Navigation – Plan du site

AccueilNuméros10ArticlesSidgwick and the Morality of Purity


Sidgwick and the Morality of Purity

Francesco Orsi


Cet article apporte un éclairage analytique sur les idées de Sidgwick en matière de morale sexuelle. Pour Sidgwick, le plaisir sexuel n’est pas mauvais en tant que tel, ni intrinsèquement, ni intuitivement, ni même de façon dérivative. Mais à l’époque où il écrit, il en vient à penser qu’il est préférable de conserver la notion de « pureté » en morale, car aller à rebours des idées communes en matière d’éthique sexuelle peut entraîner des conséquences plus graves. Comme Sidgwick lui-même nous y invite, l’article contraste ensuite ses idées avec celles de John Stuart Mill. Enfin, il montre comment l’éthique sexuelle de Sidgwick se rapporte directement à ses affirmations sur la morale ésotérique et sur le dualisme de la raison pratique.

Haut de page

Notes de l’auteur

I wish to thank two anonymous referees for their comments. This paper was written while holding a Mobilitas Grant no. MJD111 (European Social Fund).

Texte intégral

1. Introduction

1In the preface to his recently published On What Matters, Derek Parfit en passant praises Henry Sidgwick’s approach on matters of sexual ethics against the charge of moral conservatism sometimes directed at him:

  • 1 The reference is to Williams 2006: 283.

Sidgwick’s irony can make him seem stuffy, when in fact he is being subversive. Bernard Williams had been misled, for example, when he wrote that Sidgwick’s discussions of sexual morality, though sometimes mildly adventurous, ‘make fairly uncritical use of a notion of purity’. Sidgwick does ask ‘What, then, is the conduct that Purity forbids?’ But if we read him carefully, we find that his answer is: Nothing. In a Book published in England in 1874, it was more than mildly adventurous to argue, though in guarded terms, that there is no moral objection to indulging in sexual pleasure for its own sake (Parfit 2011: xxxviii).1

2I will try to make sense of both Parfit’s and Williams’ reactions by a close analysis of Sidgwick’s text. Parfit is right to the extent that Sidgwick saw nothing intrinsically wrong in getting sexual pleasure for its own sake. Indeed, we will see that Sidgwick does an unprecedented job of scrutinizing and rejecting both moralistic and what we may call “prudentialistic” views of sexuality. On the other hand, Williams is right insofar as Sidgwick does not come clean on the possibility of discarding the moral concept of sexual purity altogether—though it surely is an overstatement to claim that he continues to “make a fairly uncritical use” of it. Moreover, Sidgwick suggests that the best utilitarian code for his own times must include some norm of sexual purity.

3In order to make sense of Parfit’s and Williams’ reactions, we must thus appreciate two ethical questions that Sidgwick regarded as theoretically separate. The first is whether a certain kind of behaviour is intrinsically wrong. The second is whether, at a certain point in history, we ought to continue prohibiting that kind of behaviour, even if it is morally permissible. A negative answer to the former question does not determine a negative answer to the latter. The distinction is based by Sidgwick on familiar consequentialist reasons. Even if there are no bad consequences in every single act of a certain kind, and so there are no grounds to regard that type of act as wrong, the overall consequences of attempting to modify common sense in this respect might yet be worse, at least at a certain time, than leaving the social sanction intact. As will be shown, this seems to be Sidgwick’s overall view on the ethics of sex and moral reform. Focussing on the first question helps us understand Parfit’s appraisal, whereas focussing on the second one makes sense of Williams’ remark that Sidgwick’s view on these matters is only mildly adventurous.

  • 2 I will use The Methods of Ethics, 7th edition, Indianapolis, Hackett, 1981.

4I will proceed by critically analyzing Sidgwick’s claims regarding sex in TheMethods of Ethics (ME from now on).2 In sec. 2 his interpretations of the common sense principle of sexual purity are outlined. In sec. 3 I consider the serious and somewhat special epistemological difficulties surrounding the alleged self-evidence of purity. Sec. 4 shows how Sidgwick’s doubts extend to the derivative value of purity. For all that he says so far, it is fair to wonder whether the principle of purity is one article of common sense morality which, after critical scrutiny, could be discarded once and for all. But such a question is to be decided on utilitarian grounds. And Sidgwick’s answer is, overall, negative (sec. 5). He points out the socially beneficial consequences of a circumscribed principle of purity as conjugal fidelity—with the controversial qualification that such principle is more stringent for women than for men. Also, and crucially, he seems to favour the inculcation of a generic aversion to sexual satisfaction sought for its own sake. In sec. 6 Sidgwick’s view is briefly contrasted with John Stuart Mill’s, whom he directly mentions in this connection. Finally, sec. 7 explores how Sidgwick’s ethics of sex relates closely to his claims about esoteric morality and to the dualism of practical reason.

5The aim of this work is mainly to bring analytically to light Sidgwick’s complex views on sexual morality. Little will be said over whether his views are correct—for his times, and for ours. Part of the objective thus is to present Sidgwick’s contribution to the history of philosophical ethics of sexuality as one worthy of considerable attention, if only for its analytical detail. Secondly, Sidgwick’s discussion provides a striking example of how a destructive epistemological analysis of moral principles can, coherently, go along with a cautious social philosophy. In particular, it will become clear how intuitionism, understood as the quest for self-evident moral principles, represents here a reformist force to be contrasted with the socially prudent outcome of act-utilitarianism, thereby inspiring some scepticism for act-utilitarianism’s traditional association with a progressive moral philosophy.

2. Articulating Purity

  • 3 He probably got the term “Purity” from William Whewell. Actually, for Whewell Purity regulates bodi (...)

6The explicit treatment of the ethics of sex can be found in ME, Book III, Ch. IX (Self-regarding Virtues), § 3, and Ch. XI (Review of Common Sense), §§ 7-8; and Book IV, Ch. III, § 6. Sidgwick introduces the virtue or duty of purity, or chastity, as that part of common sense morality that regulates sexual appetites and practices. In a footnote he says that the notion of chastity is “somewhat more external and superficial” than purity (ME 330). It is not clear what this means: as the discussion shows, purity refers to sexual acts rather than merely inner sexual desire. Perhaps, by common sense’s lights, purity refers to sexual acts when these are seen as part of an overall moral ideal of sexual conduct, whereas chastity characterizes nothing beyond the acts themselves.3

  • 4 Sidgwick does not define what exactly counts as the satisfaction of sexual appetite. He can be assu (...)

7As with other common sense moral ideas, in Book III Sidgwick’s aim is to show that there is no self-evident principle of purity that can be obtained from common sense morality. In Ch. IX he sets out to critically describe various articulations of purity. “Pure conduct” in general characterizes morally permissible indulgence to sexual appetite.4 Different formulations of purity make clear what it is that may make indulgence to sexual appetite morally permissible. The typical content of a principle of purity therefore takes the form:

  • (P1) Indulging to one’s sexual appetites is wrong, unless…

8Hence there is in common sense a default assumption as to the wrongness of responding indulgently to one’s sexual desires, i.e., by wanting and so trying to satisfy them. Sidgwick does not exactly state this assumption, but considers two related claims:

  • The sexual appetite ought never to be indulged for the sake of the sensual gratification merely, but as a means to some higher end. (ME 330)

  • The gratification of merely sensual impulses is in itself something objectionable. (ME 331)

  • 5 Presumably the idea is that common sense does not forbid sexual conduct which has sexual pleasure i (...)

9The first claim both adds something to P1 and is less general than P1. Of course, it needs to be revised as it stands because of the ambiguity of “but”. It does not say: sexual appetite ought never to be indulged etc., but it ought to be indulged when it is a means to a higher end! It rather says: sexual appetite ought never to be indulged etc., but is permissibly indulged when it is a means to a higher end. Hence, the first claim adds to P1 by specifying that what is by default wrong is sex for the sake of sexual pleasure only, whereas P1 makes no mention of the character of the indulgence. On the other hand, the first claim is less general than P1, for it only allows for permissibility in the case that sexual activity is a means to a higher end.5 But, as the discussion will shortly show, common sense at times seems to recognize that sexual activity, even for its own sake, is permissible when carried out by people standing in a certain kind of relationship, legally recognized or morally permissible. Hence in some cases it may be the objective circumstances of the agents involved rather than the character of sexual conduct that which makes it permissible. So we arrive at:

  • (P2) Indulging to one’s sexual appetites for its own sake is wrong, unless…

10The second claim quoted, which Sidgwick refers to as “the ascetic opinion”, is a more general statement than P2, in that it takes gratification of all merely sensual impulses to be in itself objectionable, the sexual impulse being one of them. But Sidgwick notes that such opinion is not consistently maintained by common sense: “the most intense enjoyment of muscular exercise, or warmth, or bathing” are not normally condemned (ME 331). So there must be something specifically objectionable about the gratification of sexual impulse.

11One hypothesis is that there is a further unstated evaluative assumption, such as that sexual enjoyment, unlike the enjoyment of warmth (say), is intrinsically bad, and that therefore it ought to be avoided as far as it goes. (Note: quite obviously, to say that sexual enjoyment is a lower sort of pleasure, something which Sidgwick seems to allow for, does not settle the question whether it is in itself a bad sort of pleasure.) However, such an assumption would cry out for justification just as much as P2 itself, and indeed Sidgwick does not ascribe it to common sense.

12Another hypothesis could be a Kant-inspired idea that sexual activity essentially involves treating humanity, in others and in ourselves, merely as a means (to one’s satisfaction). Sidgwick does not mention this idea either, nor should he, for it expresses a philosophical rather than common sense claim. Nor is it clear that reflective common sense would find the Kantian claim persuasive. So the attempt to reconstruct a common sense principle of purity already brings out the groundlessness of any such principle, when interpreted as P2. But it is instructive to analyse, with Sidgwick, the different reasons why sexual conduct could be permissible, i.e. different ways of completing P2.

13A first formulation forbids indulgence to sexual appetite except within “the limits of the union sanctioned by law” (ME 330). But common sense views purity as holding over and above legally authorized unions. First, “conformity to this does not necessarily secure purity” (ibid.): presumably meaning that not any episode of indulgence to sexual appetite within a lawful union such as marriage is pure, i.e. morally permissible. Rape against one’s wife may be a case in point, though Sidgwick does not mention any here. Also, one might add, even if law makes it permissible to rape one’s wife, i.e. to have sex with her despite her denial of consent, this would not suffice to make such behaviour morally permissible. Moreover, “all illegitimate sexual intercourse is not thought to be impure” (ibid.): that is, common sense accepts that there may be morally permissible sexual conduct outside lawful unions. It would be nice to know what kind of cases Sidgwick refers to here. Perhaps he is simply thinking about heterosexual intercourse among unmarried adults.

14If common sense seems to refuse by its own lights this legalistic formulation of purity, it appears to be internally divided as to the correct one. There is admittedly a common ground, namely the claim quoted above: “the sexual appetite ought never to be indulged for the sake of the sensual gratification merely, but as a means to some higher end” (ibid.). The disagreement concerns what these higher ends are. On a stricter view, which belongs to the natural law tradition, it is the propagation of the human species, as the “primary natural end” of sexual appetite. Therefore, sexual activity is wrong, unless it is a means to procreation. This would make purity a distinctively social duty or virtue, whose point is to provide for the continuity of the human race. However, for Sidgwick common sense does not regard this as the only possible pure conduct (ME 331). He might have added that it is not even a sufficient criterion for purity: again, rape can be intended as a means to procreation but this doesn’t make it morally permissible.

15On a different, laxer, view, sexual appetites ought not to be indulged in, except as a means to the “development of mutual affection in a union designed to be permanent” (ibid.). Roughly, sex is morally permitted only as a contribution to a loving and durable relationship. Common sense thus acknowledges that the “low” pleasures of sexual enjoyment can be instrumental to the “higher” one of a durable mutual affection. This is a more liberal view, because it allows for non-procreative sex, and does not narrow down the suitable kind of unions to legally authorized ones. Moreover, it does not say that any sexual intercourse within an affective relationship is thereby permissible: sex not intended to develop or express mutual affection in a relationship is not permissible. So presumably this view forbids rape, even when it happens within a durable and overall loving relationship. Of course, some of us may feel that the “laxer” view is still too strict, as it does not remove a prohibition against all sexual encounters that do not develop mutual affection, or that happen outside a permanent relationship, or anyway not in view of a possible relationship.

  • 6 See the following footnote.

16That said, as Sidgwick notes, “the practical difference between the two views is considerable” (ibid.). The procreative view would forbid a lot of sexual activity, e.g. the use of contraceptives, that is perfectly permissible by the affective view. What is more, if we are to go by Sidgwick’s wording, the affective view would allow for homosexual intercourse, provided it contributes to mutual affection in a homosexual union designed to be permanent. It is however not clear that Sidgwick regarded even a small part of common sense morality of his time as accepting homosexuality in some form. Had he thought so, he would not have been so resolute in his advice not to publish his friend John Addington Symonds’ homoerotic poetry, and in his insistence to omit any reference to homosexuality in Symonds’ posthumous biography.6 Of course Sidgwick must have thought that common sense is incoherent insofar as it accepts the affective view, but refuses to draw the implications as to the permissibility of homosexuality. For there is no obvious reason why the values of love and mutual affection cannot be realized and expressed in a homosexual union.

17Moreover, insofar as Sidgwick here is articulating common sense, he claims that common sense would avoid any specific treatment of the issue, for a minute and detailed moral regulation of sexual activity would get the mind to contemplate precisely that kind of conduct that purity forbids, whatever that is, and be thereby tempted to impure acts (ibid.). Purity thus emerges as a possibly self-defeating moral concept: the more we know about what purity forbids, the less we are likely to abide by it. In a sense, then, the impreciseness of the duty of purity is not as bad, epistemically, as that of other common sense duties analyzed by Sidgwick: the range of forbidden and permissible activities may be stated with precision, but we need not possess more than a vague sense that there is something wrong in satisfying our sexual appetite for its own sake.

18To sum up: reflective common sense seems to reject the legalistic view of purity, but is internally divided between the procreative and the affective views. The latter have a common core, namely P2, but by the end of Book III, Ch. IX, §3, we already know that common sense rejects a possible support for P2 coming from a more general claim about the badness of sensual enjoyments (the ascetic opinion). And it is not clear what further support can be adduced. The default assumption as to the wrongness of sexual indulgence for its own sake remains groundless.

19The diagnosis is already grave from this first overview. Purity has no obvious commonsensical or philosophical justification. Further, it has a general content, but not a specific one that is uniformly shared within the same society, and much the less across different societies (it is in this connection and not, say, in the discussion of promise-keeping, that Sidgwick stresses that he is examining “the Common Sense of our age and country”). Finally, insofar as it has a specific content, it is one that is best not known in detail.

3. Purity and Self-Evidence

20In Chapter XI of Book III Sidgwick proceeds to a critical examination of common sense. The question is whether common sense principles “possess the characteristics by which self-evident truths are distinguished from mere opinions” (ME 338). The characteristics can be so described (ME 338-342):

  1. self-evident propositions must contain clear and precise terms;

  2. self-evident propositions must be able to be distinguished, after careful reflection, from mere strong sentiments or customary opinions;

  3. self-evident propositions must be mutually consistent;

  4. there must not be disagreement over self-evident propositions, at least among epistemic peers.

21Of course, these four conditions are marks of self-evident truths rather than being definitory of self-evidence. No condition in fact mentions what makes a proposition self-evident. They are only meant to provide a negative test: if a proposition fails to respect any of them, it is not self-evident. But it is not the case that if a proposition respects all of them, then it is self-evident (many ordinary empirical propositions respect all of them).

22It does not come as a surprise that principles of purity do not pass the test of self-evidence. However, it is noteworthy that he reserves a whole section to the discussion (§7). At this stage, Sidgwick says, “it seems necessary to override” the aversion to thorough scrutiny that purity itself enjoins (ME 357). The major difficulty, as we know, is to find a formulation of purity that respects condition (4). Reflective common sense rejects the legalistic view that all and only sex within a conjugal union as defined by law is permissible. As Sidgwick remarks, “we feel that positive law may be unfavourable to Purity, and that in fact Purity, like Justice, is something which the law ought to maintain, but does not always” (ibid.): particular legal codes may wrongly allow what common sense would regard as impure intercourse (like rape, one may again suppose).

23But neither the procreative view nor the affective view are able to gain general agreement. First, many reject the idea that all sexual intercourse should be prohibited, except as a means to the propagation of the human species. This would make all non-procreative conjugal intercourse wrong. Second, the affective view can be rejected by those who hold the procreative view. Interestingly, the affective view can also be rejected as itself too strict: “we do not condemn marriages without affection as impure, although we disapprove of them as productive of unhappiness” (ibid.). That is, there is in common sense the notion that mutual affection is not necessary to make conjugal sexual intercourse morally permissible, although that surely does not make for a happy marriage. Unhappy married couples can permissibly have sex for pleasure’s sake even though they have ceased loving each other, or indeed have never done so. Sidgwick attributes here to (a part of) common sense the liberal thought that morality should, to a certain extent, refrain from nosing around the bedrooms of married adults, however unhappy their relationships may be.

24This is of course sufficient to deny self-evidence to any principle of purity. But also widely shared convictions fail to pass the test. For instance, the prohibition against incestuous relationships lacks self-evidence. Insofar as the exact degree of forbidden consanguinity is left indefinite, or is the object of disagreement across different societies, conditions (1) and (4) are not met. And, even when it is made definite, it is unclear what the rationale for a degree of consanguinity rather than another should be. On rational inspection, the ban on incest seems to express no more than “a peculiarly intense moral sentiment” (ME 358), thus violating condition (2). Inter- and intra-cultural disagreement also show that it is not self-evident that marriage ought to be monogamous and even “designed” to be permanent (ibid.; see also ME 255 and 347-8). We can fairly conclude that if the only agreement on purity is around the uncompleted principle P2, then the very idea of purity is both imprecise—vs. condition (1)—and not distinguishable from a strong feeling or a customary opinion—vs. condition (2).

4. Derivative Value of Purity

25Sidgwick proceeds to consider what else, if anything, can be said in favour of at least some moral regulation of sexual conduct. He acknowledges two ends that common sense takes to be promoted by the “current sexual morality” (ME 359). One is social: “the maintenance of a certain social order, believed to be most conducive to the prosperous continuance of the human race”. The other is self-regarding: “the protection of habits of feeling in individuals believed to be generally most important to their perfection or their happiness” (ibid.). Now, first, even if a pure conduct were always a necessary means to either end, this would not make a principle of purity self-evident: “if the repression of sexual license is prescribed merely as a means to these ends, it does not seem that we can affirm as self-evident that it is always a necessary means in either case” (ibid.). Any air of self-evidence definitively vanishes.

26But is a pure conduct really necessary to achieve these social and individual ends? If so, the principle of purity could still have a derivative force, and the internalization of some principle of purity, however imprecise, intersubjectively contested and intrinsically groundless, would at least find a justification in the social or prudential interests that could be fostered. Sidgwick’s last stab is to argue that, generally speaking, purity also lacks derivative value.

27First, even if these ends were served by a regulation of sexual conduct, in practice there could hardly be a regulation that would serve both of them equally well (ibid.). There are sexual conducts which would be allowed for the sake of the prosperous continuance of the human race, but which might easily conflict with the individual’s perfection or happiness. An example could be sex within arranged marriages. On the other hand, sexual conducts which do not favour the propagation of the human species might be the best for the individual’s happiness: protected sex or homoerotic sex being two obvious cases. And, at least at this stage, we cannot rationally prefer one type of ends to another. Indeed, we see here that Sidgwick’s dualism of practical reason between utilitarianism and rational egoism—supposing the prosperous continuation of the human species to be the utilitarian’s preferred option—makes the prospect of a coherent set of principles for sexual conduct rather unlikely.

28Second, can we anyway show that sexual license is harmful to the “maintenance of population in sufficient numbers and good condition” (ibid.)? Only sociological and historical observation can help us here. Moreover, the answer will depend on exactly which sexual license is forbidden. Societies in which e.g. homosexuality was not condemned clearly could maintain their population in sufficient numbers and good conditions. Is sexual license harmful to the individual’s perfection and happiness? Here a little comment is in order. If it could be shown that sexual license—indulgence to sexual desire for its own sake—is bad for the individual, purity would become primarily a matter of prudence, or a derivative duty to oneself. In Sidgwickian terms, sexual pleasure sought for its own sake would be something that each of us, considering our existence only, would not desire, were our desires in harmony with reason. Of course, once we take “the point of view of the universe”, we take account of what is bad for any individual, and therefore we would reasonably desire that nobody indulges in sexual pleasures for their own sake. But the source of such judgment would be that mere sex makes us unhappy, or degrades us to the condition of brutes.

29Sidgwick seems keen to reject also this “prudentialistic” view. Happiness, as defined by the net balance of agreeable states of consciousness, does not seem generally to be jeopardized by sexual activity. Indeed, if sexual pleasures do not compromise the attainment of other, greater and longer-lasting pleasures, there is no reason why we should not try to satisfy them, even for their own sake. As for perfection, first, Sidgwick claims that experience does not universally show that mere sexual relations interfere with the development of the intellectual, “higher” mental faculties (ME 359). Second, Sidgwick, echoing Millian themes, admits that sex for its own sake provides for a lower kind of sentimental development of the individual than richer affective relationships do. But common sense seems to condemn such development in comparison with no sentimental development at all. And this opinion is completely unjustified: however imperfect, the development afforded by merely sexual relations, at least, cannot be worse than the lack of development that we allow in the case of celibacy (ibid.).

30In short, Sidgwick presents the common sense morality of his time as profoundly divided and unclear when it comes to sexual matters. Not only is there no principle of purity which can claim the status of self-evidence. In this respect purity is epistemically as weak as other common sense principles. Also purity’s derivative normative force is here questioned, something that does not happen to other common sense duties such as respect of promises or truth-telling.

31Where does this leave us? A conclusion one might be tempted to draw is that there is no such thing as a specific branch of morality as specified by the duty of purity. Moreover, the connection of sexual moral regulation, as defined by purity, to the general and the individual’s interest is at best shaky. Sidgwick, however, ends his review of common sense on a cautionary note:

The notions of Benevolence, Justice, Good Faith, Veracity, Purity, etc. are not necessarily emptied of significance for us, because we have found it impossible to define them with precision. The main part of the conduct prescribed under each notion is sufficiently clear: and the general rule prescribing it does not necessarily lose its force because there is in each case a margin of conduct involved in obscurity and perplexity, or because the rule does not on examination appear to be absolute and independent…The Morality of Common Sense may still be perfectly adequate to give practical guidance to common people in common circumstances (ME 360-1).

32Is the notion of purity still meaningful and clear enough to give us practical guidance in sexual matters? It does not seem that those who have followed Sidgwick’s scrutiny step by step will be able to retain a notion of purity good enough for practical purposes. If there is no plausible way of completing P2, and no obvious social and individual interests are at stake, we will simply not know whether a given sexual act or type of act is morally acceptable. And we do not need to be “uncommon” people or find ourselves in “uncommon” circumstances to feel that we lack the guidance we want. For instance, we will not know whether a perfectly common sexual act between unmarried heterosexual consenting adults is morally permissible or not.

33There remains admittedly a generic content to purity: the default prohibition against sex for its own sake. In this sense, perhaps, purity is not “emptied of significance”. But now it looks like the one conduct which is guaranteed to accomplish purity is complete sexual abstinence. For the proposed exceptions to the default prohibition (conjugal sex, procreative sex, sex as expression of affection) have not been proved satisfactory from the point of view of purity. The problem, of course, is that if the only practical guidance we are left with is the prescription of complete abstinence, we have moved too far off common sense morality itself.

34Sidgwick hesitates to acknowledge the radical nature of his analysis. But sexual morality as defined by the standard of purity seems precisely an article of common sense that calls for profound revision: indeed, the disagreements Sidgwick registers may show that moral revision is already on its way in some sectors of society.

5. Utilitarianism and Sexual Reform: Sidgwick and Mill

35Sidgwick returns to purity in Book IV, Ch. III, § 6, as he examines the relation between utilitarianism and common sense. The overall aim of the chapter is to show how utilitarianism “sustains the general validity of the current moral judgements” (ME 422), and provides the best way to resolve conflicts between common sense principles. As said at the outset, Sidgwick distinguishes the question of whether and which sexual conduct is morally wrong, from the question of which sexual permissions and prohibitions we ought to inculcate. We already know, however, that not only do the injunctions of purity lack self-evidence, but they also fail to serve social and prudential interests in any straightforward way. Given these premises, we should hardly expect utilitarianism to support the general validity of the current moral judgements about sexual conduct, and thus to favour the inculcation of the ideal of purity.

36Sidgwick indeed begins § 6 by remarking that the ascetic condemnation of sexual appetites is apparently anti-utilitarian, in that it is directed at activities whose “immediate effect is pleasure not obviously outweighed by subsequent pain” (ME 450), for they do not normally lead to dangers for health or, as he has told us above, interfere with the development of faculties and sentiments which are major sources of happiness for the individual. However, this does not mean that utilitarianism recommends the abandonment of the notion of purity: the case of purity “shows a specially complex and delicate correspondence between our moral sentiments and social utilities” (ibid.).

37Sidgwick does not try to justify on a utilitarian basis all moral sentiments related to purity. The principle that he is keen to preserve is the rather circumscribed one of conjugal fidelity. A widely internalized rule against extra-nuptial intercourse is held by Sidgwick to be a necessary means to “the maintenance…of the permanent unions which are held to be necessary for the proper rearing and training of children” (ME 450). And of course the proper rearing and training of children is indispensable for overall happiness (ME 435). Admittedly, utilitarianism also seems to justify the belief, still widespread at Sidgwick’s time—and to some extent at ours—that women’s unfaithfulness is worse than men’s: “the degradation of this standard must strike at the root of family life, by impairing men’s security in the exercise of their parental affections; but there is no corresponding consequence of male unchastity” (ME 451). Sidgwick adheres quite uncritically to the traditional idea that the internalization of a stricter requirement of chastity for women makes men sure that it is their own children they are taking care of as fathers. It is not obvious that this sexist double standard makes utilitarian sense. As Sidgwick acknowledges, female unchastity impairs the existence of the family, but male unchastity impairs its well-being nonetheless (ibid.). And if families are a necessary setting for the proper growth of children (but see ME 346), surely they had better be happy families. But of course, the point remains that, as far as utilitarianism goes, if the consequences of people accepting the double standard—including the unfairness to women—are overall better than those of accepting a non-sexist one, then people ought to adopt the double standard.

38On the other hand utilitarianism condemns, in general, sex with prostitutes, despite the opinion of “men of the world” that such conduct by men should be tolerated for alleged utilitarian reasons. The lives of prostitutes tend to be unhappy as the object of social contempt and exclusion; moreover, it is important to maintain “that higher type of sexual relations which is not, generally speaking, possible, except where a high value is set upon chastity in both sexes” (ME 452). In sum, “the Virtue of Purity [as conjugal fidelity] may be regarded as providing a necessary shelter under which that intense and elevated affection between the sexes, which is most conducive both to the happiness of the individual and to the wellbeing of the family, may grow and flourish” (ibid.).

39With this, “the main central portion of the region of duty [has been] strongly illuminated” (ME 453). But, since there are strong temptations to violate this rule, utilitarianism also favours a more general aversion to impurity, one which extends “far beyond the acts that primarily need to be prohibited, and include in its scope everything (in dress, language, social customs, etc.) which may tend to excite lascivious ideas” (ibid.). However, we cannot aspire to a more detailed regulation, as this “must vary to a great extent from age to age and from country to country” (ibid.).

40Sidgwick’s treatment is somehow double-faced. On the one hand, the utilitarian restoration of some notion of purity does not by itself signify a morally conservative retreat. First, the concept has lost its traditional ascetic connotation. In other words, no general claim about the moral badness of satisfying one’s sexual appetite is vindicated on utilitarian grounds. As is evident, the principle no longer has the form of an exception to a rule which prohibits an otherwise objectionable conduct—indulgence to sexual appetite. The prohibition is directed at a specific impure act, adultery. At bottom, utilitarianism only supports the preference for a society where adultery is forbidden to one where it is not.

41Second, even adultery is condemned only as a perturbation of a contingent social arrangement. At ME 435 Sidgwick reminds us of the obvious utilitarian reasons for providing children with adequate nourishment, protection, and education, and states the “common belief” that family, as regulated by, inter alia, the duty of conjugal fidelity, is the best or the only means of attaining these ends to a tolerable degree. But he soon adds that the topic of “domestic duties” “forms an arena for continual disputes”, over which no a priori moral judgement can be passed. Moreover, Sidgwick was free from any particular bias in favour of traditional family, as he writes while discussing parental duties: “It may be plausibly maintained that children would be better trained, physically and mentally, if they were brought up under the supervision of physicians and philosophers, in large institutions maintained out of the general taxes. We cannot decide a priori which of these alternatives is preferable” (ME 346).

42On the other hand, Sidgwick is markedly cautious in stating his views. The suggestion that, given the powerful temptation to adultery, utilitarianism recommends a generic aversion to impurity, over and above actually unfelicific sexual habits, is in this respect more telling than it seems. For one, if utilitarianism recommended us to single out only adulterous sex as morally forbidden, while leaving all other sexual activities to be regarded as morally neutral, then there would not be utilitarian reasons for continuing to use the very moral category of purity. Conjugal fidelity could be seen as an instance of the duty of promise-keeping, or as a duty of affection towards one’s spouse and children. The common sense idea of special requirements applying in the area of sexuality would cease to have not only a definite content—that much should have already been established—but a practical point. However, Sidgwick’s remark about the usefulness of a generic aversion goes precisely in the direction of preserving the moral concept of purity, however vague and widely contestable it may be. In this respect Bernard Williams is right to describe Sidgwick’s treatment of sexual morality as no more than mildly adventurous.

43Furthermore, by recommending the preservation of a vague aversion to impurity, utilitarianism, by Sidgwick’s lights, disfavours not only the aim to do without purity altogether, but even the more modest one of reforming public opinion (and eventually law) on particular sexual matters, such as e.g. homosexuality. Insofar as its being vague is necessary for the aversion to be effective, any public attempt to make the aversion more focussed, through a scrutiny of what does and does not need to be prohibited, must pro tanto weaken the aversion and thus involve some disutility.

44Now, showing that, in these and similar issues, common sense morality has some utilitarian point is not tantamount to accepting en bloc the positive morality of one’s time on utilitarian grounds (ME 463). For there may be stronger utilitarian reasons for modifying this or that aspect of common sense. As a utilitarian, and as someone closely involved with people having to cope with the burden of social and legal sanctions for their sexual inclinations, Sidgwick must have thought through, if only in private, the prospects of a moral reform in sexual matters.

45In general his claims about a possible utilitarian “innovation” of positive morality are conservative-sounding: utilitarianism will endorse a form of society “varying but little from the actual, with its actually established code of moral rules and customary judgments concerning virtue and vice” (ME 474). Moreover, “a great part of the reform in popular morality, which a consistent Utilitarian will try to introduce, will probably lie not so much in establishing new rules (whether conflicting with the old or merely supplementary) as in enforcing old ones” (ME 484).

46Applying what he says to sexual morality, on the one hand, Sidgwick is certainly aware of the unfelicific tendency of the generalized prohibitions of common sense. Talking about prostitution, he recognizes in that “so long as the social sanction is enforced, the lives of [those] against which society issues its ban must tend to be unhappy from disorder and shame, and the source of unhappiness to others” (ME 452, n.1). The same could be said about the consequences of the social sanction against homosexuality. But, for Sidgwick, an attempt at reform in his times might be even worse than the status quo. First, as just seen, insofar as the vague aversion to impurity serves the purpose of maintaining a social order where families can thrive, there is one specific utilitarian reason against moral change in this respect. Second, there are general utilitarian reasons against moral innovation as suCh. A reform in sexual morality would take a disruptive character, by working to make morally permissible or indifferent ways of conduct currently deemed immoral. And disruptive reform in general has immediate bad consequences for the reformers themselves and their close relations, and jeopardizes their other efforts:

  • 7 This passage resonates with Sidgwick’s life. As we know from Bart Schultz’s biography, Sidgwick’s e (...)

[A]s his [the utilitarian reformer’s] own happiness and that of others connected with him form a part of the universal end at which he aims, he must consider the importance to himself and them of the penalties of social disapprobation which he will incur: taking into account, besides the immediate pain of this disapprobation its indirect effect in diminishing his power of serving society and promoting the general happiness in other ways. (ME 481)7

47Moreover, Sidgwick might have thought that a disruptive sexual reform would also have wider negative consequences, for instance by weakening an agent’s overall moral habits (ME 482-3).

48Third, we have seen that Sidgwick registers the existence of divergent opinions as to when sex is morally permissible: the legalistic view, the procreative view, the affective view, and possibly also a more liberal view (i.e. the one that does not morally condemn sex without love, at least in a married couple). But such moral disagreement even within the same society may often be the utilitarian best state of affairs for Sidgwick: “it may conduce most to the general happiness that A should do a certain act, and at the same time that B, C, D should blame it” (ME 491). He in fact makes a claim about “relaxation of rules” which seems relevant here: “it may possibly be expedient on the whole that certain special relaxations of certain moral rules should continue to exist in certain professions and sections of society, while at the same time they continue to be disapproved by the rest of society” (ME 491-2). It may thus be advisable from the utilitarian point of view not to eradicate some traditional condemnation of certain types of sexual conduct.

49Finally, Sidgwick actually says something specific, if only in passing, about a possible liberalization of sexual morality. He openly criticizes, from a utilitarian standpoint, John Stuart Mill’s view in On Liberty, thus summed up: “the individual is not responsible to society for that part of his conduct which concerns himself alone and others only with their free and undeceived consent”:

  • 8 Here is the relevant passage from On Liberty: “But with regard to the merely contingent, or, as it (...)

[O]wing to the complex enlacements of interest and sympathy that connect the members of a civilised community, almost any material loss of happiness by any one individual is likely to affect some other without their consent to some not inconsiderable extent. And I do not see how it is from a utilitarian point of view justifiable to say broadly with J. S. Mill that such secondary injury to others, if merely “constructive or presumptive”, is to be disregarded in view of the advantages of allowing free development to individuality; for if the injury feared is great, and the presumption that it will occur is shown by experience to be strong, the definite risk of evil from the withdrawal of moral sanction must, I conceive, outweigh the indefinite possibility of loss through the repression of individuality in one particular direction. (ME 478)8

  • 9 See Ryan 1991: 165-7, and Ten 1991: 225.

50And in a telling footnote, he remarks: “It may be observed that Mill’s doctrine is certainly opposed to common sense: since (e.g.) it would exclude from censure almost all forms of sexual immorality committed by unmarried and independent adults” (ibid.). For instance, Mill’s doctrine would exclude from moral censure consensual homosexual love between unmarried adults (though Mill does not actually mention homosexuality in this connection). First, such behaviour, as far as it goes, would pertain to a part of conduct for which the individual is not responsible to society. Public opinion can at most issue prudential advice on these matters. Second, even if homosexuals lived an unhappy life, the reverberation of this unhappiness on others would need to be “disregarded in view of the advantages of allowing free development to individuality”. Third, a fortiori such unhappiness would need to be disregarded if, as seems to be the case, the sad condition in which homosexuals are forced to live in Sidgwick’s time is in the first place a consequence of a moral (and legal) stigma that has already been found to be prima facie unjustified.9

  • 10 One can trace a markedly conservative development from Jeremy Bentham to Sidgwick. As is well known (...)

51So, if Mill’s doctrine opposes common sense in particular in sexual matters, and is not recommendable in general on utilitarian grounds, Sidgwick must have concluded that a liberal reform of common sense in sexual matters, as informed by a doctrine such as Mill’s, would not be recommendable.10

6. Purity, Esotericism, and the Dualism of Practical Reason

52The morality of purity is far from being self-evident. Nor does it seem to serve immediate individual purposes. However, a utilitarian code (fit for Sidgwick’s time) must include some specific duty of purity—as conjugal fidelity—and inculcate a generic disposition to be averse to the gratification of the sexual appetite for its own sake. A public attempt at a liberal, Millian reform is very likely not advisable from a utilitarian point of view. But does Sidgwick suggest what is right for each of us to eventually do? Even if there were utilitarian reasons for treating conjugal fidelity as an exceptionless duty, there would remain a vast grey area regarding the intimate conduct of “unmarried independent adults”. As a utilitarian of the “act” type, Sidgwick points out the possibility of individual exceptions to rules established on utilitarian grounds,

[W]here the agent does not think it expedient that the rule on which he himself acts should be universally adopted, and yet maintains that his individual act is right, as producing a greater balance of pleasure over pain than any other conduct open to him would produce. (ME 486)

53So, from Sidgwick’s point of view, mutual sexual gratification, for its own sake, between consenting unmarried adults may be morally right or permissible, insofar as it is productive of pleasure and not impairing either’s happiness or perfection.

54However, it may be not expedient that the rule which states the permissibility of such behaviour be universally adopted, on the grounds that such adoption could weaken people’s otherwise useful disposition to purity. Sidgwick’s ethics of sex seems in fact to furnish a concrete application of his notorious claims about an esoteric morality. On utilitarian grounds, consensual harmless sex for its own sake may be right to engage in privately but it would be wrong to advocate its permissibility “in the face of the world” (ME 489). And a utilitarian probably could even desire, on utilitarian grounds, that his conclusions about the permissibility of consensual harmless sex for its own sake “be rejected by mankind generally” (ME 490).

  • 11 See note 6.

55For Sidgwick, then, Millian experiments in living of a sexual nature could permissibly be carried out, as long as they are sufficiently kept from the public eye. Now a policy of complete secrecy may also be in the best interests of people like (e.g.) Sidgwick’s homosexual friends in Victorian Britain. To that extent, a utilitarian morality and egoism would agree in their conclusions. But a life of continual concealment and fear of social (and legal) punishment involves a considerable amount of self-sacrifice over time, and it cannot be excluded that some of Sidgwick’s friends might have overall been better off by allowing some degree of publicity to their unorthodox lifestyles (e.g. by publishing works related to homosexuality under their real names11), even in the presence of social and legal sanctions. In this eventuality, Sidgwick’s well known dualism of the practical reason would present itself. The egoistically right conduct would not coincide with the morally right, i.e. utilitarian one, on account of its supposedly adverse effects on the moral dispositions of the larger public.

56It might be thought that this very fact does something to alleviate the conservative consequences of Sidgwick’s views. For one, since egoism is one ultimate principle of practical reason, the claims of open expression of one’s individuality would find a legitimate recognition within practical reason. Moreover, since, in the case of conflict between egoist and moral demands, practical reason cannot appeal to a higher principle to settle the conflict, it would not be the case that one ought, all things considered, choose a policy of complete secrecy in one’s sexual lifestyle. The choice of relative publicity would be as reasonable as the choice of secrecy. However, egoist arguments look like the wrong kind of defence. To insist on the example of homosexuality, the strive for recognition is obviously not directed at any old normative recognition, but specifically at one from the moral point of view—one that ideally rules out any form of moral disapproval towards that kind of conduct. Indeed—and this is Mill’s claim—one may struggle for others’ recognition of one’s lifestyle as morally permissible even despite of its being overall productive of unhappiness for oneself. An endorsement by practical reason on egoist grounds clearly falls short of such an aspiration.

7. Conclusion

57The aim of this work has been to show the complexity of Sidgwick’s ethics of sex. The critical scrutiny of the common sense principle of purity fairly leads to the conclusion that the indulgence of the sexual appetite for its own sake is not in itself morally objectionable. Since the principle of purity presents itself as a prohibition against sex for its own sake, to which various exceptions may be attached (e.g. sex within an affective relationship), the very form of sexual morality based on purity is shown to be philosophically indefensible. However, for Sidgwick this piece of ordinary morality still serves an important social purpose, insofar as a generic sexual restraint, and in particular one against adultery, favours the well-being of families and helps maintain a proper context for the rearing and education of children. So, while it is not wrong to engage in sexual activity for its own sake, it would be wrong to try and reform common sense morality in the direction of a relaxation of the standard of purity, and it would also be wrong to let one’s permissible and yet “impure” relationships and acts be publicly known. The limitations on personal autonomy and therefore on happiness that the Victorian status quo imposes on all those people adopting sexually “impure” lifestyles are, for Sidgwick, not sufficient to outweigh the social benefits of a sexually restrictive morality.

  • 12 See e.g. Hooker 2000: ch. 5. I am certainly not claiming that everything said here would follow fro (...)
  • 13 E.g. pre-marital sex prior to the existence of reliable methods of birth control was morally wrong, (...)

58To the extent that this is the outcome of Sidgwick’s unprejudiced act-utilitarian estimation, there is reason to nurture doubts about the liberalizing potential of act-utilitarianism, in Sidgwick’s time as in any other. This is not to say that e.g. rule-utilitarianism would fare any better in this respect. Rule-utilitarianism favours change of a rule prohibiting a certain behaviour at a certain time when at that time such a reform can be reliably predicted to increase expected happiness.12 If Sidgwick’s calculations are correct—and that is admittedly a big “if”, but here we cannot avoid taking Sidgwick’s word for it—then a change in the rule of purity in the direction of permissibility cannot be reliably predicted to increase expected happiness on the whole, in Sidgwick’s time. Thus, neither act- nor rule-utilitarianism would favour such change, in Sidgwick’s time. What is more, if the rule of purity was really part of the best code for Victorian Britain, rule-utilitarianism would conclude that whatever behaviour people in Victorian Britain judged as “impure” was also morally wrong—unlike act-utilitarianism, which at least could recognise most such behaviour, properly concealed, as morally permissible.13 For rule-utilitarianism determines an act as wrong, roughly, if it is prohibited by the best code from the utilitarian point of view. Liberalizing prospects are thus even less certain under rule-utilitarianism.

Haut de page


Bentham, Jeremy, “Offences Against One’s Self: Paederasty”, Journal of Homosexuality, 3:4 and 4:1 (1978)

Hooker, Brad, Ideal Code, Real World (Oxford, Clarendon Press 2000)

Mill, J. S., “On Liberty”, in Utilitarianism and Other Writings (New York, Meridian Books 1974)

Parfit, Derek, On What Matters (Oxford, Oxford University Press 2011)

Ryan, Alan, “John Stuart Mill’s Art of Living”, in J. S. Mill On Liberty In Focus, eds. J. Gray & G. W. Smith (London, Routledge 1991)

Schultz, Bart, Henry Sidgwick: Eye of the Universe. An Intellectual Biography (Cambridge, Cambridge University Press 2004)

Sidgwick, Henry, The Methods of Ethics, 7th edition (Indianapolis, Hackett 1981)

Ten, C.L., “Mill’s Defence of Liberty”, in J. S. Mill On Liberty In Focus, eds. J. Gray & G. W. Smith (London, Routledge 1991)

Whewell, William, The Elements of Ethics Including Polity (New York, Harper 1847)

Williams, Bernard, “The Point of View of the Universe: Sidgwick and the Ambitions of Ethics”, in The Sense of the Past: Essays in the History of Philosophy (Princeton, Princeton University Press 2006)

Haut de page


1 The reference is to Williams 2006: 283.

2 I will use The Methods of Ethics, 7th edition, Indianapolis, Hackett, 1981.

3 He probably got the term “Purity” from William Whewell. Actually, for Whewell Purity regulates bodily appetites in general, and thus comprehends Temperance and Chastity as its subspecies (Whewell 1847: 164-5). It is one of the basic, self-evident moral ideas and principles, thus stated: “the Lower parts of our nature are to be governed by, and subservient to, the Higher” (ibid.: 193). A close textual comparison would show that Whewell, insofar as he merely expounds common sense morality, is the obvious target of Sidgwick’s criticism here.

4 Sidgwick does not define what exactly counts as the satisfaction of sexual appetite. He can be assumed to refer to interpersonal sexual activity.

5 Presumably the idea is that common sense does not forbid sexual conduct which has sexual pleasure itself as one of its ends, provided that there is a higher end involved.

6 See the following footnote.

7 This passage resonates with Sidgwick’s life. As we know from Bart Schultz’s biography, Sidgwick’s efforts were unfailingly directed at protecting his friends’ inclinations from a dangerous exposition to the public. He threw into the river the key to a box which contained John Addington Symonds’ poetry, which had explicit homoerotic themes (Schultz 2004: 437); later, he strongly advised Symonds not to publish these poems (ibid.: 446); even after his friend’s death, he arranged to omit Symonds’ name and contribution (an essay on homosexuality in ancient Greece) from physician Dr. Ellis’ Studies in Sexual Inversion, and insisted to leave out of H. Brown’s biography of his friend all compromising references to his sexuality (ibid.: 708-13). Schultz in this respect speaks of an operation of massive falsification on Sidgwick’s part, albeit apparently carried out with the consent of Symonds’ friends and family. Moreover, Sidgwick feared that public exposure on this aspect would thwart his more practicable projects, such as reform for female education at Cambridge University.

8 Here is the relevant passage from On Liberty: “But with regard to the merely contingent, or, as it may be called, constructive injury which a person causes to society, by conduct which neither violates any specific duty to the public, nor occasions perceptible hurt to any assignable individual except himself; the inconvenience is one which society can afford to bear, for the sake of the greater good of human freedom” (Mill 1974: 213). Mill’s wording invites the question whether on hedonistic grounds human freedom is always to be regarded as a greater good, even when e.g. it is the freedom of a few individuals.

9 See Ryan 1991: 165-7, and Ten 1991: 225.

10 One can trace a markedly conservative development from Jeremy Bentham to Sidgwick. As is well known, Bentham defended vigorously, if anonymously, the decriminalization of homosexuality on the basis of the principle of utility (Bentham 1978).

11 See note 6.

12 See e.g. Hooker 2000: ch. 5. I am certainly not claiming that everything said here would follow from Hooker’s version of rule-consequentialism.

13 E.g. pre-marital sex prior to the existence of reliable methods of birth control was morally wrong, in that its prohibition was part of a consequentialist moral code designed to prevent children being born and growing under extremely difficult social conditions (see Hooker 2000: 176).

Haut de page

Pour citer cet article

Référence électronique

Francesco Orsi, « Sidgwick and the Morality of Purity »Revue d’études benthamiennes [En ligne], 10 | 2012, mis en ligne le 01 février 2012, consulté le 13 juin 2024. URL : ; DOI :

Haut de page


Francesco Orsi

Postdoctoral Researcher, Tartu University, Estonia

Haut de page

Droits d’auteur

Le texte et les autres éléments (illustrations, fichiers annexes importés), sont « Tous droits réservés », sauf mention contraire.

Haut de page
Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search