The principal advantage, which Juvenal discovers in the extensive capacity of the human species, is that it renders our benevolence also more extensive, and gives us larger opportunities of spreading our kindly influence than what are indulged to the inferior creation. (EPM.177)

Introduction

Hume’s Treatise, Enquiries and Essays contain plentiful material for an investigation into the moral nature of other animals and our moral relations to them. In particular, Hume pays considerable attention to animal minds. He also argues that moral judgment is grounded in sympathy. As sympathy is shared by humans and some other animals, this already hints at the possibility that some animals are morally considerable, even if they are not moral agents.Footnote 1 Many contributions to the literature on animal ethics assume one of the big three normative theories as their starting point; consequentialism, deontology or neo-Aristotelian virtue ethics.Footnote 2 However, as several philosophers have argued, Hume’s discussion of animals suggests a distinctive, alternative approach. Work in this area has varied greatly in terms of its level of engagement with Hume’s writings. Some authors have been primarily concerned with applying broadly Humean approaches to questions in applied ethics and have paid little or no attention to exegetical matters.Footnote 3 Others have focused on more or less tightly defined issues of Hume scholarship such as the necessary conditions for moral agency (See, for example, Tranöy (1959), Nuyen (1998), Boyle (2003) and Pitson (2003)). My aim here is to assemble the textual basis for a Humean approach to animal ethics, systematically and from the ground up, via a careful study of Hume’s texts.Footnote 4 The result is an approach to animal ethics grounded in a general philosophy of nature, a naturalistic methodology and broadly plausible psychological assumptions that has obvious attractions. Although I shall do little to defend or apply the approach, this is an ambitious task in some respects. It requires a broad scope as I consider what Hume has to say concerning, inter alia, questions of moral psychology, ethics and epistemology. Drawing together the necessary elements inevitably involves covering some territory that is familiar to students of Hume from other contexts. However, while much of the exposition should be fairly uncontroversial, I shall also need to consider more contentious interpretative questions as I address some puzzles and difficulties along the way.

The plan ahead is as follows. After sketching Hume’s general approach to studying human and animal minds as parts of nature and his account of moral judgment grounded in sympathy (Sects. 2 and 3), I go on to explain how sympathy operates across species (Sect. 4) and the methodological basis for ascribing sympathy to animals (Sect. 5). I then take up questions concerning the appropriateness and extent of our moral judgments (Sect. 6 and 7). Moral judgments have proper objects and involve an imaginative shift of perspective as we approximate the common point of view. Given the possibility that sympathy, methodized and corrected, appropriately extends to animals there is no reason for Hume to exclude them as proper objects of moral concern. Nevertheless, Hume denies that we have obligations of justice towards animals and also denies that animals are moral agents. Following Driver (2011), I argue that Hume’s claims concerning justice do not undermine the central thesis that animals are proper objects of moral concern (Sect. 8). I also propose a novel explanation for Hume’s denial of moral agency to animals based on his essay Of Polygamy and Divorce (Sect. 9).

Natural Continuity

Like other Scottish Enlightenment thinkers, Hume devotes considerable attention to animal minds. However, he is not primarily concerned with our moral relations to animals or with comparing human virtues to the imagined virtues of animals in a state of nature.Footnote 5 Rather, he wants to illustrate the continuity between the rational and emotional capacities of humans and animals. He is out to undermine the Cartesian doctrine of animals as natural automata and to challenge the sharp bifurcation between the mental and the non-mental. This is all part of his larger project of introducing the experimental method of reasoning into moral subjects, or, as we might say, naturalizing mind and morals.

On the whole, Hume’s strategy is one of leveling down. He wants to puncture the pretensions of human reason and especially the pretensions of moral rationalists like John Locke, William Wollaston and Samuel Clarke. Reason is not an infallible, God-given faculty. It cannot justify itself and even tends to undermine and subvert its own conclusions (T.183; T.267–268). Reason is a natural faculty governed by psychological principles of association of ideas, driven by custom and habit. The faculty, and the principles that govern it, are amenable to empirical investigation, primarily careful observation of everyday life and introspection. The same is true of the passions. Moreover, there are basic similarities between the faculties of reason or passion as they are found in human and animals. While differences remain, human reasoning is less special and distinctive than commonly supposed if animals can reason too. The same is true for moral judgment if it depends on passions that animals possess as well. We should study the human mind in much the same way that we study the natural world, and understand ourselves as belonging to that world.Footnote 6

So, Hume emphasizes the mental complexity of animals in order to accentuate the similarities between their minds and ours. Indeed, it is possible to mine from his writings an extensive list of traits that he thought animals and humans have in common. A list drawn just from the Treatise would include ‘thought and reason’ (T.176), the ability to adapt ‘means to ends’ for self-preservation, attaining pleasure and avoiding pain (T.176), the possession of structurally complex emotions such as ‘pride and humility’, ‘contempt’ and ‘vanity’ (T.326) that have both a cause and an object and require an idea of self, principles of associative reasoning, namely, ‘resemblance, contiguity and causation’ (T.327), passions including ‘love and hatred’ (T.397), ‘affection’ (T.398), ‘courage’, ‘anger’. ‘grief’, ‘envy and malice’ (T.398), and ‘will’ (as opposed to mere inclination) (T.448). Hume, then, attributes relatively sophisticated mental lives to at least some other animals. Moreover, he attributes these capacities widely ‘thro’ the whole animal creation’ (T.328).

Virtue, Vice and Moral Judgment

In order to draw out the significance of Hume’s assertions concerning the rich mental lives of animals, we need in front of us a sketch of his account of moral judgment. For Hume, it is the virtues, not particular actions, which are the principal targets of moral appraisal (T.411; T.477–478; T.575). It is true that we judge a person to be virtuous or vicious, in this respect or that, because we approve or disapprove of their actions. However, this is because a person’s actions are the only evidence we possess of their inner character or motive. Pace moral rationalists, judgments about virtue depend upon the passions rather than reason (at least in the first instance, as I discuss below). It is because our ideas of a person’s character produce certain ‘agreeable’ or ‘uneasy’ (T.470–471) sentiments of approbation or disapprobation that we judge the person to be virtuous or vicious.

Why, though, should a person’s character (or the actions that spring from their character) produce either pleasurable or painful sentiments in an observer? Hume’s answer appeals to the role of sympathy in our moral psychology. Sympathy, he writes, ‘is the chief source of moral distinctions’ (T.618).Footnote 7 It is due to our capacity for sympathetic responsiveness to the inner lives of others that we come to approve or disapprove of character traits. We approve of character traits, and label them as virtues, insofar as they are useful or agreeable to their possessor or to others. To say that a character trait is useful or agreeable to its possessor or to others is to say that the trait tends to produce pleasurable impressions.

So, we judge that a person is virtuous or vicious depending on whether their character tends to produce pleasure or pain in the minds of those who are affected by their actions and towards whom we feel sympathy. As we feel a natural sympathy towards all other humans (T.605), we judge people to be virtuous or vicious insofar as they cause humans pleasure or pain. However, there is nothing in the structure of Hume’s theory that depends on species membership. Instead, it suggests that insofar as we feel sympathy towards other animals, and insofar as a person’s character tends to produce pleasurable or painful impressions in the minds of animals, then we will tend to form judgments about the virtue or vice of that person’s character. So, for instance, we will tend to judge that someone who causes animals to suffer is thereby vicious.

How Sympathy Works

Hume’s project of naturalizing mind and morals combined with his emphasis on sympathy suggests that other animals may appropriately figure into judgments concerning virtue and vice. How, though, more precisely, does sympathy play its role in our moral psychology? Hume’s account is pleasingly detailed (T.316–324; T.575–576). We might distinguish the following six stages:

  1. 1.

    We observe either the outward behavior of others or other outward signs likely to cause passions or sentiments, such as a surgeon’s instruments laid out for an operation.

  2. 2.

    We infer the other’s sentiments from that behavior or evidence, which produces an idea of that sentiment in us.

  3. 3.

    Due (principally) to our resemblance to the other, and also due to the relations of contiguity and causation, we associate in our imagination the idea of the other’s sentiment with an impression of ourselves.

  4. 4.

    This association increases the vivacity of the idea of the other’s sentiment. The greater the resemblance and contiguity, the greater the idea’s vivacity.

  5. 5.

    Given sufficient vivacity, the idea is enlivened and becomes a secondary impression of the same sentiment in us.

  6. 6.

    This sentiment motivates us to act insofar as it is pleasurable or painful; qualities which we are naturally disposed to try to promote or diminish.

Hume describes the manner in which sympathy operates between humans. ‘The minds of men,’ he tells us, ‘are mirrors to one another’ (T.365). However, there is nothing about the mirroring, on Hume’s account, that obviously requires a high level of cognitive ability or abstraction that is unique to us. Therefore, there is no obvious bar on animals either being capable of sympathizing or being the objects of sympathy. While examining each stage in sufficient detail to establish this would take too long, it is worth looking at one stage in some more detail. This will give us a better sense of how Hume’s associationist psychology explains apparently sophisticated patterns of thought without relying on forms of theoretical reason that animals seem not to possess.

The transition of thought described in (2) requires that we infer the idea of another’s feelings from their behavior. This might be thought to require relatively sophisticated cognitive capacities, such as, in our terms, a theory of mind. However, Hume’s approach depends solely on principles of association that lead minds from one idea to another without any requirement of rational governance or oversight. It is a brute fact that our minds are constructed in such a way that ideas of the observable behavior or circumstances of others become associated with ideas of their inner mental states. For example, our own experience reveals that the impression of a flame is constantly conjoined to the impression of heat. As a result, the idea of a flame leads to the idea of heat and we form the idea that the former causes the latter (T.87). So, whenever we perceive a flame we form the idea of heat. Now, given our basic resemblance to others, principles of association move our mind from the idea of another person being exposed to a flame to the idea of another person feeling heat.

Although this is an inference from someone’s behaviour or circumstances to their mental states, it depends on probabilistic reasoning that can be explained entirely in terms of associationist principles. It is to be counted as reasoning in the descriptive sense that the mind makes regular transitions between ideas in accordance with experience. However, there is no need to invoke higher-order, rational governance or oversight. Reason is just the description we give to the orderly transition of ideas in accordance with the psychological principles that arise from our natural constitution. The following passage captures the idea well:

Nothing shews more the force of habit in reconciling us to any phaenomenon, than this, that men are not astonish’d at the operations of their own reason, at the same time, that they admire the instinct of animals, and find a difficulty in explaining it, merely because it cannot be reduc’d to the very same principles. To consider the matter aright, reason is nothing but a wonderful and unintelligible instinct in our souls, which carries us along a certain train of ideas. (T.178–179)

There is no reason to limit reasoning in this descriptive sense to humans. Nor do other stages in the process require uniquely human abilities. Indeed, Hume explicitly observes that some other animals are capable of sympathy. We can, thinks Hume, ‘observe the force of sympathy thro’ the whole animal creation, and the easy communication of sentiments from one thinking being to another’ (T.363). Moreover, it is ‘evident that sympathy, or the communication of passions, takes place among animals, no less than among men’ (T.398).

Sympathy Between Species

Hume’s account of sympathy depends on the resemblance between the sympathizing mind and the object of its sympathy. The two must be relevantly similar for the idea of someone else’s passions to be associated in the imagination of the sympathizing mind with an impression of self and become enlivened into a secondary impression. Humans feel sympathy towards one another because ‘nature has preserv’d a great resemblance among all human creatures’ (T.318), both mental and physical. There is a similar relation of resemblance between animals of the same species who also feel sympathy towards each other. As Hume observes:

Grief likewise is receiv’d by sympathy; and produces almost all the same consequences, and excites the same emotion as in our species. The howlings and lamentations of a dog produce a sensible concern in his fellows. (T.398)

Given the necessity of an impression of self for the operations of sympathy, this is a striking claim.

We might ask, however, whether the communication of feelings can cross the barriers between species. Given that both humans and animals are capable of sympathy, and that the capacity doesn’t depend on capacities peculiar to us, it should be no surprise that Hume’s answer is “yes”. He affirms this in several places, for example:

Love in animals, has not only for its object animals of the same species, but extends itself farther, and comprehends almost every sensible and thinking being. A dog naturally loves a man above his own species, and very commonly meets with a return of affection. (T.397; see also T.481 and EPM.300)

The fact that Hume makes the point repeatedly indicates its importance to his overall project. While humans and animals are mentally and physically different in many ways, the common capacity for sympathy suggests, by itself, a relevant like-mindedness.

A wonderful anecdote also suggests that the young Hume, at least, felt sympathy for other animals. Mossner relates it as follows:

In his youthful period Hume’s moral fibre was tested at a dinner party given by Lady Dalrymple in Edinburgh. Lady Anne Lindsay relates the story as told by her grandmother.

You know the Truthfulness of his Honest Nature … as a Boy he was a fat, stupid, lumbering Clown, but full of sensibility and Justice, - one day at my house, when he was about 16 a most unpleasant odour offended the Company before dinner … “O the Dog … the Dog.” Cried out everyone “put out the Dog; ‘tis that vile Beast Pod, kick him down stairs … pray.” –

Hume stood abashed, his heart smote him … “Oh do not hurt the Beast” he said … “it is not Pod, it is Me!”

I think this is capable of being made a very good proverb of, “It is not Pod, it is me!”

How very few people would take the evil odour of a stinking Conduct from a guiltless Pod to wear it on his own rightful Shoulders. (Mossner 1980, 65)

It may seem to be a commonplace that most humans feel sympathy for other animals—even if not all of us would follow Hume in shouldering the stinking conduct falsely attributed to Pod. Hume, however, has a principled, psychological account to explain this phenomenon (albeit one that is hostage to empirical fortune in its details and that we should not now accept in its entirety).

Hume, moreover, has methodological grounds for expecting a like-mindedness between humans and animals. He was interested in the minds of other animals primarily because he thought it possible to make inferences back to human psychology (many evolutionary psychologists make the same, reasonable assumption). In order to do so, he relies on a general, analogical principle licensing the inference from like effects to like causes (See further Boyle (2003, 6–10)). He expounds the principle as it specifically relates to humans and animals in a couple of places in the Treatise. Here is one example:

‘Tis usual with anatomists to join their observations and experiments on human bodies to those on beasts, and from the agreement of these experiments to derive an additional argument for any particular hypothesis. ‘Tis indeed certain, that where the structure of parts in brutes is the same as in men, and the operation of these parts also the same, the causes of that operation cannot be different, and that whatever we discover to be true of the one species, may be concluded without hesitation to be certain of the other. (T.325; see also T.176–177, T.397, EHU.104–5)

For instance, animals’ causal and inductive inferences are apparently non-rational in the sense that they involve neither demonstrative reasoning nor second-order rational governance. Given the same event (patterns of causal and inductive inference) we can reason analogically to the same cause in humans (non-rational processes of psychological association). This is an important strand in Hume’s argument against the presumed rationality of induction.

Hume, then, is committed to applying principles of analogical reasoning from like effects to like causes across species. His concern with animals is part and parcel of his project to place the moral sciences on an experimental footing; we should study humans as we study other animals, by careful observation, not revelation or a priori metaphysical speculation. Hume’s principle concerning analogical reasoning licenses us to give the same sort of accounts of human and animal psychology. Evidence of some behavior in animals gives us reason to attribute the same mental states to them as we would do to humans, and vice versa. Just as everything material is governed by the same Newtonian laws, so too all minds are governed by the same, empirically investigable, associationist principles. Unless particular observations suggest otherwise, we should assume a basic similarity between the mental states of humans and animals.Footnote 8

Reason and Moral Judgment

As we can feel sympathy towards animals, we can be affected by their pleasures and pains. It follows that we will tend to judge that a person can be virtuous or vicious not only with respect to their effects on other humans, but also to the extent that they cause pleasure or pain to other, non-human animals. This seems to be in line with common sense, if not common practice.

Following from Hume’s descriptive account, a further, more obviously philosophical, normative question arises concerning the correctness of these judgments. It might occur to some readers that the question of the correctness or appropriateness of our moral judgments does not arise within the context of Hume’s moral psychology. Moral judgments are a species of passions, and, at times, Hume writes as though the passions are brute, arational states without correctness conditions:

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; see also T.458)

It is hard to know precisely what to make of these claims. On the one hand, they are repeated, clear and unequivocal. On the other hand, it is not obvious that they are consistent with other principles that are integral to Hume’s final, considered account of moral judgment. Thus, Baier somewhat notoriously describes the ‘original existence’ passage as “one very silly paragraph that has perversely dominated the interpretation of … [Hume’s] moral psychology” (1991, 160). Part of the explanation, she suggests, is that they are hyperbolic comments addressed to a particular audience, namely the moral rationalists. When engaged in strategic battle, it is not unheard of for a philosopher to state the bold opposite of their opponents’ position and then qualify their way back towards the sensible center. Moreover, it must be remembered that the Treatise records an intellectual voyage still at sea in some ‘leaky weather-beaten vessel’ (T.263). Hume addresses different problems as they arise in ways that are not always mutually consistent because he employs ideas that are pertinent and available at a particular stage in the voyage. The sense of intellectual movement helps to explain both the fascination of the Treatise and the mature Hume’s dissatisfaction with his early masterpiece. It might also be that his assertions concerning the arationality of the passions are examples of the sorts of abstruse reasoning that leads to conclusions that ‘vanish, like the phantoms of the night on the appearance of the morning’ (T.455). They cannot be made to cohere with a fuller survey of our moral lives.

In any case, it is clear that reason plays a number of roles in Hume’s final account of moral judgment. Reason not only ‘excites a passion’ (T.459) by revealing facts of the matter and the likelihood that a plan of action will succeed, it also plays important roles in the making of moral judgments, the shaping of ends and the correction of sentiment. Hume emphasizes the role of reason most clearly in the second Enquiry:

But in order to pave the way for such a [moral] sentiment, and give a proper discernment of its object, it is often necessary, we find, that much reasoning should precede, that nice distinctions be made, just conclusions drawn, distant comparisons formed, complicated relations examined, and general facts fixed and ascertained. (EPM.173)

Because reason paves the way, there are indirect standards of reasonableness for the passions. A passion can be indirectly unreasonable insofar as it is produced by a false belief or faulty chain or reasoning.Footnote 9 Moreover, he also seems to suggest in various places that a passion can be unreasonable in itself.Footnote 10 For example, he observes that someone ‘that has a real design of harming us, proceeding not from hatred and ill-will, but from justice and equity, draws not upon him our anger, if we be in any degree reasonable’ (T.350, italics added). Elsewhere, he suggests that ‘nature provides a remedy in the judgment and understanding, for what is irregular and incommodious in the affections’ (T.489).

By allowing that reason and imagination play a role in moral judgment, Hume avoids a potential difficulty. If moral judgments were identical to any pleasurable response then we would be forced to say that the pleasure involved in drinking wine constitutes a form of moral praise with respect to the wine (T.472). Instead, Hume thinks of moral judgments as involving a particular type of pleasurable or painful sentiment (T.517). Pleasures and pains can differ in several ways, including with respect to their phenomenal feel, their causes and their objects. Furthermore, their objects can be imagined in different ways. Not all pleasures and pains are passions with moral objects. Rather, passions of moral praise or blame have their own proper objects that distinguish them from, say aesthetic or prudential passions. Examples of the proper objects of the moral passions include ‘such actions as tend to the peace of society’ (T.533) or ‘to the good of mankind’ (T.577) or, at least, character traits that lead a person ‘to be serviceable and useful within his sphere’ (T.602).

With respect to the passions of love and hatred, Hume also tells us that their object is ‘some other person, of whose thoughts, actions and sensations we are not conscious’ (T.329). Similarly, in the first Enquiry, Hume writes that ‘the only proper object of hatred or vengeance, is a person or creature, endowed with thought and consciousness’ (EHU.98; emphasis added). These comments are significant given that Hume thinks that a sentiment of praise or blame ‘is nothing but a fainter and more imperceptible love or hatred’ (T.614). It follows that such sentiments have proper objects and, therefore, improper objects too.

Hume makes much of the object/cause distinction in his discussion of passions such as pride, humility, love and hatred. Baier has argued that this would make little sense if passions contained no representative quality or reference beyond themselves (1991, 161–163). While I am sympathetic to her argument, there are alternative readings that do not require us to admit that the ‘original existence’ passage is incompatible with Hume’s considered view. For example, Cohon (1994) explains the object-directedness of the passions in terms of their causal relations rather than their representational content. Similarly, Radcliffe argues that, for Hume, passions can have objects without thereby representing them or being truth-apt (2018, 98–101). In this way, she suggests, passions are like feelings of pleasure or pain.

Whether or not Hume’s account of the passions requires them to have representational content, it is clear that they can have proper (and therefore improper) objects. In particular, the proper objects of moral judgments are character traits that tend to cause pleasures and pains in those creatures with whom we sympathize. Or, more precisely, they are character traits that cause pleasure or pain as a general rule regardless of their actual effects in particular cases. This object-directedness gives rise to correctness conditions insofar as the passions can be directed towards their proper objects or, say, too narrowly when we fail to adopt the common point of view—more shortly. Perhaps it is possible to account for this entirely in non-representational, causal terms as Cohon and Radcliffe suggest. The question of the normativity of causal relations is a notoriously difficult and deep one. On the face of it, however, the phenomenon is easier to explain if we allow that the passions have representational content contrary to the ‘original existence’ passage.

Sympathy, Methodized and Corrected

Given that the passions involved in moral judgments have proper objects, it makes sense to consider their extent and the appropriateness of that extent. Is there perhaps some reason why animals are not among the proper objects of moral concern? To be morally motivated by another’s situation, we must be capable of responding to them sympathetically. It is through sympathy that we acquire the pains and pleasures that move us to act. All normal adults, according to Hume, have this capacity [which is perhaps why he thinks that self-professed psychological egoists are disingenuous (EPM.169)]. However, the degree of pain or pleasure that we feel, and the consequent influence on the will, depends upon various contingent factors without obvious moral significance. The closer someone is to us, the more closely they resemble us, or the more likely they are to benefit us, the more intensely we feel pain or pleasure in response to their situation (T.581). As a result, sympathy moves us more strongly with respect to fellow countrymen (T.316–317), ‘relations of blood’ (T.318), and those who are similar to us in terms of ‘manners, or character, or country, or language’ (T.318), or of ‘the same trade, profession, and even name with ourselves’ (T.352). Hume’s explanation is that the mind moves more easily from the idea of the pain or pleasure of someone who closely resembles us to the idea of our own pain or pleasure.

The biases inherent to sympathy inevitably lead to conflict in public life as we favor those who are closest or most similar to ourselves. The results can range from nepotism to war. In an attempt to partially counter these biases, Hume proposes that moral judgment involves adopting the ‘general’ or ‘common’ point of view:

‘Every particular man has a particular position with regards others; and ‘tis impossible we could ever converse together on any reasonable terms, were each of us to consider characters and persons, only as they appear from his particular point of view. In order, therefore, to prevent these continual contradictions, and arrive at a more stable judgment of things, we fix on some steady and general points of view; and always, in our thoughts, place ourselves in them, whatever may be our present position.’ (T.581–582; see also T.591; EPM.229)

Moral agency requires an imaginative shift in perspective away from our immediate point of view. In fact, there is a double movement in imaginative thought. First, we should try to adopt the standpoint of a ‘judicious spectator’ (T.581). Second, we should consider the situation of another from that impartial viewpoint. In doing so, we aim to eliminate irrelevant factors from our moral reasoning and learn to expand our original sympathies. Thus, we ‘correct the different sentiments of virtue, which proceed from its different distances from ourselves’ (T.585). How far beyond our ‘narrow circle’ (T.602) of self, family and friends should our viewpoint expand? In the second Enquiry, Hume suggests that the general point of view extends to all humankind (EPM.272). In the Treatise, he has something more restricted in mind and limits the general point of view to all those directly affected by a particular individual’s character (T.582, T.591, T.602, T.606).Footnote 11

Hume’s discussion is couched as descriptive psychology, rather than in normative terms; ‘we fix on some steady and general points of view’ not ‘we ought to fix’. However, as Hume discusses in various places, we frequently in fact fix on a more local, and even entirely selfish, point of view (e.g., T.583). Moral judgment involves overcoming this tendency.Footnote 12 Thus, Hume writes that ‘experience soon teaches us this method of correcting our sentiments, or at least, of correcting our language, where the sentiments are more stubborn and inalterable’ (T.582). This seems to be both an arch allusion to public hypocrisy and to a necessary condition for public discourse (the allusion being all the archer for the necessary condition, public discourse being necessarily dependent on hypocrisy). Hume has humans in mind, but nothing in his account implies that this is the correct limit even if it is Hume’s own limit. Indeed, there is no obvious, principled reason why it should stop short of all those with whom we share the bonds of sympathy.

Moral Implications

The internal logic of Hume’s account suggests that all minds with whom we can sympathize are equally appropriate objects of moral concern when considered from the impartial, common point of view. We might expect, then, that animals are no less morally considerable then, say, distant strangers. Of course, animals are, overall, more dissimilar to us, and to each other, than we are to other humans. As Hume notes, ‘every human creature resembles ourselves, and by that means has an advantage above any other object, in operating on the imagination’ (T.359). We should also expect, therefore, that our actual sentiments towards animals are weaker than they are towards other humans. Moreover, we should expect that our sentiments towards some animals are stronger than towards other animals. Due to their greater resemblance to ourselves, we might care more about mammals than, say, cephalopods. In fact, it is a clear strength of Hume’s account that it predicts and explains our moral bias against other species of animal. Questions remain as to whether, or to what extent, that bias is justified, and as to the moral implications of our answers.

Hume’s own comments about the treatment of animals are scattered and rather inconclusive. When he discusses hare-hunting and justice, it is justice with respect to the hunter and not the hare that concerns him (T.506). This is in keeping with the idea that justice is an artificial virtue that arises in the context of a social contract to which the hare in not a party. Hume’s most substantial comments on human-animal relations appear in the context of his discussion of justice in the second Enquiry:

Were there a species of creatures, intermingled with men, which, though rational, were possessed of such inferior strength, both of body and mind, that they were incapable of all resistance, and could never, upon the highest provocation, make us feel the effects of their resentment; the necessary consequence, I think, is, that we should be bound, by the laws of humanity, to give gentle usage to these creatures, but should not, properly speaking, lie under any restraint of justice with regard to them, nor could they possess any right or property, exclusive of such arbitrary lords. Our intercourse with them could not be called society, which supposes a degree of equality; but absolute command on the one side, and servile obedience on the other. Whatever we covet, they must instantly resign: Our permission is the only tenure, by which they hold their possessions: Our compassion and kindness the only check, by which they curb our lawless will: And as no inconvenience ever results from the exercise of a power, so firmly established in nature, the restraints of justice and property, being totally useless, would never have place in so unequal a confederacy. This is plainly the situation of men, with regard to animals; and how far these may be said to possess reason, I leave it to others to determine. (EPM.190–191)

Hume’s position seems to follow from his account of the origins and foundations of the artificial virtue of justice. However, there is textual reason to resist this conclusion that a capacity for resistance is a necessary condition for being covered by the restraint of justice. For, whether or not other animals might inflict damage upon us is an appeal to our particular self-interest. Elsewhere, Hume argues as follows:

‘Tis only when a character is considered in general, without reference to our particular interest, that it causes such a feeling or sentiment, as denominates morally good or evil. (T.472; see also T.499, T.591)

While Hume astutely observes that we are liable to conflate moral concern and self-interest, he also notes that ‘a man of temper and judgment may preserve himself from these illusions’ (T.472). In large measure, this is the point of adopting the common point of view from which we make moral judgments.

Whether or not we have duties towards animals as a matter of justice, it does not follow, of course, that animals are in no way morally considerable.Footnote 13 Indeed, Hume says that ‘we should be bound, by the laws of humanity, to give gentle usage to these creatures’. But this is also odd given that compassion arises from sympathy, which, as I have explained, has nothing to do with humanity per se. The explanation may be the role of the sentiment of humanity in the second Enquiry, which, at one stage, he fully identifies with moral sentiment (EPM.235–236; see further (Taylor 2009)). The sentiment of humanity explains the concern that we feel for other humans and is rooted in species membership. Arnold (1995) argues that it is this sentiment that explains Hume’s asymmetric treatment of human and animal morality in both the Treatise and the Enquiries. However, as Pitson (2003: 648) observes, Hume is clear that other animals do also show sympathetic concern—a “disinterested benevolence” (EPM.300)—towards both conspecifics and other animals including humans. Moreover, Arnold’s view that a universal sentiment of humanity grounds Hume’s moral theory in the Treatise as well as the second Enquiry is hard to square with the following:

In general, it may be affirm’d, that there is no such passion in human minds, as the love of mankind, merely as such, independent of personal qualities, of services, or of relation to ourself.’Tis true, there is no human, and indeed no sensible, creature, whose happiness or misery does not, in some measure, affect us, when brought near to us, and represented in lively colours: But this proceeds merely from sympathy, and is no proof of such an universal affection to mankind, since this concern extends itself beyond our own species. (T.481)

Moreover, if Hume did come to think that moral sentiment is identical to the sentiment of humanity, and that this sentiment is necessarily grounded in species-membership, such a view would be unsupported by the analogical methodology and detailed moral psychology outlined in the Treatise as discussed above. In fact, it would be more charitable to read the sentiment of humanity as a rather loose way of talking about a general sentiment of benevolence that is capable of extension to other minds, human or otherwise. When Hume writes ‘that we should be bound, by the laws of humanity, to give gentle usage to these creatures,’ it is plausible that he is using ‘humanity’ just as we might use the word ‘humane’ to describe the outlawing of egregious practices of animal husbandry (which is to say as a form of self-flattery that appeals to the better possibilities of our nature). A sentiment of benevolence directed only towards humans would seem to be indefensible from the moral, common point of view, corrected by reason and imagination.

Animals and Moral Agency

Hume repeatedly emphasizes the similarity between human and animal minds. He also argues that some animals are capable of sympathy, that moral distinctions are based on sympathy, and that other animals have moral sentiments. One might then expect Hume to conclude that some other animals are moral agents capable of right and wrong action. In fact, he asserts the contrary:

Animals have little or no sense of virtue or vice; they quickly lose sight of the relations of blood; and are incapable of that of right and property. (T.326)

Animals … want of a sufficient degree of reason may hinder them from perceiving the duties and obligations of morality. (T.468)

While Baier (1985, 147) thinks that the qualification ‘little or no’ is significant, I am inclined to doubt that Hume is showing any genuine uncertainty. He uses the idea that animals have moral duties as the conclusion of a reductio argument against moral rationalism (T.467–468). Moreover, in the first Enquiry, Hume uses the idea of a virtuous horse, along with a golden mountain, to illustrate the creative power of combining ideas that are not found together in experience (EHU.19). Hume’s position would appear to be that animals are incapable of moral agency (rather than, say, being lowly or imperfect moral agents).

In his essay On the Dignity or Meanness of Human Nature, he most thoroughly emphasizes the extent of our differences from other animals:

In forming our notions of human nature, we are apt to make a comparison between men and animals, the only creatures endowed with thought that fall under our senses. Certainly this comparison is favourable to mankind. On the one hand, we see a creature, whose thoughts are not limited by any narrow bounds, either of place or time; who carries his researches into the most distant regions of this globe, and beyond this globe, to the planets and heavenly bodies; looks backward to consider the first origin, at least, the history of human race; casts his eye forward to see the influence of his actions upon posterity, and the judgments which will be formed of his character a thousand years hence; a creature, who traces causes and effects to a great length and intricacy; extracts general principles from particular appearances; improves upon his discoveries; corrects his mistakes; and makes his very errors profitable. On the other hand, we are presented with a creature the very reverse of this; limited in its observations and reasonings to a few sensible objects which surround it; without curiosity, without foresight; blindly conducted by instinct, and attaining, in a short time, its utmost perfection, beyond which it is never able to advance a single step. What a wide difference is there between these creatures! And how exalted a notion must we entertain of the former, in comparison of the latter! (Essays, 82)

It is clear that Hume thinks there is at least one difference between humans and animals that is also a necessary condition for moral agency. Various candidates have been proposed, each with some merit. For example, Nuyen (1998) argues that on Hume’s account animals cannot be moral agents because they lack the artificial virtues, and, therefore, lack a sense of justice. Tranöy (1959) suggests that it is because animals are incapable of demonstrative reasoning that Hume does not credit them with moral agency [see further Boyle (2003)]. While Hume thinks that ‘beasts are endow’d with thought and reason as well as men’ (T.176), he has in mind probabilistic rather than demonstrative reason. The minds of animals, Hume writes, ‘are not so active as to trace relations, except in very obvious instances’ (T.397). Pitson (2003) defends a broadly similar explanation and argues that although animals are capable of sympathy, on Hume’s view, it is more like emotional contagion or associative sympathy than the cognitive perspective-taking that is required for the common point of view. Adopting the common point of view also requires imagination, with respect to which Hume sees animals as relatively deficient, being ‘but little susceptible either of the pleasures or pains of the imagination’ (T.397).

One disadvantage of explaining animals’ lack of moral agency by way of their lack of relatively basic capacities such as human-like sympathy or imagination is that it sits awkwardly with Hume’s principle of analogical reasoning. However, there is another feature that seems to preclude animals from moral agency and that has not yet been commented upon. Moral agency requires the capacity not only to act, but to critically evaluate one’s actions and one’s own nature. In his essay Of Polygamy and Divorce, Hume writes:

Among the inferior creatures, nature herself, being the supreme legislator, prescribes all the laws which regulate their marriages, and varies those laws according to the different circumstances of the creature … But nature, having endowed man with reason, has not so exactly regulated every article of his wedding contract, but has left him to adjust them, by his own prudence, according to his own particular circumstances and situation. (Essays, 183)

One needn’t be a fully-fledged Kantian to think that what nature legislates for us in terms of our natural constitution is merely a starting point for a process of moral reflection. The Treatise is primarily a descriptive account of human nature, and the operations and limits of human understanding. However, it remains of interest more due to Hume’s investigations into normative questions than due to his descriptive psychology, sociology and political science. Given that our minds operate thus and so, that we form beliefs about this or that under these circumstances or those, are we justified in doing so? Is induction or our belief in the existence of the external world rationally justified? The problem is that when reason turns to normative questions, including those concerning the justification of its own operations, we find ourselves faced with the sorts of skeptical worries that Hume so memorably and brilliantly describes in the conclusion to Book I. Reason is driven by the passions to defend itself, but subverts itself instead. Yet, the question of justifying our nature comes to seem pressing and unavoidable as we reflect and as we do philosophy. ‘‘Tis almost impossible,’ Hume writes, ‘for the mind of man to rest, like those of beasts, in that narrow circle of objects, which are the subject of daily conversation and action’ (T.271). We may be the only animals capable of the sort of higher-order thought that reveals our own existence as an existential problem to be solved. Our capacity for moral agency, or the kind of moral agency that is distinctive of human life, requires the possibility of grappling with the problem of our own existence and legislating for ourselves. We can consistently suppose that animals possess capacities for sympathy and imagination, licensed by the principle of analogical reasoning, and yet lack the capacity to critically examine their own actions and natures.

Conclusion

According to Hume’s moral theory, for an action to be morally wrong is for it to cause an idea in an observer, which produces painful sentiments of disapprobation when the observer has adopted the general point of view without particular regard for self-interest. So, Hume thinks that virtue and vice depend upon objects producing sympathetic responses in observers who have corrected their passions by reason and imagination. If the thing acted upon cannot produce feelings of sympathy in a suitable observer (or, like a corpse, produces feelings of sympathy that reason judges to be inappropriate because they depend on false beliefs) then we cannot act wrongly with respect to that object. If it were the case, then, that human sympathy suitably methodized and corrected were unresponsive to the suffering and joys of other animals, then it would be right to conclude by Hume’s lights, that they lack moral considerability. However, this is not the case. Animals are relevantly like-minded in such a way that we can come to sympathize with their pains and pleasures once we make the imaginative shift in perspective towards the general point of view. This is so even if animals are not moral agents.

There is nothing in Hume’s account that depends on species membership. Of course, we are more like other humans than we are like pigs, cows or chickens and so we are likely to find it easier and more natural to sympathize with other humans. However, the basis of moral concern between two sentient beings is that their minds are sufficiently alike to allow for the communication of passions in the particular ways that I have described above. Hume’s position must be that humans and some other animals are suitable moral relata. Any animal that cannot methodize and correct its sentiments cannot be a moral agent. However, animals with whom we are relevantly like-minded are the proper objects of our moral concern and humans who cause joy or suffering in those animals are the proper objects of our moral judgments. The project of naturalizing minds and morals is naturally associated with a view of ourselves as continuous with the rest of nature, which tends, in turn, to undermine moral distinctions grounded in species membership.Footnote 14