Hume's Constitutivist Response to Scepticism

1. Introduction

In the concluding section of the Book One of A Treatise of Human Nature,¹ entitled “Conclusion of this Book” (henceforth, “Conclusion”), Hume confronts radical scepticism about the standards of correct reasoning. According to the naturalistic interpretations (e.g., Garrett 1997; Kemp Smith 2005), Hume resolves this scepticism by appealing to some psychological facts. Exactly what kind of psychological fact plays a part varies, according to different scholars. A common problem with these interpretations is what Ainslie (2015: 233) calls the “normativity problem” that the alleged naturalistic epistemic standards fail to accommodate their normative force (cf. Qu 2020: 147). In escaping his scepticism, Hume seems to express merely his psychological inclinations toward the use of reason. It remains unclear why this seemingly mere report of his psychology can provide a principled reason to overcome his scepticism.²

This paper seeks to supplement the naturalistic interpretations by focusing on the specific psychological item, the “indirect passions”, which few scholars have examined in this context. The indirect passions such as pride, humility, love, or hate are essentially directed at persons, and hence are closely related to Hume’s account of person. In this paper, I will argue that Hume’s discussions of the indirect passions and person provide a constitutivist ground for the naturalistic epistemic standards in the “Conclusion”. Being the object of the indirect passions constitutes what kind of person one is, and being the kind of person (philosopher in Hume’s case) gives non-optional reason to pursue certain kinds of reasoning.

2. The Naturalistic Interpretations of Hume’s Response to Scepticism

In this section, I give a brief summary of the scepticism in the “Conclusion” and describe the naturalistic interpretations of Hume’s response to it. Let me begin by recognizing that until Hume begins his sceptical discussions in T 1.4, he appears to accept the legitimacy of reason by default.³ For example, he lists “Rules by which to judge of causes and effects” (T 1.3.15) and considers these rules to be “all the LOGIC I think proper to employ in my reasoning” (T 1.3.15.11), which shows Hume’s endorsement of causal reasoning. Hume seems to think that demonstrative reasoning, which purports to discover relations of ideas, also has a high epistemic status, since it is the only reasoning that can produce something worthy of the name “knowledge” (T 1.3.11.2). Hume regards mental operations that produce irrational beliefs, such as credulity (T 1.3.9.12) and prejudice (T 1.3.13.12), as unjust, which also shows his endorsement of the legitimacy of reason. Hume seems to be committed to the default epistemic standard that “reason ought to be assented to” (Garrett 2015: 230).

However, in the “Conclusion”, Hume reminds himself of his discovery that causal reasoning is merely the propensity of the imagination to enliven an idea (cf. T 1.3.14.20) and this insight destabilizes his default trust in reason.⁴ The enlivening effect of the imagination is “so trivial” (T 1.4.7.3), “inconstant” (T 1.4.7.4), and does not teach us “the original and ultimate principle” (T 1.4.7.5) of objects. These considerations lead to what Hume calls “a very dangerous dilemma” (T 1.4.7.6). Hume says that “if we assent to every trivial suggestion of the fancy” on the grounds that reasoning and all other idiosyncratic operations of the imagination are on the same footing, this would “lead us into such errors, absurdities, and obscurities, that we must at last become asham’d of our credulity” (T 1.4.7.6).⁵ On the other hand, if we only follow stable operations of the mind, that is, “the general and more establish’d properties of the imagination” (T 1.4.7.6), this would result in “a total extinction of belief” (T 1.4.1.6), as was explained in the section, “Of Scepticism with Regard to Reason” (T 1.4.1). Given the fallibility of our cognitive faculties such that our reasoning sometimes errs, we are required to check the reliability of our judgment by forming a second-order judgment evaluating it. But then, we have to confirm the reliability of the second-order judgment by forming a third-order judgment. Thus, “we are oblig’d by our reason to add a new doubt”, and this endless addition of doubt ends up with “a total extinction of belief and evidence” (T 1.4.1.6). We can avoid this scepticism and have a belief “only by means of that singular and seemingly trivial property of the fancy” (T 1.4.7.6). This brings us back to the credulity. Thus, the dilemma is that we have “no choice left but betwixt a false reason and none at all” (T 1.4.7.6). Reason is more fragile than traditionally thought, and following the default principle that “reason ought to be assented to” leads to this destructive dilemma. This dilemma leads Hume to fall into “philosophical melancholy and delirium” (T 1.4.7.9).

However, this does not seem to be Hume’s final position. Amusement (e.g., backgammon) cures Hume’s melancholy, and he recollects himself and comes to have “indolent belief in the general maxims of the world” (T 1.4.7.10). But Hume is not satisfied with the indolence and returns to the use of reason. Here the naturalistic interpretations claim that Hume replaces the default attitude of endorsing reason with some naturalistic standards of correct reasoning, appealing to some psychological feature. For example, Garrett focuses on the following passage:

Where reason is lively, and mixes itself with some propensity, it ought to be assented to. Where it does not, it never can have any title to operate upon us. (T 1.4.7.11)

Garrett takes this as the “normative epistemic principle”, and names it the “Title Principle” (Garrett 1997: 234). For example, Hume seems to endorse reason accompanied by the passionate propensities of curiosity and ambition (T 1.4.7.12). As Qu (2020: 129–30) proposes, we can take this epistemic norm as a response to the dangerous dilemma: overly sceptical reasoning is neither sufficiently lively nor accompanied by motivating passions, and lively and active reasoning mingling with passions rejects mere fancy. The view that the Title Principle, or the passage in which the principle is claimed to be found, constitutes Hume’s epistemological response to scepticism has many advocates (e.g., Owen 1999: ch. 9; Ridge 2003; Allison 2008: ch. 12; Schafer 2014; Qu 2020: ch. 6).

One can take a naturalistic interpretation without committing to the Title Principle. For example, Kemp Smith thinks that, for Hume, the source of doxastic normativity is “natural belief,” something that our belief-forming processes inevitably incline us to have: “The beliefs which ought to be accepted are, [Hume] teaches, beliefs that Nature itself marks out for us. In their fundamental forms, as ‘natural’ beliefs, we have no choice but to accept them; they impose themselves on the mind” (Kemp Smith 2005: 388). Natural beliefs include beliefs about external objects and beliefs about causality. They have no rational basis, but are justified in that they are the product of our natural, irresistible imaginative tendencies.⁶ Kemp Smith does not examine the “Conclusion” in detail, but there Hume certainly seems to have in mind the weak psychological force of radical scepticism: “[v]ery refin’d reflections have little or no influence upon us” (T 1.4.7.7). His interpretation would say that in the “Conclusion”, it is ultimately nature that provides the normative distinction between the operations of the imagination.

3. The Normativity Problem

However, the naturalistic interpretations confront the normativity problem; it is not clear why we should follow these alleged standards. Several scholars have pointed out that the Title Principle lacks justification (Williams 2004: 269; Durland 2011: 83; Ainslie 2015: 233; Qu 2020: 147). As Qu puts it, Hume “helps himself to it [the title principle] without providing any good philosophical reasons for doing so” (Qu 2020: 147). It seems legitimate to require some ground for the norm, because its plausibility is by no means self-evident. Our psychological propensities can change from moment to moment. Just being in a good mood might enliven our reasoning. So it might seem to follow that the same reasoning could be justified at one time and not justified at another time, depending on one’s psychological states, which is implausible (Durland 2011: 80; Ainslie 2015: 233).

Donald Ainslie points out that Kemp Smith’s natural beliefs do not explain why they should be endorsed either (Ainslie 2015: 232). Kemp Smith’s interpretation, which emphasizes the irresistibility of natural beliefs, suggests the view that an inference or belief has (or loses) authority when we have no choice but to agree with it (or dismiss it). Ainslie states, “That would be a case not of our finding out what we should believe, but of our finding out what we do believe” (Ainslie 2015: 232, emphasis in original).⁷ Kemp Smith’s interpretation only states that we cannot help following various imaginative tendencies in various situations, and does not seem to tell us “why” we should follow them.

The problem is not only that without explaining its ground, Hume’s naturalistic epistemology is implausible in our contemporary eyes. As Qu points out (Qu 2020: 148–49), the lack of the ground for the epistemic norm must be also problematic from Hume’s own perspective. As a possible response to the dangerous dilemma, Hume considers the view that “no refin’d or elaborate reasoning is ever to be receiv’d” (T 1.4.7.7) so that we do not fall into the extreme scepticism. However, this proposal appeals to the “trivial” part of the imagination in order to avoid scepticism. Then, Hume claims, the proponent of this view “by a parity of reason must embrace all of them [conclusions of the fancy]” (T 1.4.7.7). Here Hume’s point seems to be that there is no principled ground for choosing the prohibition of refined reasoning over assenting to the fancy. Then, if there is no reason to adopt the alleged naturalistic standards, this seems problematic even by Hume’s own standards. To put it another way, without explaining how the psychological properties, to which the naturalistic interpretations appeal, are not “trivial,” the naturalistic interpretations fail to show how Hume can resolve the “dangerous dilemma” in the first place.

One might respond to the worry by adding a restriction to the “propensity” Hume appeals to in the “Conclusion”. For example, Hume raises curiosity as a passion that motivates him to use reason (T 1.4.7.12). From this, one may propose that Hume specifically endorses reason that satisfies curiosity.⁸ Curiosity, a distinctively epistemic passion, might play a privileged role in grounding the use of reason.⁹ However, Hume’s actual discussion of curiosity indicates that the passion is not rigid enough to resolve the dangerous dilemma. The object of curiosity is truth, but according to Hume, we do not pursue it as such. Rather, we pursue truth because of the pleasure that comes with the pursuit itself (T 2.3.10.3). Hume concedes that “[t]he truth we discover must also be of some importance” (T 2.3.10.4), but this is so “only because ’tis, in some measure, requisite to fix our attention” (T 2.3.10.6). As Gelfert notes, “[i]t is ultimately the pursuit that matters” (Gelfert 2013: 722). Thus, curiosity consists in the innocent pleasure of exercising one’s reason. Although this discussion is interesting on its own, it is not clear how it can address the dangerous dilemma. For the pleasure-seeking passion appears to be quite “trivial” (T 1.4.7.6). What prevents Hume from concluding that if we should follow this pleasure, then “by a parity of reason” (T 1.4.7.7) we should also follow all trivial propensities? Hume’s account of curiosity does not seem to answer this question.¹⁰

Another option for the naturalistic interpretation to respond to the concern about the ground is the kind of interpretation according to which, in the “Conclusion”, Hume provides a moral justification for the use of reason (e.g., Owen 1999: ch. 9; Ridge 2003; Sasser 2022: ch. 5). Hume characterizes virtue as a character trait that is agreeable to oneself or to others, or useful to oneself or to others (T 3.3.1.30). According to these criteria, the proponents of this view argue, the disposition to a certain kind of reasoning that Hume endorses in the “Conclusion” qualifies as virtuous. For example, Hume aims at reasoning that contributes to “the instruction of mankind” (T 1.4.7.12), that is, reasoning that is useful to society. Moreover, lively reasoning that satisfies the title principle is presumably agreeable to the reasoner. Thus, Hume’s rich theory of morality might provide a solid basis for his return to the use of reason. However, it has been pointed out that attributing moral justification for the use of reason to Hume comes with several interpretive costs. Here I describe two of them. First, Book Three of the Treatise, in which Hume develops his moral theory, was published separately from Book One and Book Two. This makes it unlikely that Hume appeals to morality at such a crucial moment of Book One (Garrett 2015: 232). In particular, it is not clear how Hume can move on to Book Two and use reason to explicate the passions, without having first justified reason. Second, the moral interpretation could collapse the distinction between epistemic normativity and moral normativity (Qu 2014), a distinction Hume seems committed to when saying, “Laudable or blameable . . . are not the same with reasonable or unreasonable” (T 3.1.1.10). Even if these costs do not totally undermine the moral interpretations, it would be good if we could offer an interpretation without these costs. In the following, I will present such an interpretation.

4. Why Indirect Passions?

Few scholars have attended to the indirect passions in the present context.¹¹ So, in this section, I will briefly motivate my turn to the indirect passions. As we saw above, the “Conclusion” has a negative phase that includes the “dangerous dilemma” and a positive phase where Hume appears to have overcome the scepticism and returned to the use of reason under some alternative epistemic standards. In both phases, the indirect passions occupy important places. In the negative phase, for example, his “desponding reflections” (T 1.4.7.1) resulting in the dangerous dilemma cause others’ “enmity”, “hatred”, “anger, calumny and detraction” (T 1.4.7.2). They are the indirect passions or intimately connected to them. The importance of the indirect passions in Hume’s sceptical discussions is also suggested in A Letter from a Gentleman to his Friend in Edinburgh. In response to the charge of “Universal Scepticism” (LG 14) for Hume’s discussion in T 1.4.7, Hume maintains that his purpose there is not to endorse the scepticism, but:

to abate the Pride of mere human Reasoners, by showing them, that even with regard to Principles which seem the clearest, and which they are necessitated from the strongest Instincts of Nature to embrace, they are not able to attain a full Consistence and absolute Certainty. Modesty then, and Humility, with regard to the Operations of our natural Faculties, is the Result of Scepticism. (LG 21, emphasis mine)

The point of the sceptical discussion is to make dogmatic thinkers less proud and more humble, by revealing that the reason they trust is more fragile than they think.¹² In the positive phase, Hume cites the passion of “ambition” of “contributing to the instruction of mankind” and “of acquiring a name by my inventions and discoveries” (T 1.4.7.12). Ambition is classified as an indirect passion (T 2.1.1.4), and Hume describes it as a kind of pride that has authority or power as its cause (T 2.1.8.4, 2.2.8.14). And the ambition of “acquiring a name” in particular seems intimately connected to the “love of fame” (T 2.1.11), which is the pride caused by others’ admiration. Thus, the indirect passions are prominent in the “Conclusion”.

Another feature of the “Conclusion” that has not often been mentioned is Hume’s existential concerns, which, as I will argue later, seem best captured with reference to the indirect passions. Hume, who falls into the radical scepticism, appears to be:

some strange uncouth monster, who not being able to mingle and unite in society, has been expell’d all human commerce, and left utterly abandon’d and disconsolate. Fain wou’d I run into the crowd for shelter and warmth; but cannot prevail with myself to mix with such deformity. I call upon others to join me, in order to make a company apart; but no one will hearken to me. Every one keeps at a distance, and dreads that storm, which beats upon me from every side. (T 1.4.7.2)

When the sceptical mood reaches its climax, Hume’s existential concerns become obvious:

Where am I, or what? From what causes do I derive my existence, and to what condition shall I return? Whose favour shall I court, and whose anger must I dread? What beings surround me? and on whom have I any influence, or who have any influence on me? I am confounded with all these questions, and begin to fancy myself in the most deplorable condition imaginable, inviron’d with the deepest darkness, and utterly depriv’d of the use of every member and faculty. (T 1.4.7.8)

In particular, he seems to be concerned about himself as a philosopher, when asking, “with what confidence can we afterwards usurp that glorious title [philosopher], when we thus knowingly embrace a manifest contradiction?” (T 1.4.7.4) Likewise, in the positive phase, Hume’s self-identification as a philosopher is prominent: “if we are philosophers, it ought only to be upon sceptical principles, and from an inclination, which we feel to the employing ourselves after that manner” (T 1.4.7.11). This passage suggests that in returning to the use of reason, Hume seems to be interested in some psychological inclination that bestows upon him the identity of philosopher.¹³ As I will argue in the following sections, these existential concerns seem to be intimately connected to the indirect passions. Hume himself hints at such a connection: through the indirect passions, Hume seems to try to accommodate some aspects of person, as he refers to “personal identity . . . as it regards our passions” (T 1.4.6.5).

Moreover, the structure of the Treatise gives us an additional reason to look to the indirect passions. Hume pronounces that Book One and Book Two are closely intertwined, by saying, “[t]he subjects of the Understanding and Passions make a compleat chain of reasoning” (T Adv.).¹⁴ If Book One and Book Two are continuous, then we expect a particularly close relationship between the “Conclusion” and Hume’s account of the indirect passions: Book Two begins with Hume’s discussion of the indirect passions, which means that it comes just after the “Conclusion”. Also, of the three parts of Book Two, the first two parts are devoted to discussions of the indirect passions. Thus, Hume’s self-proclaimed continuity of the Treatise and the centrality of the indirect passion in Book Two lead one to expect that there is some connection between the dramatic ending of Book One and Hume’s account of the indirect passions.

5. The Indirect Passions and the Notion of Person

Then, what are the indirect passions, and how are they related to the notion of person? Hume characterizes the indirect passions in terms of what he calls the “double relation of ideas and impressions” (T 2.1.5.5). The indirect passions are in themselves simple feelings, and take a positive or negative valence (T 2.1.2.1, T 2.2.1.1, T 2.2.2.3). But Hume describes the feelings in terms of their causes and objects. The causes consist of an impression of pain or pleasure, and an idea of the object which the hedonic quality accompanies (e.g., a perception of a beautiful house involves an impression of beauty and an idea of a house). The causes are associated with an idea of a person (the beautiful house has to be conceived as my house, to stir pride). When they bring about the indirect passions, our attention is directed toward the objects of the passions, that is, self or other people (the owner of the beautiful house). The idea of the object which the hedonic qualities accompany and the idea of the person are connected via principles of associations of ideas. The impression of pain or pleasure and the impression of the indirect passions are connected via the principle of the association of impressions, that is, resemblance. Thus, the process in which the indirect passions are produced can be described as the “double relation of ideas and impressions” (T 2.1.5.5).

In this account, however, Hume does not clearly explain how the framework of the “double relation” contributes to an account of person. Still, we can extrapolate the relation between the passions and person from what Hume says in the texts. Many commentators have pointed out that the indirect passions lead us to form ideas of our identity as bearers of the causes of the passions (Rorty 1990; Ainslie 1999; Ainslie 2005; Taylor 2015: ch. 1). For instance, when I am proud of the music I have written, the pride produces the idea of myself as a musician. When I love my friend for her beautiful house, the love makes me conceive her as a homeowner. This interpretation fits well with Hume’s account of the indirect passions where the person in the “double relation” is always to be understood in relation to a specific cause. And the causes of the indirect passions seem to have a special relationship with the persons when Hume says that the cause of the passions is “consider’d as connected with our being and existence” (T 2.1.8.8). Hume also states that pride tells us “our own merit” (T 3.3.2.8) and “our rank and station in the world, whether it be fix’d by our birth, fortune, employments, talents or reputation” (T 3.3.2.11), which suggests that through pride we learn about our attributes. Thus, the interpretation that the indirect passions are meant to accommodate the person as the bearer of the causes of the passions is textually supported. To distinguish this kind of personal identity from the traditional concept of personal identity that is concerned with the diachronic sameness of individuals, I will refer to the former as “social identity”, or more simply “identity”.

But what kind of “bearer” relation is at issue here? It might be associations of ideas. I myself am associated with various properties through the associations of ideas, namely through relations of resemblance, spatial or temporal contiguity and causation. For example, there is a causal connection between myself and the music I composed, and in this respect, I may be a bearer of a musical disposition. However, this suggestion makes the associations of impressions or the presence of the indirect passions redundant. If the idea of the bearer of something is available independently of the passions, we have a hard time figuring out Hume’s point of introducing “personal identity . . . as it regards our passions” (T 1.4.6.5). Indeed, Hume seems to be interested in the idea of self that is almost inseparable from the indirect passions. Pride “never fails to produce” (T 2.1.5.6) the idea of self, and “[w]hen the self enters not into the consideration, there is no room either for pride or humility” (T 2.1.2.2). Hume sometimes speaks as if pride is literally part of person:

A certain degree of generous pride or self-value is so requisite, that the absence of it in the mind displeases, after the same manner as the want of a nose, eye, or any of the most material feature of the face or member of the body. (EPM 7.10)

These suggest that the “bearer” relation in the present context and the notion of person based on it must be ones that essentially refer to the indirect passions.¹⁵

According to the standard interpretations, the “bearer” relation is some kind of normative relation expressed by the indirect passions.¹⁶ Hume thinks that the causes of the indirect passions and the person are related in a way that allows for normative evaluation of the person for the causes. An object does not produce the indirect passions when it “attends us during so small a part of our existence” (T 2.1.6.7). If there is no such constant relationship between a person and an object, “’tis impossible he can, upon its account, become the object of punishment or vengeance” (T 2.3.2.6). Actions that do not cause the indirect passions are not subject to moral evaluation: “Actions themselves, not proceeding from any constant principle, have no influence on love or hatred, pride or humility; and consequently are never consider’d in morality” (T 3.3.1.4). When a friend unknowingly steps on my foot, I do not feel anger towards her (or my anger will soon subside), and she is not blameworthy for the action. Hume also writes, “Inanimate objects may bear to each other all the same relations which we observe in moral agents; though the former can never be the object of love or hatred” (EPM App. 1.17). Since Hume takes it for granted that inanimate objects are not subject to moral evaluation (T 3.1.2.4), in this passage Hume seems to distinguish between entities that are subject to normative evaluation and those that are not, with reference to the indirect passions. Wine, no matter how great the taste it gives us, is not loved or respected in its own right. The indirect passion seems to express such normative relations between the cause of the passion and the person, rather than mere factual relations.¹⁷ Borrowing an expression from Taylor (2015: vi), I call such relations “commitments.”¹⁸ For what I am not committed to (e.g., athletic ability or some unintentional actions), I feel no pride or humility, and others would not love or hate me for it. What I am committed to (e.g., musical ability or moral decency) causes those passions. The wine itself does not have any commitment, and thus does not produce the indirect passions.

Thus, when Hume discusses “personal identity . . . as it regards our passions” (T 1.4.6.5) in connection with the indirect passions, he seems to be interested in the notion of a person whose identity is determined by commitment. There is a sense in which the presence of commitment or the indirect passions determines identity. Take the example of an identity as a musician. Intuitively, if one does not take pride in her good music and is not humiliated by her bad music, we would not call her a musician.¹⁹ Instead, it seems that anyone who feels pride or humility about the quality of the music she makes is a musician. In short, it seems that those who are committed to the quality of music are musicians, and those who are not are not musicians. This claim should be distinguished from the claim that a person who is poor at making music is not a musician. An incompetent musician might still be a musician. For she might still want to improve her music, that is, she might feel humiliated by her bad music. The idea is that the commitment to the quality of one’s music, rather than good or bad music itself, makes a person a musician. The same story would go for identity as a philosopher. The fact that I am humiliated by logical fallacies in my reasoning shows that I am a philosopher, a kind of person who engages in a certain kind of reasoning (assuming that avoiding logical fallacies is one of the norms in philosophy).²⁰ To feel proud of good reasoning and to feel humiliated by bad, namely to be committed to the quality of reasoning, is part of my identity as a philosopher. Again, the fallacies do not immediately undermine my identity. Rather, being humiliated by the fallacies shows that I am a philosopher although a bad one. This conception of person makes it intelligible why Hume particularly appeals to the indirect passions to characterize it.

Although in the above account, I only mentioned the self-directed passions of pride and humility, the same account applies to the other-directed passions of love and hate. Those who love or hate me for my music see me as a musician, one who is committed to the quality of music. If people do not have the negative indirect passion toward me for my bad music, they do not see me as a musician. Now, the question is what the relationship is between the self-directed and the other-directed passions. Hume seems to think that the self-directed passions have priority over the other-directed ones in the formation of one’s identity. For he claims that other’s admiration of our traits does not move us if we ourselves do not value them: “The praises of others never give us much pleasure, unless they concur with our own opinion, and extol us for those qualities, in which we chiefly excel” (T 2.1.11.13). For example, a merchant does not feel pleasure even if others admire or love his learning, in which he does not take pride (T 2.1.11.13). This remark suggests that love and hate have force so long as they concur with pride and humility. However, this is not to say that love and hate are peripheral to one’s identity. Rather, when concurring with pride or humility, love or hate has a great impact on us, and sometimes becomes the main source of our pride and humility. Hume writes:

the possessor [of wealth] has also a secondary satisfaction in riches arising from the love and esteem he acquires by them, and this satisfaction is nothing but a second reflection of that original pleasure, which proceeded from himself. This secondary satisfaction or vanity becomes one of the principal recommendations of riches, and is the chief reason, why we either desire them for ourselves, or esteem them in others. (T 2.2.5.21)

Being rich is pleasant on its own, but the admiration from others which the wealth causes becomes the main reason for aspiring to be rich. This suggests that the most important source of pride or humility regarding wealth is love or hate from others. The significance of the indirect passions from others would apply not only to wealth, but to other attributes as well. Hume points out, as an empirical fact, that we are heavily influenced by sympathy with others, and that therefore the praise or blame of others is a typical and indispensable cause of pride or humility (T 2.1.11). This entails that love and hate play a significant role in the formation of one’s identity. Namely, feeling pride and humility make me who I am, but these passions often come from love and hate from others. This point seems plausible: taking pride in one’s music makes one a musician, but since love is an indispensable cause of the pride, the significant part of the identity amounts to being loved by fellow musicians and audiences. Feeling proud that one’s music is loved and feeling humiliated that one’s music is blamed seem to be an essential part of being a musician.

Here I address one possible concern with the above interpretation. I have suggested that the indirect passions are expressions of commitments to the cause of the passions, and the commitments make one who one is. However, one might think that the notion of commitment does not fit well with Hume’s view that the causes of the indirect passions are not necessarily something one has chosen for oneself. According to Hume, something’s being a cause of an indirect passion does not necessarily require that the person at which the passion is directed intends it (T 2.2.3). Virtue and vice, which are instances of causes of the indirect passions, include involuntary traits such as natural ability (T 3.3.4). A high social “rank” generates pride, insofar as it is generally regarded as a cause of pride (or the connection between high rank and pride is established as a “general rule”), even if the person in question does not originally take pride in that rank (T 2.1.6.8). Thus, there may seem to be a gap between the commitment and the indirect passions, since the latter involves matters not of one’s own choice.

In reply, we need not always understand “commitment” in terms of deliberate choice or intention. Schauber (1996) distinguishes between active and passive commitment. Active commitment to something involves one’ s decision to get involved in it. A promise is an example. Passive commitment “do[es] not exist at will” (Schauber 1996: 122), but “one can find oneself caring deeply about the fate of [the object of the commitment]” (Schauber 1996: 121).²¹ An example is friendship. I propose to attribute to Hume’s account of the indirect passions a broad concept of commitment that encompasses these two kinds of commitments. When my natural abilities and social status, which I myself did not decide to have, stir pride or humility, these traits are the object of my deep care and concern.²² Here, I seem to be committed to these traits in the passive sense. To draw on the previous examples, playing music or engaging in reasoning may not have been what musicians or philosophers intended to get involved in, but even so, they are deeply invested in them. The indirect passions seem to accommodate this broad sense of commitment and thereby contribute to the formation of our identity.

In sum, being proud and humiliated about some kinds of things, which often amount to being loved and hated for them, expresses one’s commitment to them, where the commitment is understood in a broad sense. The indirect passions, and the commitments embedded in them, give us a specific identity.²³

6. A Constitutivist Ground for the Use of Reason

With Hume’s views on person and the passions in hand, I attempt to find in the “Conclusion” something similar to “constitutivism” about normativity (e.g., Korsgaard 2009; Velleman 2009; Katsafanas 2013; Schafer 2019). There are different types of constitutivism, but according to Katsafanas (2013: 38–41), constitutivists have the following lines of thought in common. They claim that agents (or actions or practical identities, depending on different authors) have a constitutive aim: seeking the aim is exactly what makes them agents.²⁴ Then, assuming that aims generate reasons for actions,²⁵ we as agents have reasons to do certain things by virtue of what we are. To see how this suggestion works, the example of a chess-player is useful. Chess-players have the constitutive aim of checkmate, and if they are just moving pieces around randomly without the aim, they are no longer chess-players. So, chess-players, by virtue of who they are, have a reason to do the actions that are conducive to checkmate. Of course, chess-players can also have a non-constitutive aim (e.g., to enjoy the game), and the aim can give them reasons for actions. The difference between non-constitutive aims and constitutive ones of chess-players is that the latter is completely non-optional for them: you can play chess without enjoying it, but you can’t play chess without aiming at checkmate. In this respect, the constitutive aim provides an inevitable, normative binding force. In the same way, constitutivists believe that there is a constitutive aim that agents necessarily have, and given that we are inescapably agents, from the aim we get normative claims that apply to all of us. For example, Korsgaard thinks that conforming to the categorical imperatives is the constitutive feature of agency (Korsgaard 2009: 81). If so, being agents would give us a reason to follow certain norms, insofar as they are derived from the categorical imperatives. However, constitutivists disagree over what the constitutive aim of agency exactly is. Although it is controversial as to which type of constitutivism is plausible, or whether constitutivism in general is defensible, the aim of this paper is interpretive and I do not enter into these issues.

From the previous section, we can say that the indirect passions are constitutive of person of a specific kind in that to be proud or loved (e.g., for good music) and to be humiliated or hated (e.g., for bad music) are the very things that make us who we are (e.g., a musician).²⁶ Now, in this framework, to be a certain kind of person involves the aims of getting what produces pride (love) and avoiding what produces humility (hate). Pride in type X of things involves aiming to get the type X of things: if you are not interested in getting them, then you are not proud of having them. Humility for type X of things involves aiming to avoid the type X of things: if you do not care, then you are not humiliated by having them. Thus, from the indirect passions, we get the two aims of getting what produces pride and avoiding what produces humility. What is constitutive of a musician is to aim at making pride-producing music and to aim at avoiding humility-producing music. What makes one a philosopher is to aim at pride-producing reasoning and to aim to avoid humility-producing reasoning. Of course, given that love and hate are often main sources of pride and humility, aiming at love-producing reasoning and aiming to avoid hate-producing reasoning are also important parts of being a philosopher. Abandoning these aims is not optional for philosophers.

However, one might think that while these aims are non-optional for philosophers, being a philosopher is optional. In other words, one could quit being a philosopher, and thereby ignore its constitutive aims. However, Hume seems to think that most of us rarely change what kind of things we feel pride or humility in. Hume writes:

Almost every one has a predominant inclination, to which his other desires and affections submit, and which governs him, though, perhaps, with some intervals, through the whole course of his life. (EMPL The Sceptic, 160)

The “predominant inclination” to which one’s other desires submit can be construed as the indirect passions. For Hume describes the indirect passions as governing desires and other affections: they give “additional force to the direct passions, and encrease our desire and aversion to the object” (T 2.3.9.4). Also, Hume states that his own “ruling passion”, which seems interchangeable with the “predominant inclination”, is “love of literary fame” (EMPL My Own Life: xl). Love of fame is an indirect passion, that is, pride caused by the admiration from others (T 2.1.11). These suggest that, for Hume, pride and humility in certain kinds of objects are constant “through the whole course of his life”.

However, it might seem to be an exaggeration to say that we can never change our indirect passions. But at least it follows from the nature of the indirect passions that we cannot easily escape the commitments embedded in the indirect passions. First, the indirect passions are feeling, that is, something “that depends not on the will” (T App, 2). This means that we cannot voluntarily stop feeling the indirect passions, and therefore cannot voluntarily cease to be philosophers. Second, the indirect passions involve durable commitment: “[a]ctions themselves, not proceeding from any constant principle, have no influence on love or hatred, pride or humility” (T 3.3.1.4, my emphasis). If we are proud of our philosophical thinking, we have to be philosophers, at least for a while, and we cannot casually change our constitutive aims.

This framework of the indirect passions gives a basis for Hume’s normative claims in the “Conclusion”. In the negative phase, Hume’s uncritical, default adherence to reason leads to “desponding reflections” (T 1.4.7.1), including the dangerous dilemma. As seen in Section 4, Hume’s sceptical discussions make other people feel the negative passions such as “hatred” or “anger” (T 1.4.7.2) toward him. I also noted that Hume has existential concerns: he asks himself “Where am I, or what?” (T 1.4.7.8), and “with what confidence can we afterwards usurp that glorious title [philosopher]”? (T 1.4.7.8). The above discussion shows that these existential concerns and the negative indirect passions are closely linked. Hume is a philosopher, that is, someone who aims at pride-producing reasoning and aims to avoid humility-producing reasoning. However, as Hume shows in discussing the dangerous dilemma, strict adherence to reason requires either credulity or total extinction of belief. Both conclusions provoke the negative passions in others, and Hume himself feels shame about them. And he does not know how to escape the dilemma: “For my part, I know not what ought to be done in the present case” (T 1.4.7.7). If you have no idea how to avoid a certain thing, then it seems difficult for you to aim to avoid it. Then, since he cannot even aim to avoid humility-producing reasoning, he fails to meet the constitutive feature of a philosopher and appears to cease to be who he is. This understanding provides a constitutivist rationale for why Hume should not place authority by default on reason. Being a philosopher gives him a reason to avoid cognitive activities that only produce the negative passions such as hate or humility.

In the positive phase, according to the naturalistic interpretations, Hume replaces the default reliance on reason with standards of reason based on some psychological feature. Again, as suggested in Section 4, these renewed naturalistic criteria seem to be closely related to the indirect passions and Hume’s existential concerns. Garrett’s title principle says, “[w]here reason is lively, and mixes itself with some propensity, it ought to be assented to” (T 1.4.7.11). This passage comes with Hume’s self-identification as a philosopher: “if we are philosophers, it ought only to be upon sceptical principles, and from an inclination, which we feel to the employing ourselves after that manner” (T 1.4.7.11). Or as Schafer (2014) emphasizes, Hume refers to curiosity as what drives him to philosophical inquiry (T 1.4.7.12). But Hume pairs curiosity with the indirect passion of ambition and love of fame (T 1.4.7.12). According to my interpretation, the indirect passions and social identity as a philosopher are intertwined: aiming for reasoning that generates pride or satisfies ambition is exactly what makes Hume a philosopher. We can now find a constitutivist basis for Hume’s naturalistic response to scepticism. Lively reasoning or curiosity-satisfying reasoning can be a cause of pride and person-constituting, and therefore being a philosopher provides a non-optional reason to pursue such reasoning.²⁷ If my interpretation is correct, Hume’s response to scepticism is not groundless in such a way that it exclusively relies on free-floating psychological propensities, but has a constitutivist basis.

It might be said that radical scepticism, such as that which leads to the dangerous dilemma, is still something philosophers take pride in, insofar as it is valid, and failure to follow that scepticism is a shame for a philosopher. In fact, in the argument that strict adherence to reason would lead to the annihilation of all beliefs, Hume “can find no error” (T 1.4.1.8). ²⁸ Then isn’t excessive scepticism something philosophers admire and part of their identity? I think, for two reasons, that radical scepticism cannot be the constitutive aim of philosophers. First, radical scepticism is painful and cannot be a cause for pride, no matter how faithful philosophers may be to the scepticism. Excessive scepticism arising from the default adherence to reason makes philosophers’ minds “uneasy” (T 1.4.1.10) and leads to “despair,” “melancholy” (T 1.4.7.1) and “hate” (T 1.4.7.2). All of these are painful and cannot be the cause of pride, because as a fact of human psychology, the cause has to be somehow pleasurable (T 2.1.5.5). The problem is that even though philosophers successfully follow the default norm of reason (“no error” in the radical scepticism), they feel pain and thus humility. It is not that they feel humiliated because they fail to follow the norm. Thus, the only way for philosophers to maintain their identity or avoid humility-producing reasoning is to modify the norms of reason to something they can be proud of. That would be some alternative naturalistic standard. Second, according to Hume, it is impossible for anyone to remain fully committed to the radical scepticism that demands the annihilation of all beliefs (T 1.4.1.7). In order to act, philosophers must believe in causality, external existence, etc. If so, they have to partially ignore the norms of reason while declaring their commitment to the annihilation of all beliefs. Where total reliance on reason is part of philosophers’ identity, this impossibility of following the norm threatens their identity, as Hume laments, “with what confidence can we afterwards usurp that glorious title [philosopher], when we thus knowingly embrace a manifest contradiction?” (T 1.4.7.4).²⁹ Since philosophers cannot fully submit to radical scepticism, they must search for a feasible epistemic policy that they can consistently pursue and allows them to maintain their identity. This would be some naturalistic norm that would generate pride.

One might think that this proposal ends up with something similar to Kemp Smith’s interpretation that appeals to the irresistibility of beliefs and inferences. My interpretation also relies on the fact that someone like Hume has to accept the identity of a philosopher. If so, then the criticism directed at Kemp Smith’s natural beliefs, that they are explanations of what we do believe, not of what we ought to believe (Ainslie 2015: 232), may be directed at my interpretation as well. Following Katsafanas’s response to a similar concern about constitutivism in general (Katsafanas 2018: 384–85), we can respond as follows. Kemp Smith seems to derive the normativity of belief from its irresistibility: “The beliefs which ought to be accepted are, [Hume] teaches, beliefs that Nature itself marks out for us. In their fundamental forms, as ‘natural’ beliefs, we have no choice but to accept them; they impose themselves on the mind” (Kemp Smith 2005: 388). However, what is irresistible in my interpretation is to aim at pride-producing reasoning and to aim to avoid humility-producing reasoning. My interpretation does not derive normativity from the irresistibility of beliefs or inferences, but from the irresistibility of aiming at a certain kind of reasoning. This picture is an account of what we “ought” to do, without ruling out the possibility that we fail to make good reasoning.

The proposed framework implies that the normativity of reason is relative to one’s identity. Hume seems happy to accept this consequence. In a later part of the “Conclusion”, He refers to “many honest gentlemen, who being always employ’d in their domestic affairs, or amusing themselves in common recreations” and says, “of such as these I pretend not to make philosophers” (T 1.4.7.14). This remark can be understood as saying that those who are not philosophers, that is, those who do not feel pride or humility in their philosophical reasoning, are not required to commit to such reasoning. How, then, can Hume accommodate the normativity of non-philosophers’ reasoning? In the “Conclusion,” Hume seems to concentrate on the question of what kind of reasoning he should pursue, and his answer to the question concerning non-philosophers is not clear.³⁰ The framework of the indirect passions might be extended to the epistemic normativity for non-philosophers. For example, assenting “to every trivial suggestion of the fancy” (T 1.4.7.6) seems to be shameful even for non-philosophers. Still, since the main topic of the “Conclusion” seems to be the normativity of reason for philosophers, this paper, which purports to interpret the “Conclusion”, cannot do justice to this possiblility here.

7. Concluding Remarks on the Scope of My Interpretation

I would like to conclude by stating what this framework is not supposed to explain. The purpose of this paper is to supplement the naturalistic interpretations, not to offer an alternative to it. Indeed, I remained neutral about exactly what kind of reasoning Hume thinks philosophers or more generally people are proud of. The naturalistic interpretations have offered such standards in various forms, such as the Title Principle, natural beliefs or some curiosity-based epistemic norms. The present paper, however, is not meant to determine which of these options best captures Hume’s thoughts and constitutes Hume’s identity as a philosopher. What can be said here is that all of the above options are compatible with the constitutivist framework of the indirect passions: all of the lively reason, natural or irresistible reason, and curiosity-satisfying reason can be admired by others, and therefore can be a cause of pride. Thus, the indirect passions can give normative force to these epistemic norms.³¹ In this sense, my interpretation supplements the naturalistic interpretations.³²

Nevertheless, it should be emphasized that the constutivist framework is not a device that automatically confers binding force on any naturalistic epistemic standards. Rather, it imposes significant constraints on such standards. Hume says that the causes of pride and humility satisfy the following conditions (T 2.1.6), many of which are already implied in my account so far. (1) The causes are intimately connected to a person: the correct/incorrect reasoning that we happen to make does not produce pride/humility. (2) They are peculiar to the bearers of the causes: reasoning that is too easy does not cause pride, and failure in reasoning that is too difficult is not humiliating. (3) They are publicly approved or disapproved: reasoning that is not accompanied by love/hate does not produce pride/humility. (4) They are durable to some degree: making correct/incorrect reasoning only once is not enough to stir pride/humility. (5) They conform to general rules: if the standards of correct/incorrect reasoning are not shared in a community, pride/humility is not aroused. Reasoning that does not meet these conditions does not produce the indirect passions, and hence is not person-constituting. The indirect passions would exclude normatively insignificant reasoning (e.g., reasoning accompanied by momentary desire) from reasoning that appears to satisfy naturalistic epistemic criteria.

Importantly, it is not possible to present a complete interpretation of Hume’s response to scepticism and the epistemology that results from it without specifying what kind of reasoning deserves the philosopher’s pride. In particular, the indirect passions alone cannot fully exclude epistemically flawed reasoning, especially superstition.³³ In a community united by superstitious doctrine, superstitious, false reasoning would produce pride and admiration. What, then, is the difference between reasoning worthy of the pride of superstitious people and reasoning worthy of the pride of philosophers? For the constitutivist framework to be fully persuasive, we have to wait for the answer to this question.

In this paper, however, I am content if I can show that Hume’s naturalistic response to scepticism comes with normative force, involving more than mere psychological description. To reiterate, Hume’s identity as a philosopher gives him non-optional reasons to seek pride/love-producing reasoning and to seek to avoid humility/hate-producing reasoning. This interpretation fits well with the text in that it accommodates the prominence of the indirect passions and Hume’s existential concerns in the “Conclusion”, and the close connection between Book One and Book Two of the Treatise that Hume himself declares.