diagnosis what is wrong with Stoicism's claim that rational passions convey representations of the cosmic moral order causing them. The argument of OE, which is in keeping with his Copy Principle and his general account of impressions and ideas, is that the impressions of a passion are "original existences" that do not copy anything. Reason has no traffic with the passions qua simple impressions (emotions) because they present no copy that reason could check against the original—the passions qua simple impressions are the original. In sum, Hume's initially bewildering statement that the passions are "original existences" not only follows from the application to the passions of his general epistemic commitments, but it is also key to his rejection of the Stoic theory of rational passions.

4. metaethical implications

The inadequacies of the standard reading of OE suggest caution about the metaethical commitments that many authors take Hume to broadcast in that passage. In this section, I briefly argue that my more accurate reading of OE does not support the standard reading of Hume as a kind of moral non-cognitivist who denies any role to reason in ethical thought or action, as such. I also suggest that Hume's preoccupation with Stoicism deflates the non-cognitivist bite of the neighboring passage in which he proclaims that reason is "the slave of the passions." [End Page 632]

4. 1. OE and Moral (Non-) Cognitivism

Standard readers hold that OE is key to Hume's Mutual Exclusivity Argument: Reasoning is the checking of mental representations for truth or falsity, but the passions are non-representational "original existences," therefore reason and the passions have, as Henry Allison puts it, "no point of contact."60 Putting this apparent conclusion of the Mutual Exclusivity Argument together with his Argument from Inertness, on which reason alone can never motivate us, standard readers find Hume committed to the moral non-cognitivist view that our action and decision-making is in some deep sense irrational, because it is motivated by these passions or "original existences" that are immune to rational evaluation.

The problem with the standard reading is that it fundamentally misreads Hume by equating his target in OE with a general position called cognitivism in ethics, and consequently misses his more restricted intentions. As my historically-contextualized reading of OE reveals, he is targeting a particular Stoic claim about what the rationality of our passions must consist in—namely, the representation of moral truths afforded by the passions' somehow resembling those impressions to be had from the point of view of the divine administrator of the cosmos. Yet, Hume's account of the passions as compounds of impressions and ideas allows him to reject that particular representational story without prejudice to other ways in which the passions might represent. Most Humean passions can represent (even in virtue of resemblance) insofar as they are also part ideas, and there is no obvious barrier to the impressional-components of the passions representing in ways other than resemblance.

Although I cannot here conclusively demonstrate that Hume allows for the passions to represent in these other ways, my interpretation is consistent with a number of other passages that conflict, however, with the standard reading. For instance, in the Abstract that serves as his own précis of the Treatise, Hume writes, "it is by means of thought only that any thing operates upon our passions" (T Abs. 35/SBN 662). Hume frequently uses 'thought' as a synonym for 'reason,' which makes it hard to see how he could make this claim given the standard reading of OE as establishing the mutual exclusivity of reason and the passions.61 Either the standard reading is wrong about OE, or Hume is wrong about his own commitments. As I have argued, the apparent meaning of OE shifts when we return it to its proper historical context, and therefore the mistake lies with the standard reading. My historically contextualized reading is thus congenial toward a growing acknowledgment of Hume's broadly cognitivist view that "belief is almost absolutely requisite to the exciting [of] our passions" (T 120).62 Specifically, my interpretation of OE as a rejection of the Stoics' Rational Passions Thesis—rather than of moral cognitivism in general—removes a longstanding [End Page 633] obstacle to recognizing his insistence that rational beliefs can influence our motivating passions.63

4. 2. Stoicism and Reason's Enslavement to the Passions

If I am right that OE does not commit Hume to moral non-cognitivism, then the redoubt of non-cognitivist interpretations of Hume is surely his dramatic claim that "reason is, and ought only to be the slave of the passions" (hereafter, RSP) (T 415). It seems that there is only one possible reading of RSP: We are governed by the sensitive or emotional side of our nature, not the cognitive. However, my focus on OE does, I think, suggest a different reading.

RSP and OE are, in fact, neighboring passages: The text runs from RSP almost immediately into OE. The significance of this fact becomes clearer if we attend to the organization of this stretch of text. Hume offers a two-part argument against Stoicism's two-part Rational Passions Thesis. He devotes one paragraph to each part of his argument. The first paragraph targets the Stoics' causal claim; RSP appears toward the end of that paragraph. After RSP, the first paragraph ends with the sentence in which Hume explains that he needs to round out his argument "by some other considerations." The second paragraph is devoted to OE, which targets the Stoics' representational claim. RSP therefore links the two halves of Hume's argument against Stoicism's Rational Passions Thesis. Given RSP's place in this textual schema, my suggestion is that what appears to be RSP's declaration of non-cognitivism is in fact Hume's hyperbolic inversion of the rhetoric of the Stoics, who held the view that the passions ought to be the slaves of reason.64

Historical precedents provide some support for my suggestion. Hume would have been familiar with Bayle's earlier sly inversion of Stoic rhetoric concerning the relationship between reason and the passions. In Remark H to his article on [End Page 634] Ovid, Bayle names the "Stoicks" as among those "'philosophers… [who] were unacquainted with the disposition of the springs which put the heart of [humans] in motion, and had no knowledge, no suspicion of the change wrought upon [post-lapsarian humans], by which [their] reason became a slave to [their] passions'" (Dictionary, 4:441).65 Bayle is mocking the Stoics' view that reason enjoys a relation of dominance over our passions in two respects: first, they hold that reason alone can cause passions; second, they hold that reason has the competency to judge whether the impressional-component of a passion represents its (cosmic rational) cause in virtue of resembling it. Stoics such as Cicero use the rhetoric of masters and slaves to express reason's domination of the passions, as in the following passage: "It is [a person's] duty to enable reason to have rule over that [yielding] part of the soul which ought to obey. How is that to be done?[,] you will say. Even as the master over the slave" (Tusculan Disputations, II.xx.47–xi.48). As Cicero sees it, if we harness our affectivity to the yoke of reason, then correct judgments of value will inform our actions and we shall live well. Bayle's reversal of the roles of reason and the passions re-appropriates Cicero's imagery in order to critique its Stoic picture of moral psychology.

My interpretive suggestion, therefore, is that Hume's placement of RSP at the crux of his critique of Stoicism's Rational Passions Thesis indicates that he is following Bayle's lead in offering a rhetorical inversion of the picture presented by Stoic moral psychology. In Hume's order of presentation, after discrediting the Stoic claim that reason alone can cause passions, he echoes Bayle in declaring that "reason is, and ought only to be the slave of the passions, and can never pretend to any other office than to serve and obey them." On my reading, Hume employs the rhetoric of RSP to draw attention to the fact that he has just discredited one of the vaunted offices ascribed to reason by the Stoics, for it is impossible that reason alone can serve as sire of the passions. However, the Stoics also think that reason has another office from which to exercise domination over the passions. It is to block that claim that Hume moves to "confirm" his argument with OE, which aims to show that reason alone cannot exercise the office of epistemic judge, at least not in the way that the Stoics envisioned judging the accuracy of the passions' representations.

Admittedly, the Stoics were not the only philosophers that Hume (or Bayle) would have recognized as recommending the passions' subordination to reason, for that recommendation was widespread: It is present in Clarke and Wollaston, as well as in Augustine, Calvin, and a number of Hume's now obscure British and French contemporaries.66 What was not widespread, however, was the causal claim that reason has the power to produce passions of its own. While I cannot therefore [End Page 635] rule out the possibility that other views besides the Stoics' come under the critical orbit of RSP, the Stoics surely do, and there is historical precedent in Bayle for contextualizing RSP as a kind of ad hominem against the Stoics. On my reading, consequently, in RSP Hume is saying that the passions are subordinate to reason in neither of the ways Stoicism predicts. My reading of RSP is therefore deflationary in the following sense: If RSP is accurately contextualized as a rhetorical strike at the Stoics' particular conception of the relationship between reason and the passions, then we should not assume that RSP provides us with either an accurate picture of Hume's own moral psychology or evidence of his animosity to moral cognitivism as such.

4. 3. Summary

We run the risk of fundamentally misreading Hume if we conflate his particular arguments against Stoicism with a general attack on "cognitivism" in ethics. When read in context, what the two passages regularly cited from Treatise 2.3.3 as evidence of Hume's non-cognitivism actually reveal is a more limited moment in the development of his views about Stoicism. The advantage of my reading is that it renders OE (and RSP) consistent with Hume's oft-stated, if routinely overlooked, commitment to the rational influence of belief over the passions; the standard reading of OE conflicts with that commitment.


The general point that emerges from my inquiry is that when we place OE back in its proper historical context, we find an interesting reading of Hume that we have missed. Far from being the centerpiece of a program that denies any role to reason in thought or action as such, OE is an integral part of Hume's rejection of Stoicism's Rational Passions Thesis. Placing OE against this Stoic backdrop yields three important interpretive gains: (i) it unlocks Hume's hitherto puzzling conception of (indirect) passions as compounded effects of impressions and ideas, (ii) it clarifies that Hume's ban on the passions' representational capacity is only limited to their impressional- components' (emotions') capacity to represent their causes in virtue of resemblance, and (iii) it restores the lost Stoic focus of key passages that have been mistakenly interpreted as salvos on behalf of a sweeping non-cognitivist agenda in ethics. In light of my more accurate reading of OE, the larger question most worth pursuing is whether the ethical tradition that bears Hume's name can still credibly cite him as an ally.67 [End Page 636]

Jason R. Fisette

Jason R. Fisette is Lecturer of Philosophy at the University of Nevada, Reno.

bibliography and abbreviations

Alanen, Lili. "The Powers and Mechanisms of the Passions." In Traiger, The Blackwell Guide to Hume's Treatise, 179–98. ["Powers and Mechanisms"]
Allison, Henry. Custom and Reason in Hume: A Kantian Reading of the First Book of the "Treatise." Oxford: Clarendon Press, 2008. [Custom and Reason]
Anderson, Robert. Hume's First Principles. Lincoln: University of Nebraska Press, 1966.
Augustine. The City of God against the Pagans. Edited and translated by R. W. Dyson. Cambridge: Cambridge University Press, 1999. [City of God]
Baier, Annette. Death and Character: Further Reflections on Hume. Cambridge: Harvard University Press, 2008. [Death and Character]
———. A Progress of Sentiments. Cambridge: Harvard University Press, 1991.
Bayle, Pierre. Historical and Critical Dictionary, 4th ed. 5 vols. Translated by P. des Maizeaux. New York: Garland Inc., 1984. [Dictionary]
Beauchamp, Tom. "Editor's Annotations." In Hume, A Dissertation on the Passions.
Berkeley, George. The Works of George Berkeley Bishop of Cloyne. 9 vols. Edited by A. A. Luce and T. E. Jessup. London: Thomas Nelson and Sons, 1949. [Works]
Betegh, Gábor. "Cosmological Ethics in the Timaeus and Early Stoicism." In Oxford Studies in Ancient Philosophy, vol. XXIV, edited by David Sedley, 273–302. Oxford: Oxford University Press, 2003. ["Cosmological Ethics"]
Blundell, Mary Whitlock. "Parental Nature and Stoic Oikeiosis." Ancient Philosophy 10 (1990): 221–42. ["Parental Nature"]
Boethius. The Consolation of Philosophy. Translated by David R. Slavitt. Cambridge: Harvard University Press, 2008. [Consolation]
Box, M. A. The Suasive Art of David Hume. Princeton: Princeton University Press, 1990.
Bricke, John. Mind and Morality: An Examination of Hume's Moral Psychology. Oxford: Oxford University Press, 1996. [Mind and Morality]
Brown, Charlotte. "Is Hume an Internalist?" Journal of the History of Philosophy 26 (1988): 68–87.
Calvin, John. Institutes of the Christian Religion. Edited by John T. McNeill and translated by Ford Lewis Battles. Philadelphia: The Westminster Press, 1960. [Institutes]
Capaldi, Nicholas. Hume's Place in Moral Philosophy. New York: Peter Lang, 1989.
Cicero. Tusculan Disputations. Translated by J. E. King. London: William Heinemann, 1927.
Clarke, Samuel. A Discourse of Natural Religion. In Raphael, The British Moralists, 189–225. [DNR]
Cohon, Rachel. Hume's Morality: Feeling and Fabrication. Oxford: Oxford University Press, 2008. [Hume's Morality]
———, and David Owen. "Hume on Representation, Reason and Motivation." Manuscrito 20 (1997): 47–76. ["Hume on Representation"]
Collier, Mark. "Hume's Science of Emotions: Feeling Theory without Tears." Hume Studies 37 (2011): 3–18. ["Hume's Science"]
Dancy, Jonathan. Practical Reality. Oxford: Oxford University Press, 2000.
Davidson, Donald. "Actions, Reasons and Causes." In Essays on Action and Events, 3–19. Oxford: Clarendon Press, 1980.
Descartes, René. Oeuvres des Descartes. Edited by Charles Adam and Paul Tannery. 11 vols. Paris: J. Vrin, 1897–1913. [AT]
———. The Philosophical Writings of Descartes. Translated by John Cottingham, Robert Stoothoff, and Dugald Murdoch. 2 vols. Cambridge: Cambridge University Press, 1985. [CSM]
———. The Philosophical Writings of Descartes. Volume 3 : The Correspondence. Translated by John Cottingham, Robert Stoothoff, Dugald Murdoch, and Anthony Kenny. Cambridge: Cambridge University Press, 1991. [CSMK]
Diogenes Laertius. Lives of Eminent Philosophers. 2 vols. Translated by R. D. Hicks. Cambridge: Harvard University Press, 1965. [Lives]
Epictetus. The Discourses as reported by Arian. 2 vols. Translated by W. A. Oldfather. Cambridge: Harvard University Press, 1966. [Discourses]
Fisette, Jason. "Hume on the Lockean Metaphysics of Secondary Qualities." Hume Studies 40 (2014): 95–136. ["Lockean Metaphysics"]
Fogelin, Robert. Hume's Skepticism in the Treatise of Human Nature. London: Routledge, 1985. [Hume's Skepticism]
Frede, Michael. "The Stoic Doctrine of the Affections of the Soul." In The Norms of Nature: Studies in Hellenistic Ethics, edited by Malcolm Schofield and Gisela Striker, 93–110. Cambridge: Cambridge University Press, 1986. ["The Stoic Doctrine"] [End Page 637]
Garrett, Don. Cognition and Commitment in Hume's Philosophy. Oxford: Oxford University Press, 1997. [Cognition and Commitment]
———. "Hume's Naturalistic Theory of Representation." Synthese 152 (2006): 301–19.
Gauthier, David. "Reason and Maximization." Canadian Journal of Philosophy 4 (1975): 411–34.
Hampton, Jean. "Does Hume Have an Instrumental Conception of Practical Reason?" Hume Studies 21 (1995): 57–74. ["Practical Reason"]
Harris, James. Hume: An Intellectual Biography. New York: Cambridge University Press, 2015.
Hume, David. Enquiries concerning Human Understanding and concerning the Principles of Morals. Edited by L. A. Selby-Bigge. 3rd ed. by P. H. Nidditch. Oxford: Clarendon Press, 2006. [SBN]
———. An Enquiry concerning Human Understanding. Edited by Tom L. Beauchamp. Oxford: Clarendon Press, 2006. [EHU]
———. An Enquiry concerning the Principles of Morals. Edited by Tom L. Beauchamp. New York: Oxford University Press, 2003. [EPM]
———. Essays Moral, Political, and Literary. Edited by Eugene F. Miller. Indianapolis: Liberty Fund, 1985. [EMPL]
———. "Hume's Early Memoranda, 1729–1740: The Complete Text." Edited by Ernest Campbell Mossner. Journal of the History of Ideas 9 (1948): 492–518. [Memoranda]
———. A Letter from a Gentleman to his Friend in Edinburgh. Edited by John. V. Price. Edinburgh: University of Edinburgh Press, 1967. [Gentleman]
———. The Letters of David Hume. 2 vols. Edited by J. Y. T. Grieg. Oxford: Clarendon Press, 1932. [Letters]
———. A Treatise of Human Nature. Edited by David Fate Norton and Mary J. Norton. Oxford: Oxford University Press, 2005. [T]
———. A Treatise of Human Nature. Edited by L. A. Selby-Bigge. 2nd ed. by P. H. Nidditch. Oxford: Clarendon Press, 2009. [SBN]
Hursthouse, Rosalind. "Hume: Moral and Political Philosophy." In British Philosophy and the Age of Enlightenment, edited by Stuart Brown, 179–202. London: Routledge, 1996.
Inoue, Haruko. "The Origin of the Indirect Passions in the Treatise: An Analogy Between Books 1 and 2." Hume Studies 29 (2003): 205–22. ["The Origin of the Indirect Passions"]
Karlsson, Mikael M. "Reason, Passion, and the Influencing Motives of the Will." In Traiger, The Blackwell Guide to Hume's Treatise, 235–55. ["Reason, Passion"]
Kenny, Anthony. Action, Emotion, and Will. London: Routledge & Kegan Paul, 1963.
Kosgaard, Christine. "Skepticism about Practical Reason." In Creating the Kingdom of Ends, 311–34. New York: Cambridge University Press, 1996.
Kydd, Rachael. Reason and Conduct in Hume's Treatise. London: Oxford University Press, 1946.
Lapidge, Michael. "Stoic Cosmology." In Rist, The Stoics, 161–85.
Locke, John. An Essay Concerning Human Understanding. Edited by Peter Nidditch. Oxford: Clarendon Press, 1979. [ECHU]
Long, A. A. "Representation and the Self in Stoicism." In Psychology, edited by Stephen Everson, 102-20. Cambridge: Cambridge University Press, 1991. ["Representation"]
Long, A. A., and D. N. Sedley, eds. and trans. The Hellenistic Philosophers, vol. 1. London: Cambridge University Press, 1987. [LS].
Mackie, J. L. Ethics: Inventing Right and Wrong. London: Penguin, 1990. [Ethics]
———. Hume's Moral Theory. London: Routledge & Kegan Paul, 1980.
McDowell, John. "Non-Cognitivism and Rule-Following." In Mind, Value, and Reality, 198–218. Cambridge: Harvard University Press, 1998.
McIntyre, Jane. "Hume's 'New and Extraordinary' Account of the Passions." In Traiger, The Blackwell Guide to Hume's Treatise, 199–215. ["Account of the Passions"]
Miller, Jon Charles. "Hume's Citation of Strabo and the Dating of the Memoranda." Hume Studies 39 (2013): 197–202. ["Hume's Citation"]
Millgram, Elijah. "Was Hume a Humean?" Hume Studies 21 (1995): 75–93.
Norton, David Fate, and Mary J. Norton. The David Hume Library. Aberystwyth: Edinburgh Bibliographical Society, 1996.
Norton, David Fate, and Jacqueline Taylor, eds. The Cambridge Companion to Hume. Cambridge: Cambridge University Press, 2009.
Owen, David. "Hume on the Mechanics of Mind: Impressions, Ideas, and Associations." In Norton and Taylor, The Cambridge Companion to Hume, 70–104. ["Mechanics of Mind"]
Parfit, Derek. Reasons and Persons. Oxford: Oxford University Press, 1987.
Penelhum, Terence. David Hume: An Introduction to His Philosophical System. West Lafayette: Purdue University Press, 1992. [David Hume] [End Page 638]
———. "Hume's Moral Psychology." In Norton and Taylor, The Cambridge Companion to Hume, 238–69.
Perrson, Ingmar. "Hume—Not a 'Humean' about Motivation." History of Philosophy Quarterly 14 (1997): 189–206. ["Hume—Not a 'Humean'"]
Platts, Mark. "Moral Reality and the End of Desire." In Reference, Truth and Reality, edited by Mark Platts, 69–82. Oxford: Basil Blackwell, 1981. ["Moral Reality"]
Radcliffe, Elizabeth. "Hume on the Generation of Motives: Why Beliefs Alone Never Motivate." Hume Studies 25 (1999): 101–22. ["Hume on the Generation of Motives"]
———. "Hume on the Passions and 'Original Existences.'" In Humean Readings, edited by Lorenzo Greco and Alessio Vaccari, 211–31. Rome: Storia e Litteratura, 2012.
Raphael, D. D., ed. The British Moralists: 1650-1800, vol. 1. Indianapolis: Hackett Publishing Company, 1991.
———. "Hume's Critique of Ethical Rationalism." In Hume and the Enlightenment: Essays presented to Ernest Campbell Mossner, edited by W. Todd, 14–29. Edinburgh: University of Edinburgh Press, 1974.
Rist, John, ed. The Stoics. Berkeley: University of California Press, 1978.
Scanlon, T. M. What We Owe to Each Other. Cambridge: Harvard University Press, 1998.
Schroeder, Mark. Slaves of the Passions. Oxford: Oxford University Press, 2007.
Sellars, John. Stoicism. Berkeley: University of California Press, 2006.
Shaftesbury, Anthony Ashley Cooper. Characteristics of Men, Manners, Opinions, Times. Edited by Lawrence E. Klein. Cambridge: Cambridge University Press, 1999. [Characteristics]
Smith, Michael. The Moral Problem. Cornwall: Wiley-Blackwell, 1994.
Snare, Francis. Morals, Motivation and Convention: Hume's Influential Doctrines. New York: Cambridge University Press, 1991. [Morals, Motivation and Convention]
Stewart, M. A. "The Dating of Hume's Manuscripts." In The Scottish Enlightenment: Essays in Reinterpretation, edited by P. Wood, 265–314. Rochester: University of Rochester Press, 2000.
Strange, Steven K. "The Stoics on the Voluntariness of the Passions." In Stoicism: Traditions and Transformations, edited by Steven K. Strange and Jack Zupko, 32–51. New York: Cambridge University Press, 2004. ["The Voluntariness of the Passions"]
Stroud, Barry. Hume. London: Routledge, 1995.
Todd, Robert B. "Monism and Immanence: The Foundations of Stoic Physics." In Rist, The Stoics, 137–60. ["Monism and Immanence"]
Traiger, Saul, ed. The Blackwell Guide to Hume's Treatise. Malden: Wiley-Blackwell, 2005.
Tweyman, Stanley. Reason and Conduct in Hume and His Predecessors. The Hague: Martinus Nijhoff, 1974. [Reason and Conduct]
Wollaston, William. The Religion of Nature Delineated. In Raphael, The British Moralists, 239–58. [RND]
Wright, John P. Hume's 'A Treatise of Human Nature': An Introduction. New York: Cambridge University Press, 2009. [Hume's 'A Treatise'] [End Page 639]


1. I cite Hume's works as follows. References to 'T' are to A Treatise of Human Nature, followed by book, part, section, and, where appropriate, paragraph numbers from the 2005 Norton and Norton edition (references to paragraphs of the Appendix to the Treatise are preceded by 'App'). The corresponding page numbers in the 2009 Selby-Bigge edition, revised by Nidditch (SBN), follows each of these citations. References to 'EHU' are to An Enquiry concerning Human Understanding, followed by section and paragraph numbers from the 2006 Beauchamp edition, and the corresponding page numbers in the 2006 Selby-Bigge edition, revised by Nidditch (SBN). References to EPM are to An Enquiry concerning the Principles of Morals, followed by section and paragraph numbers from the 2003 Beauchamp edition, and the corresponding page numbers in the 2006 Selby-Bigge edition, revised by Nidditch (SBN).

3. I am indebted to Zed Adams for the following way of putting things.

6. For assessments of this so-called Humean theory of motivation as a matter of philosophical psychology, see Baier, Death and Character, 6; Bricke, Mind and Morality; Jonathan Dancy, Practical Morality, 10; Donald Davidson, "Actions, Reasons and Causes"; David Gauthier, "Reason and Maximization"; Mark Platts, "Moral Reality"; and Smith, The Moral Problem, 92–129.

10. See also T 78, and T 278.

13. Commentators disagree as to whether Hume's Argument from Inertness provides for any kind of practical reasoning. See Hampton, "Practical Reason"; Korsgaard, "Skepticism about Practical Reason"; Millgram, "Was Hume a Humean?"; Derek Parfit, Reasons and Persons, 120; Penelhum, "Hume's Moral Psychology," 251; T. M. Scanlon, What We Owe to Each Other, 19–20; Schroeder, Slave of the Passions; and Stroud, Hume, 155. However, the debate is not directly relevant to my argument.

14. Cohon's Hume's Morality, ch. 3, is a notable exception, as is her co-authored article with David Owen, "Hume on Representation." Other commentators recognizing this as a two-part argument include: Korsgaard, "Skepticism about Practical Reason," 313; Ingmar Perrson, "Hume—Not a 'Humean,'" 190; and Radcliffe, "Hume on the Generation of Motives," 103.

18. However, Cohon has noted (in correspondence) that it is unclear whether Clarke is saying that the passions themselves are unreasonable or merely that such passions make the will influenced by them unreasonable.

19. In the second Enquiry, Hume includes Clarke in a list of moral philosophers whose theorizing "excludes all sentiment, and pretends to found everything on reason," which would also seem to rule out rational passions (EPM 3.34/SBN 197 n. 1).

20. Radcliffe argues that Hume may be targeting Descartes's and Leibniz's conception of the passions. However, Radcliffe herself appears to concede that neither believes that reason alone can cause passions, and indeed, in a letter to Elisabeth, dated May 21, 1643, Descartes says that it is "the body… acting upon the soul [that] caus[es] its feelings and passions" (AT III.665/CSMK 218). Radcliffe's suggestion then leaves unanswered the question of whom Hume is addressing in this stretch of Treatise 2.3.3. See her "Hume and the Passions as 'Original Existences,'" 216 n. 17. Jean-Luc Solère has suggested to me (in conversation) that Descartes does allow for the existence of passions caused by the soul or mind or reason alone. However, I remain unconvinced by the passages in Passions to which Solère points. As I understand Descartes's views there, he does allow that, in a loose sense of the term (see AT XI.347–48/CSM 337–38, sect. 25), we can say that there are passions (such as intellectual joy) that, qua perceptions of volition, are caused by the soul etc. (see AT XI.396–97/CSM 360–61, sect. 91; AT XI.440–42/CSM 381–82, sect. 147–48; AT XI.488/CSM 404, sect. 212). However, it also seems that Descartes does not think these are passions "strictly speaking," because passions caused by the soul lack the sensible or neural agitation that characterizes passions caused by the body. It is not improbable that these passions caused by the soul may therefore be better classified as affects, in the technical sense of 'affects' deployed in The Principles of Philosophy as a mode of substance (see AT VIIIA.22–23/CSM 208, sect. 48, and AT VIIIA.26/CSM 211, sect. 56).

21. For the image of reason as fire or breath, see the remarks of Aristocles, Aetius, Galen, Alexander, and Cicero at LS 273, 274, 282, and 323, respectively.

24. I cite Diogenes Laertius as follows: References to 'Lives' are to Lives of Eminent Philosophers, translated by R. D. Hicks, followed by volume and page number.

25. The Stoic view allows for passions to be irrational in an epistemic, not a causal, sense. Roughly, their view is that pathe, i.e. passions caused by an indirect manifestation of divine reason (i.e. through the logos immanent in bodily impulses) routinely introduce epistemic error because of their selfish perspective. For discussion, see Gábor Betegh, "Cosmological Ethics," 295–99, and Mary Whitlock Blundell, "Parental Nature," 230–34.

26. Galen writes: "For in defining distress, he [Chrysippus] says that it is 'a shrinking at what is thought to be something to avoid,' and he says pleasure is 'a swelling up at what is thought to be something to pursue'" (LS 412).

27. Seneca states that "impulse never exists without the mind's assent" (LS 419), and Diogenes Laertius reports that "They [the Stoics] hold the emotions to be judgements" (Lives, 2.218).

28. Sextus Empiricus provides a nearly identical formulation; see LS 243.

30. Following the translation of Lives at LS 346.

31. Following the translation of Lives at LS 412.

33. Hume, Enquiries concerning Human Understanding and concerning the Principles of Morals, 2. Hume's dating may be an exaggeration, however, given that he is trying to downplay the Treatise as a youthful "error."

34. John P. Wright argues that a reference to Shaftesbury appears in a letter that Hume wrote to his friend Michael Ramsay in 1727. See Wright's Hume's 'A Treatise, ' 14, and Letters 1.10.

35. I retain Shaftesbury's italicization of the Stoic works he is citing; the citations and non-italicized remarks are Shaftesbury's own.

36. I am grateful to David Owen for urging me to develop this and the following historical point.

38. See Discourses III.xxiv.86–88. See also III.xxiv.54–59.

40. Hume is emphatic that ideas always represent their impressions (T 37; and T 157).

41. See also T 38; EHU 12.9/SBN 152; and possibly T 317–18.

42. The point about the representational capacity of simple impressions of sensation is controversial. For a rejection of that claim, see Cohon and Owen, "Hume on Representation"; for a defense, see Garrett, "Hume's Naturalist Theory of Representation." I should note that Garrett agrees with Cohon and Owen, however, in thinking that the passions are non-representational impressions of reflection, so my claim is a departure from both. My thanks to Don Garrett, David Owen, and John Wright for pressing me on this point.

43. Impressions of reflection comprise, in addition to the "passions, desires, and emotions" (T 8), awareness of mind's operations in, say, determining causal associations (T 165–66), experience of "beauty and deformity" (T 276), and consciousness of will or volition (T 399). I confine myself to the passions and the emotions.

44. See Locke, ECHU II.31.14 and Berkeley, The Principles of Human Knowledge, sect. 8, in Works 2:44. References to Locke's Essay are by book, chapter, and section numbers.

45. I pass over the issue of direct passions, which Hume believes can arise "immediately" or without the "interposition" of any idea. Hume calls the indirect passions "secondary" at T 275, and "compound" at, e.g. T 394. Hume seems to mean by "compound" what he meant (in Book I) by "complex"; i.e. potentially analytically separable.

46. Readings of the Humean passions as somehow complex or compound are also suggested by Alanen, "Powers and Mechanisms," 186; Baier, A Progress of Sentiments, ch. 7; and Haruko Inoue, "The Origin of the Indirect Passions."

47. Donald Ainslie and Don Garrett have both suggested (in conversation) that these two references to "component parts" are not on all fours, for in the context of the seventh rule Hume is speaking of material or quantitative complexity, whereas in the context of the passions he is speaking of analytical or qualitative complexity. However, Hume speaks of the material complexity of the passions on several occasions, especially when he is appealing to the "animal spirits," which Tom Beauchamp characterizes as "nerve fluid[s] or humour[s] inside narrow tubes and pores that mak[e] sensation and voluntary motion possible" ("Editor's Annotations," 110). Hume conceives of the animal spirits as spatially located in different "cells" (T 60–61) and cells are presumably materially or quantitatively complex. Hume invokes movement in the parts of the animal spirits to explain the complexities of the passions, which suggests that the passions are both materially and analytically complex (T–11/SBN 185; T 318; T 325–26; and T 420). To be sure, Hume is hesitant about invoking the explanatory power of the animal spirits (T–2/SBN 275–76), but he also says that it is "specious [i.e. attractive] and plausible" to do so (T 60–61), leading some commentators to surmise that he "avoids offering… [such] explanations… until he finds he cannot make his meaning clear without it" (see Robert Anderson, Hume's First Principles, 122; cf. Wright, Hume's A Treatise of Human Nature: An Introduction, 52 and n. 29).

48. See also T 174.

49. I am assuming that 'compound' and 'complex' are interchangeable, at least for my purposes. And this is a safe assumption, for Hume defines complexity as what "may be distinguished into parts" (T 2), which is precisely how he talks about something that is compound (what is compound has "component parts" or "a number of parts").

50. For the association of ideas, see T 279; and T 285. For the association of impressions, see T 285.

51. See T 141; T 347; T 380.

52. For instance, Mark Collier identifies emotions and passions without discussion in his "Hume's Science." Two commentators more alert to the possible distinction are Baier, A Progress of Sentiments, 164 and 310 n. 9, and Owen, "Mechanics of Mind," 85. Norton and Norton provide a separate entry for the term 'emotion' in the glossary of their edition of the Treatise, but their gloss leaves the question of an emotion's relationship to passions unresolved (575).

53. There are several examples in the Treatise of Hume appearing to regard 'emotions' and 'passions' as interchangeable terms. For the most salient examples, see T 7–8; T 275; and T 421.

54. The Treatise contains numerous examples of Hume using 'emotion' in the technical sense of the impressional component of a passion. For the most salient examples, see T 85–86, 628; T 630–31; T 283; T–8/SBN 286–88; T 305–6; T 350; T 373–74; T–9/SBN 417–18; T 418–19; T 422–23; T 437–38; and T 442–43.

55. Whether the direct passions are simple or complex impressions of reflection is unclear. If complexity is tokened by an idea being one of the component causes of a passion, then the following statement may imply that the direct passions are simple inasmuch as they are caused by brute physiological processes: "The impressions, which arise from good and evil most naturally, and with the least preparation are the direct passions of desire and aversion… hope and fear.… The mind by anoriginal instinct tends to unite itself with the good, and to avoid the evil.… Beside good and evil, or in other words, pain and pleasure, the direct passions frequently arise from a natural impulse or instinct, which is perfectly unaccountable" (T, 8/SBN 438–39). However, when Hume first names those very passions, he explains them as being caused by ideas of pleasure and pain, which, in keeping with the interpretation I have suggested, would seem to make ideas component causal parts of the direct passions: the direct passions arise, he says, when the "idea of pleasure or pain… returns upon the soul, produc[ing] the new impressions of desire and aversion, hope and fear, which may properly be called impressions of reflexion" (T 8). I suspect that at least some direct passions (those arising not from our ideas of pain or pleasure, but from instincts) are simple impressions of reflection, but I cannot resolve the issue here.

56. Hume classifies "thirst" as an impression of sensation (T 8). Height is a more difficult case, given that height or extension is a primary quality, our ideas of which modern philosophers such as Locke hold to resemble the qualities in external bodies causing those ideas in us. Although I cannot argue for the point here, Hume rejects the view that our perceptions of extension resemble their causes, which would make his invocation of height concordant with anger, thirst, and sickness insofar as these examples designedly rebuke a causal-resemblance theory of representation. For defense of the claim about Hume and primary qualities, see Jason Fisette, "Lockean Metaphysics."

57. Hume acknowledges this narrow sense of "passion" at least twice. First, he claims that there can be passions that are calm or exist "without producing any sensible emotion" and that we consequently conflate with reason (T 417). Second, he states that "What we commonly understand by passion is a violent and sensible emotion of mind" (T 437–38). Indeed, there is a sharp drop-off in the frequency of the word "emotion" from Book II (thirty-six instances) to Book III (two instances). That fact may indicate that as Hume leaves the technical discussion of the passions behind, he increasingly sets aside his wider sense of "passion" (which refers to the compound of hedonic impressions and ideas) and adopts the narrower sense of "passion" (which refers to their emotional or impressional-components) in keeping with the terminological convention of the vulgar. My thanks to Colin Chamberlain for help with this point.

58. In T, Hume invokes the passions' originality with respect to three items: other passions, volitions, and actions. The first two are consistent with a Stoic target. (i) Other passions: For Stoics, rational passion are copies of other passions; namely, those obtained from a cosmic or divine point of view. (ii) Volitions: Volition is "the internal impression we feel and are conscious of, when we knowingly give rise to any new motion of our body, or new perception of the mind" (T 399). Hume finds it helpful, if not entirely accurate, to classify volitions as among the direct passions (T–4/SBN 438–39), presumably because volitions are genetically similar to the direct passions in being impressions of reflection (or secondary impressions) that arise immediately from hedonic perceptions without the interposition of ideas of self or others (T 574). If volitions are "original" existences, and if volitions are akin to direct passions, then Hume is denying that passion-like states represent their causes in virtue of resembling them. Stoicism is again a good match for this remark about volition. 'Volition' is a term that Hume uses repeatedly (five times) with reference to the Stoics' causal claim about the rational passions in the paragraph that precedes the invocation of originality in T, and here in T the representational claim about originality is again preceded by a rejection of the causal claim that "reason alone" can "excite passions" (T–8/SBN 457–58). (iii) Actions: In the paragraphs preceding his declaration of actions' originality, Hume rejects the idea of actions caused by reason alone, which is a view held not by the Stoics, but by Clarke and Wollaston. I suspect that the invocation of "actions," along with "other passions and volitions," is an echo of the order of argument found in T 2.3.3, in which Hume first rejects the rationalists' motivational claims with his Argument from Inertness, and then targets the Stoics' causal and representational claims about the rational passions. My thanks to an anonymous referee for urging me to clarify Hume's references to these three items.

59. In addition to his rejection of the Stoic's specific representational claim, Hume is also skeptical—on more general epistemic grounds—of the very idea of the rational cosmic moral order presupposed by Stoicism's Rational Passions Thesis. In the Enquiry Concerning Human Understanding, for instance, Hume takes issue with the "ancient Stoic" view that "the WHOLE [of nature], considered as one system, is… ordered with perfect benevolence," remarking that such a view is premised on "remote and uncertain speculations" (EHU 8.34–36/SBN 101–3). My thanks to an anonymous referee for noting this point.

61. For example, see T 1–2; T 18–19; T 463; EHU 5.8/SBN 46–47; EHU 6.2/SBN 56–57; and EHU 7.7/SBN 63.

63. Hume's claim that the passions can represent and involve beliefs also bears on the Stoic doctrine of the pathe. The pathe are passions indirectly caused by the divine reason immanent in our bodies and are judgments of what appears harmful or beneficial. However, the Stoics typically regard the pathe (unlike the eupatheiai) as epistemically irrational insofar as they are reducible to false beliefs about what is really harmful or beneficial as judged from the cosmic or divine point of view (see LS 410–19). Hume's account of such passions differs from the Stoics' in being more optimistic about their potential for epistemic rationality or at least correctability. A number of the pathe are analogues of what Hume calls the indirect passions: anger, pride etc. Hume would seem to agree with the Stoics that these passions are representations of value, but disagree that these judgments are invariably false. For instance, while acknowledging that pride and humility can distort value insofar as they "judge of objects more from comparison than from their real and intrinsic value," he claims that through experience we can discover principles that "settl[e] the just value of every thing" (T 291; and T 294). Other pathe are akin to what Hume calls the direct passions: aversion, fear etc. Hume holds that the direct passions are responsive to the beliefs of probability afforded by our "understanding"; for instance, fear "gradually and insensibly decays" "according as the probability" of the feared event tapers off (T–12/SBN 440–41). Hume does not hold that the direct passions are compounds of impressions and ideas resulting from the double association of impressions and ideas, so the truth-apt beliefs of probability that influence the direct passions are presumably extrinsic to the passions themselves. However, given that on his view it can be rational to fear something, Hume would reject Andronicus's claim that "fear is an irrational shrinking, or avoidance of an expected danger" (LS 411). I am grateful to an anonymous referee for urging me to contrast Hume's views on the passions with the Stoics' views on the pathe.

64. For a similar treatment of Hume's rhetoric in RSP (minus the Stoic context), see M. A. Box, The Suasive Art of David Hume, 181–82.

65. Bayle names the Stoics as a fitting target for the passage that he is, in turn, quoting approvingly from Du Verdier.

66. Clarke and Wollaston speak of our passions as in need of a "bridle" (RND 253) to "govern" them (DNR 211). Clarke also explicitly warns that if reason does not govern, we shall become "enslaved" by our passions (DNR 191). For analogous passages in Augustine and Calvin, see The City of God against the Pagans, 450–51 (Bk. XI, ch. 2), 624 (Bk. XIV, ch. 23), and 626 (Bk. XIV, ch. 24), and Institutes of the Christian Religion, 289 (Bk. II, ch. 3, sect. 1), 292 (Bk. II, ch. 3, sect. 3), and 296 (Bk. II, ch. 3, sect. 5). Jane McIntyre collects similar passages from Hume's contemporaries in "Account of the Passions."

67. I am very grateful to the editors, Jack Zupko, Alan McLuckie, and Hank Southgate, as well as an anonymous referee, for their helpful comments and queries. I presented versions of this paper at the 42nd International Hume Society Conference at Stockholm University on July 21, 2015, the Scottish Seminar in Early Modern Philosophy at the University of St. Andrews on May 8, 2015, the Atlantic Canada Seminar in Early Modern Philosophy at Dalhousie University on July 10, 2014, and the New England Colloquium in Early Modern Philosophy at Brown University on May 10, 2014, and I am thankful for my audiences on those occasions for fruitful discussions. I especially benefited from feedback from the following individuals: Zed Adams, Donald Ainslie, Omri Boehm, Justin Broackes, Colin Chamberlain, Rachel Cohon, Alice Crary, Steve Daniel, Jeff Edwards, Don Garrett, Lorenzo Greco, Paul Guyer, James Harris, Yitzhak Melamed, Jeff McDonough, Alison McIntyre, Robert Miner, Alan Nelson, David Owen, Jean-Luc Solère, Jackie Taylor, Margaret Watkins, Christopher Williams, and John P. Wright.

Additional Information

Print ISSN
Launched on MUSE
Open Access
Back To Top

This website uses cookies to ensure you get the best experience on our website. Without cookies your experience may not be seamless.