§4. Let us pass now to what I have called the duties of the Affections, the rules that prescribe either love itself in some degree, or the services that naturally spring from it in those relations where it is expected and desired. Here, in the first place, the question how far we are bound to render these services when we do not feel the affection is answered differently in many cases by different persons, and no determination of the limit seems self-evident. And similarly if we ask whether affection itself is a duty; for on the one hand it is at least only partially within the control of the will, and in so far as it can be produced by voluntary effort, there is thought to be something unsatisfactory and unattractive in the result; and on the other hand, in certain relations it seems to be commonly regarded as a duty. On those points the doctrine of Common Sense is rather a rough compromise between conflicting lines of thought than capable of being deduced from a clear and universally accepted principle. And if we confine ourselves to the special relations where Common Sense admits no doubt as to the broad moral obligation of at least rendering such services as affection naturally prompts, still the recognised rules of external duty in these relations are, in the first place, wanting in definiteness and precision: and secondly, they do not, when rigorously examined, appear to be, or to be referable to, independent intuitions so far as the particularity of the duties is concerned. Let us take, for example, the duty of parents to children. We have no doubt about this duty as a part of the present order of society, by which the due growth and training of the rising generation is distributed among the adults. But when we reflect on this arrangement itself, we cannot see intuitively that it is the best possible. 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; we have to refer to psychological and sociological generalisations, obtained by empirical study of human nature in actual societies. If, however, we consider the duty of parents by itself, out of connexion with this social order, it is certainly not self-evident that we owe more to our own children than to others whose happiness equally depends on our exertions. To get the question clear, let us suppose that I am thrown with my family upon a desert island, where I find an abandoned orphan. Is it evident that I am less bound to provide this child, as far as lies in my power, with the means of subsistence, than I am to provide for my own children? According to some, my special duty to the latter would arise from the fact that I have brought them into being: but, if so, it would seem that on this principle I have a right to diminish their happiness, provided I do not turn it into a negative quantity; since, as without me they would not have existed at all, they can, as my children, have no claim upon me for more than an existence on the whole above zero in respect of happiness. We might even deduce a parental right (so far as this special claim is concerned) to extinguish children painlessly at any point of their existence, if only their life up to that point has been on the whole worth having; for how can persons who would have had no life at all but for me fairly complain that they are not allowed more than a certain quantity?  I do not mean to assert that these doctrines are even implicitly held by Common Sense: but merely to show that here, as elsewhere, the pursuit of an irrefragable intuition may lead us unaware into a nest of paradoxes.
It seems, then, that we cannot, after all, say that the special duty of parents to children, considered by itself, possesses clear self-evidence: and it was easy to show (cf. chap. iv.) that as recognised by Common Sense its limits are indeterminate.
The rule prescribing the duty of children to parents need not detain us; for to Common Sense it certainly seems doubtful whether this is not merely a particular case of gratitude; and we certainly have no clear intuition of what is due to parents who do not deserve gratitude. Again, the moral relation of husband and wife seems to depend chiefly upon contract and definite understanding. It is, no doubt, usually thought that Morality, as well as law prescribes certain conditions for all connubial contracts: and in our own age and country it is held that they should be (1) monogamic and (2) permanent. But it seems clear that neither of these opinions would be maintained to be a primary intuition. Whether these or any other legal regulations of the union of the sexes can be deduced from some intuitive principle of Purity, we will presently consider: but as for such conjugal duties as are not prescribed by Law, probably no one at the present day would maintain that there is any such general agreement as to what these are, as would support the theory that they may be known a priori.
If, then, in these domestic relations---where the duties of affection are commonly recognised as so imperative and important---we can find no really independent and self-evident principles for determining them, I need not perhaps spend time in showing that the same is the case in respect of the less intimate ties (of kindred, neighbourhood, etc.) that bind us to other human beings. Indeed, this was made sufficiently manifest in our previous discussion of those other duties.
No doubt there are certain obligations towards human beings generally which are, speaking broadly, unquestionable: as, for example, the negative duty of abstaining from causing pain to others against their will, except by way of deserved punishment (whether this is to be placed under the head of Justice or Benevolence); and of making reparation for any pain which we may have caused. Still, when we consider the extent of these duties and try to define their limits,---when we ask how far we may legitimately cause pain to other men (or other sentient beings) in order to obtain happiness for ourselves or third persons, or even to confer a greater good on the sufferer himself, if the pain be inflicted against his will,---we do not seem able to obtain any clear and generally accepted principle for deciding this point, unless the Utilitarian formula be admitted as such. Again, as regards Reparation, there is, as we have seen, a fundamental doubt how far this is due for harm that has been involuntarily caused.
Similarly, all admit that we have a general duty of rendering services to our fellow-men and especially to those who are in special need, and that we are bound to make sacrifices for them, when the benefit that we thereby confer very decidedly outweighs the loss to ourselves but when we ask how far we are bound to give up our own happiness in order to promote that of our fellows, while it can hardly be said that Common Sense distinctly accepts the Utilitarian principle, it yet does not definitely affirm any other.
And even the common principle of Gratitude, though its stringency is immediately and universally felt, seems yet essentially indeterminate: owing to the unsolved question whether the requital of a benefit ought to be proportionate to what it cost the benefactor, or to what it is worth to the recipient.