Mill and Indecency Mark 2

Mill, Indecency and the Liberty Principle

Jonathan Wolff

Department of Philosophy

University College London

Mill, Indecency and the Liberty Principle
In this paper I want to do two things. One concerns Mill’s attitude to public indecency. In On Liberty Mill expresses the conventional view that certain actions, if conducted in public, are an affront to good manners, and can properly be prohibited. I want to come to an understanding of Mill’s position so that it allows him to defend this part of conventional morality, but does not disrupt certain of his liberal convictions: principally the conviction that what consenting adults do in private is no-one ‘s concern but their own. The difficulty is to find an argument that Mill could have used to defend the position that some things which, though acceptable in private, can rightly be stopped if attempted in public.

The other thing I want to do is consider the impact of Mill’s view of indecency on the interpretation of the Liberty Principle. There remain difficulties here which, to my knowledge, have not been adequately explored. So I want to look at a range of interpretative alternatives.

In the first part of the paper I shall raise and explore the issue of the interpretative problems. In the second part I shall look at some ways of trying to justify Mill’s view of indecency on characteristically Millian grounds. And in the final part I shall explore the somewhat surprising consequences of the discussion of the second part for the interpretative questions raised in the first.
1. The Interpretation of the Liberty Principle
When Mill tells us that ‘one simple principle’ should regulate the limits of state action, it is tempting to take him at his word. ‘The only purpose’, he tells us, ‘for which power can be rightfully exercised over any member of a civilized community, against his will, is to prevent harm to others.’¹ To take Mill literally - - or, as we might say, naively - - is to take him as proposing this principle as setting out a necessary and sufficient condition for the justification of state intervention. The state can interfere with your freedom of action only when you cause, or threaten to cause, harm to another, and it can interfere whenever you cause, or threaten to cause, harm to another.

If we decide to read Mill this way then the next thing we must do is look at the policies Mill thought he could justify, and see if they can subsumed under this principle. Much of On Liberty concerns freedom of thought, which, for Mill should be unbounded. This, of course, immediately requires us to focus on Mill’s understanding of ‘harm’. Allowing you freedom of thought can certainly lead to my distress: it might gravely upset me to know that you believe in the false god. Some would say that such upset constitutes harm. Clearly, then, we cannot rely on an unanalysed notion of harm.

Various proposals can be made. Rees influentially argues that for Mill harm means ‘harm to interests’.² What are interests? We are accustomed, at least, to ideas of physical and financial interests. Wollheim, on the other hand, introduces the concept of what has subsequently been called ‘morality-dependent harm’, where a morality-dependent harm is a harm one suffers only as a result of holding a particular moral view. For example, if the fact of Sunday entertainments upsets me, this can only be because I hold the moral view that Sunday entertainments are immoral. On Wollheim’s interpretation harm, for Mill, means ‘non-morally dependent harm’: a harm one would suffer irrespective of the moral beliefs one holds.³

These interpretations are normally thought to be competing, although it is logically possible that they should be compatible: perhaps Rees-harm and Wollheim-harm are coextensive. Be that as it may, both interpretations are consistent with Mill’s example of freedom of thought. Mill argues that freedom of thought should never be prohibited, and it is plausible both that it cannot cause harm to anyone’s interests, and that any suffering caused to me by other’s thoughts will plausibly be a consequence of my own moral (including religious) beliefs. Yet Mill also suggests that freedom of expression can be limited if it is likely to lead to riot. Riots obviously cause or threaten harm in both technical senses under consideration and so on both interpretations freedom of expression can properly be limited, in certain cases.

But even if some technical reinterpretation of the notion of harm enables us to follow the naive interpretative strategy pursued so far, there remain problems. Mill clearly states that economic competition and competitive examinations are not to be prohibited. Yet he equally clearly states that they cause harm: losers in economic competition might lose a great deal of money; those who do badly in competitive examinations might have to suffer lower life prospects. Such harms are non-morality dependent and are harms to interests. How can we accommodate such apparent harms?

At this point we face an interpretative fork. On the one hand we can continue with the naive strategy and try to refine the concept of harm so economic competition and so on, turn out to be harmless in the relevant sense, while theft and force remain harmful, or on the other we can try a new interpretative strategy. The most promising alternative is to move to what we can call the ‘necessary condition’ interpretation, which is to say that harm is a necessary condition of justified intervention, but not a sufficient condition. Or to put this another way, the possibility of harm is enough to put an action or type of action into the public domain, but whether it should be prohibited is a further question, to be settled on other grounds.

Wollheim’s interpretation does not obviously leave us with any room to manoeuvre if we wish to persist with the naive strategy, so if we accept his understanding of harm we must, it seems, move to the necessary condition interpretation. In the case of Rees, we could, if we wished, adopt a distinction expressed by David Lloyd Thomas⁴ and distinguish ‘rights-based interests’ from ‘desire-based’ interests: harm to interests would now be understood as harm to rights-based interests. You only have just complaint if another’s action violates your rights: your rights-based interests. This is certainly supported by Mill’s statement that we can prohibit harm to those interests which ‘ought to be considered as rights’.⁵

Obviously whether we choose to continue with the naive strategy, or move to the necessary condition strategy, we need an account of why exactly economic competition is to be permitted. That is, on the naive interpretation we need to come to an understanding of the basis of the distinction between rights-based and desire-based interests, while on the necessary condition analysis we need to come to an understanding of why some matters, while in the public domain, escape prohibition. Clearly these projects take us deeper into the heart of Mill’s position, and we will look at this shortly. Unfortunately we are not yet finished with the interpretative complications.

Both interpretative strategies appear to be compromised by one notorious passage; the main subject of this paper:
... there are many acts which, being directly injurious only to the agents themselves, ought not to be legally interdicted, but which, if done publicly, are a violation of good manners, and coming thus within the category of offences against others, may rightly be prohibited. Of this kind are offences against decency; on which it is unnecessary to dwell, the rather as they are only connected indirectly with our subject, the objection to publicity being equally strong in the case of many actions not in themselves condemnable, nor supposed to be.⁶
The expression here is coy and confusing, and Mill appears to be in something of a hurry to move on to the next topic. The passage begins by referring to acts which are direct injure only the individuals performing them, yet their public performance - - unlike their private performance - - may properly be prohibited. It is not easy to think what Mill has in mind here: perhaps masturbation was thought to fall into such a category. Another possible example might be certain forms of self-mutilation. But then Mill remarks that the objection to publicity is equally as strong in cases of indecent acts which do no harm to those engaging in them. Perhaps the simplest example would be sexual intercourse between a legally married husband and wife. No moral or religious code would normally forbid such a thing if conducted in private, but equally few such codes would permit it in public. This asymmetry between public and private performance - - whether or not it harms the participants - - I shall call the ‘indecency policy’.

The indecency policy will be considered reasonable and sensible by many people. It is very similar to the view put forward by H.L.A. Hart in Law, Liberty, and Morality.⁷ The question to be considered here is not whether Mill and Hart are correct, but whether this policy can be defended on the basis of the central doctrines of Mill's On Liberty. Clearly Mill's own argument here seems remarkably weak.⁸ His only explicit defence is that public indecency is bad manners, and bad mannered behaviour can be subject to legal regulation. But at the least it is not at all clear that this argument sits happily with the liberalism of On Liberty. The point is that the upset, shock or disgust caused by exposure to indecent behaviour is not - - or not necessarily - - a harm to the exposed person’s physical or financial interests, nor is it commonly a non-morality dependent harm: shock may simply be a result of an affront to one’s moral views. Consequently on Mill’s view harm no longer appears to be a necessary condition of justified intervention. If so neither the naive, nor the necessary condition interpretation can be defended.

Now it might be said that this argument is too quick. Suppose exposure to public indecency does cause harm. Then it could be prohibited by the Liberty Principle. All we need to do is to extend the concept of harm. However, the difficulty here is that if we say public indecency is a harm then how can we resist the thought that upset caused by private actions can also be a harm? And if we say this then, of course, we may find that illiberal interventions into people's private behaviour can also be justified on these grounds.

Perhaps it will be suggested that we can avoid the problem by assuming that there is a different sort of effect, and only a genuine sense of harm, in the case of public actions. But as Jonathan Riley remarks, it is hard to see why the dislike of a public action is in kind different from the dislike of a private action. How can the source of the dislike make it a harm in one case and not in the other?⁹ To put this another way, if the sight of certain acts causes grave offence, then the knowledge that others are engaging in similar acts in the privacy of their own homes may cause similar offence. Here it might be replied that it is rarely, if ever, the case that the same type of act will cause the same level of distress whether performed in public or private. But be this as it may, it is hard to see how any such difference in degree can give rise to a principled distinction of type that underpins the indecency policy.¹⁰

Not surprisingly, then, this passage appears to cause real obstacles to our understanding of the Liberty Principle, and reactions to the quoted passage by Mill's interpreters have been mixed. At one extreme is the supposition that Mill was, by his own lights, making a mistake when he wrote it down. On this view the passage should simply be deleted. Thus David A. Conway writes:

The passage [under consideration] is, however patently inconsistent with Mill's over-all position in On Liberty ... Henceforth, I shall not consider [this passage] as being part of 'Mill's position'.¹¹

Rees takes the somewhat different view that the passage is authentically part of Mill's view, and it is understandable that it should be. However, like Conway, Rees suggests that the indecency policy is not consistent with the remainder of On Liberty.¹² John Skorupski comments that such inconsistency is only to be expected, as the passage is introduced at a point where Mill is spelling out certain ‘obvious limitations’ to the Liberty Principle, and so Mill himself was aware of the inconsistency.¹³

Now it is not certain that Skorupski is right here. Mill uses the phrase ‘obvious limitations’ in the paragraph preceding the one in which the indecency policy is contained, and it just is not obvious whether the scope of the term ‘limitations’ is intended to cross the paragraph. Thus it is simply unclear whether Mill accepted that the indecency policy conflicts with the Liberty Principle or he believed that it could be subsumed under it.¹⁴

Yet how one views this must be fundamental for one’s understanding of the Liberty Principle. On the ‘compatibilist’ approach, one must struggle to find a reading of the Principle that reconciles the two, while on Skorupski’s ‘conflict’ approach one must avoid doing so, while at the same time find a deeper explanation of why Mill commits himself to conflicting principles.

In fact we can distinguish four general lines of interpretation:

i) The naive interpretation: Mill intended the Liberty Principle to justify all the social policies he lists, and only those social policies which naturally form part of a liberal position. This requires us to generate a conception of harm such that ‘harm to others’ is a necessary and sufficient condition for justified intervention, and justifies exactly those interventions Mill condones. In effect this means following and extending Lloyd Thomas’s modified version of Rees: i.e. the ‘harm to rights-based interests’ interpretation.
ii) The necessary condition interpretation: Mill intended the Liberty Principle as a doctrine to determine whether an action is of public concern. Its being of public concern is a question of whether it harms others. Only in cases of public concern can we ever be justified in intervening, but not all cases of public concern generate justified interventions. Thus ‘harm to others’ is a necessary condition of intervention, yet other considerations are required to generate a sufficient condition for justified intervention. On this interpretation we need a conception of harm such that in every case of justified intervention someone is harmed, and a further account of when such harm justifies intervention. Once more we might follow Lloyd Thomas, arguing this time that harm to rights-based interests is a necessary, but not sufficient, condition for justified intervention, or we could follow Wollheim, saying the same thing about non-morality dependent harm.
iii) The unknown inconsistency interpretation: Mill intended either (i) or (ii) above, but did not succeed. There is no understanding of harm that both can reasonably be attributed to Mill, and does the job required. Hence Mill’s project is a failure, although Mill did not recognise this. His position must be amended. This is Conway’s view.
iv) The known inconsistency interpretation: Mill knowingly exaggerated the scope of the Liberty principle in setting it out. He thought that harm to others was generally a necessary and/or sufficient condition for justified intervention but he recognised exceptions. Thus one does not need to generate a particularly sensitive understanding of harm, but a deeper understanding of Mill’s overall position which allows one to see that the Liberty Principle is true only for the most part. That is, we need to understand that Mill was prepared to modify the scope of the Principle, and for this reason we need not substantially amend its content. On this basis we can return to Rees’s or Wollheim’s interpretation, or some other, as a ‘necessary condition for the most part’. Broadly this is Skorupski’s position.
I don’t pretend to be able to settle the question of which is the right approach to take, but we can make some headway. In any case considering the indecency policy will be an important test for any given interpretation. In the next section, then I shall explore the question of what type of defence of the indecency policy can be given, and then, in the final section, return to the question of how this impacts on one’s understanding of the Liberty Principle.

2. Defending the Indecency Policy

In essence, the question before us is that of why offence taken at public acts is a ground for regulation while offence taken at private acts is not. C.L. Ten, on Mill’s behalf, appeals to the concept of ‘offensive nuisances’, which are ‘offensive acts ... committed in a place frequented by those who are offended, and not easily avoided by them.’¹⁵ That an act is an offensive nuisance is enough, thinks Ten, to place it in the public realm, subject to regulation, but whether it should be prohibited is a further question requiring the balancing of various considerations.¹⁶

While reasonable enough, it is not obvious how this account is to be defended on Millian grounds. Ten appears to defend the indecency policy on grounds of fairness. If people find certain forms of public behaviour offensive, and it is hard to avoid them as they take place in a ‘common environment’, then such people will either have to suffer offence or sacrifice their legitimate activities. But it is unfair that people should have to make this sacrifice. So, for example, if there is only one beach, and nude sunbathing offends me, then nude sunbathing on the beach is unfair to me, because I can only avoid being offended by avoiding the beach. But it is unfair that I should have to sacrifice my beach-going, so nude sun-bathing should be subject to public regulation.

An apparent problem with this argument is that no one is stopping me from going to the beach, it is simply that I don't go because I don't want to be offended by nudity. And why is this any different from not going because, although I like swimming and sun-bathing, I find the vulgar accents of the other beach users offensive? It seems that there will be a difference - - and unfair treatment - - only if I have a right not to be offended by nudity (and no right against offensive vulgar accents). But if we admit this, then we need an explanation of why I have one right but not the other. Further we are yet again faced with the problem of why I have no right to be protected against offence caused by other people's private behaviour.

In reply, however, it is open to Ten to allow that all sorts of things - - including vulgar accents - - can be offensive nuisances, and thus part of the public domain. But whether they should then be prohibited is a further matter. Thus Ten could rightly argue that there are independent good reasons for not regulating accents, and these do not apply in the nude sunbathing case. Appeal could be made to the value of freedom of expression, or the difficulty and cost of framing and enforcing appropriate legislation, given that any nuisance actually suffered is likely to be rather trivial.

The real difficulty with Ten's proposal is not, then, that it is incapable of generating the right distinctions, for perhaps it can. Rather, it simply seems ad hoc to extend Mill's view in this way. Unlike, say, Feinberg, ¹⁷Mill does not explicitly appeal to an 'offence principle', and there seems little internal motivation for letting the concept of offensive nuisance do the work Ten requires of it. We should see what can be done to support the indecency policy on characteristically Millian grounds, even if Mill did not do this for himself. This is what I shall now attempt.

Let us accept that, whether or not we ought to use the concept of a right in interpreting the Liberty Principle, Mill’s overt policy is to assign people certain rights: including rights not to be exposed to indecent behaviour, together with rights to act however one wishes (with other consenting adults) in private. But what, according to Mill, defines or determines the catalogue of rights we ought to have? We cannot be sure that it should be those rights that appear in standard catalogues of natural rights, because Mill, following Bentham, believes there are no natural rights. Nor can it be those rights given by current conventions, for Mill wishes to use his doctrine to reassess convention, not merely to endorse it.

Clearly this is where utilitarianism steps in. When Mill says ‘I forgo any advantage which could be derived to my argument from the idea of abstract right, as a thing independent of utility’,¹⁸ he presents more than a hint that we are to be offered a utilitarian theory of rights. And this makes sense of much Mill says. The utilitarian advantages of free disposal of property, and of economic trade, require private property rights which, although they can do harm in particular cases, produce a sufficient surplus of good to overbalance this.¹⁹ And given the current sorry state of human nature no other system could do better. (Although, Mill writes, with the inevitable progress in the moral development of human nature, socialism will eventually work better still.²⁰)

How do we extend this reasoning to encompass the indecency policy? In trying to come to an understanding of Mill it will be necessary to give some attention to the utilitarian structure of On Liberty. But let us first concede the point made by Riley and Conway. The mere knowledge that certain acts are taking place (or even that it is possible that such acts should be taking place) can be just as upsetting as the sight of other acts of which public performance is banned on grounds of decency or offence. Knowing that my neighbours are indulging in violent sado-masochism, incest or bestiality can be far more disturbing than witnessing nudity in the supermarket.

However to come to a satisfactory defence of the indecency policy we need to look not merely at the nature of the offence, nor merely at its source, but also at the consequences of prohibiting the actions which cause offence. Suppose we successfully prohibit, say, homosexual acts between consenting adults in private. If we do this then the consequences may be extremely far reaching. We will be ruling out certain plans of life: stifling the individuality of those involved. This has two sorts of cost. First, for Mill, ‘the free development of individuality is one of the leading essentials of well-being’.²¹ Second, and more important, we will be preventing some people from carrying out ‘experiments of living’. As is well known Mill argued that we should encourage free experimentation because mankind is progressive, capable of learning from experience. Through the experiments of some individuals we may learn things of great value, for the permanent benefit of mankind. Of course we might not believe that we will learn much from any particular experiment, but then we have often been wrong about this in the past. By restricting certain patterns of private behaviour we run the risk of shutting out lifestyles that could be of immense long-term value.

However, in general, restrictions on public behaviour will not have such costs. Most of those acts which are offensive if performed in public can still be performed in private, and thus can still constitute part of a person's active plan of life. So it will often be the case that legislation against public indecency will not constrain anyone's plan of life, neither countering their individuality nor ruling out their experiments in living.

Mill's indecency policy, then, can be justified in the following utilitarian terms. Offence or upset caused by private actions, however grave the offence, is never a sufficient reason for prohibiting those actions, for such prohibition runs the risk of ruling out something that will be of permanent benefit to mankind. But offence or upset caused by public actions is sometimes (but, of course, not always) a sufficient reason for prohibiting those actions. If the offence is grave enough, and actions of that type can still be performed in private even if banned in public, then prohibition may be acceptable. To use a phrase common in the condemnation of indecent or other ill-mannered behaviour: it is ‘so unnecessary’. Mill, at least, can tell us what it is unnecessary for.

It is important to notice that I am not relying on the following reasoning: prohibiting private performance of upsetting acts would sometimes be desirable, but the costs and difficulties of enforcing such a policy would outweigh any good that such a policy would do. That is quite a different argument, relying on the costs of enforcement. Mill, of course, would be quite entitled to make that argument too, but I claim that the distinctive Millian argument refers not to the costs of enforcement but to the consequences of any successful prohibition. The objection is that such a prohibition rules out certain plans of life.

Is it plausible to attribute this argument to Mill? I do not claim that it lies buried in the text. Yet the argument combines characteristic positions from Mill to defend his indecency policy. It also relies on a form of argument that can be applied in other problem cases. So I would claim that it is both more authentic and more useful, than Ten's appeal to offensive nuisance'.

To make my interpretation clearer, and before considering objections, I want to bring out more explicitly the type of utilitarianism I wish to attribute to Mill. Suppose we wish to carry out Mill's project of setting out ‘the nature and limits of the power which can be legitimately exercised by society over the individual.’²² Suppose too that we wish to adopt a utilitarian framework. For various reasons, and in particular, the utilitarian advantages of stable expectations and the rule of law, we may wish to develop a doctrine of rights as an indirect utilitarian strategy. But the set of rights we should adopt is partly an experimental question: we want that set of rights most likely to do the most long term good, given human nature.

At this point all sorts of potential pleasures and pain should be considered: mere offence, nosy preferences, sadistic pleasures, the damage caused by competition, as well as standard sources of pleasure and pain. All of these are utility losses and gains, of different kinds and degrees. Proposals for theories of rights might give some of these sources of pleasure and pain no weight at all, and something protected as a right in one account need not be so protected in another. Eventually we will decide upon the utilitarian best set of rights. And once we have carried out the utilitarian calculations we are likely to have decided that some sources of utility losses should not count as an infringements of rights. For example, as we have seen, Mill is clear that the damage one suffers through losses in normal economic competition is not to be considered a violation of rights. However, a few clear cases aside, we cannot in advance say what rights we have, until the utilitarian calculation has been made.

Once we understand this structure, we can agree with Riley and Conway that the sight of something is not necessarily a worse harm than the bare knowledge that it is taking place. Yet we can also agree with Mill and Hart that while knowing that something is taking place never infringes your rights, seeing it sometimes will. Why? Given the importance of allowing experiments in living, the utilitarian consequences of that set of rights is better than any competing version.

The utilitarian case for the indecency policy is that prohibiting offensive private actions stifles individuality and rules out experiments in living, and therefore has great actual or potential costs, whereas prohibiting offensive public actions does not have such costs as these actions can still be conducted in private. Conway, however, writing against a similar point made by Hart, argues that this is not so. 'One can be nude in private, but the point of so doing (a feeling of freedom in the supermarket, or whatever) may be lost.'²³

This particular example seems too under-described to know what to say about it,²⁴ but broadly, it seems, we can distinguish at least three different kinds of cases where the rationale of the act requires that it be done in public. First, an action may simply be done as a form of exhibitionism, to shock unsuspecting bystanders. Second, a shocking action may be performed for public spirited reasons: for educative purposes or to break irrational taboos. Diogenes, it is said, would masturbate in the market place in order to show how easily sexual desire could be satisfied. Whatever we think of the merits of this lesson, this seems a different case to mere exhibitionism. Third, some actions may be done by means of protest: spitting on the flag, for example. This again seems essentially public. Spitting on the flag in the privacy of one's own home does not make the same point.

The first thing to say about these cases is that strictly speaking they are irrelevant. For the indecency policy only concerns those offensive actions which can either be done in public or private, and so if certain actions can only be performed in public then they are not covered by the policy. However, the utilitarian doctrine presented here to support that policy should also have things to say about other related cases. So what does it say?

In which of these cases does one have the right to be shielded from such offensive spectacles? Again the answer turns on the benefits of individuality and experimentation: what harm to these would be done by prohibition? It is plausible that in the case of exhibitionism very little harm will be done. But education and taboo breaking seem to require more tolerant treatment. Perhaps we will learn a lesson for the future enrichment of mankind. There will always be the question of whether an actual demonstration is a necessary component of the educative purpose, but sometimes this will be the only, or at least most effective means by which a message can be conveyed. No doubt there will be hard cases. Finally, flag spitting as protest seems to be more closely related to issues of freedom of expression, and this surely would outweigh any consideration based only on decency. However, nothing depends on the particular answers I have given. The utilitarian framework provides a way of coming to make decisions, although there is room for debate in each case about what those decisions should be.

Rees suggests that it is hard to see how Mill can consistently allow the prohibition of obscenity while opposing the prohibition of the eating of pork in predominately Muslim countries.²⁵ But on the basis of what I have said there is no difficulty here. Mill does not make clear whether he is talking about public pork eating or merely private pork eating, but his view would allow the prohibition of public displays of pork eating while allowing it in private. One can imagine the eating of pork handled in a rather similar matter to that of off-track gambling or pornographic peep-shows in Great Britain. Premises where these activities take place are clearly marked, but required by law to arrange their doorways and windows so that no one should be able to peer in from the street. Consequently all who witness the scenes within do so knowingly and voluntarily. 'Offensive sequences' in the theatre and cinema can be handled in a similar way.

This does raise the further question of how such places should be allowed to present and advertise themselves. Ideally they should be discreetly marked, although this is not always the case, and private activities can be publicly advertised in ways which cross the public/private boundary. Kurt Baier illustrates this with the fanciful example of someone putting a vivid neon sign on their house proclaiming:

Cannibalism, Bestiality, Incest. Tickets $5.00. Meals $25. Close relatives half price.²⁶

But as long as pork butchers and restaurateurs exercise more restraint, then the consumption of pork in Muslim countries remains perfectly consistent with the indecency policy.

Finally, perhaps the most pressing objection to the utilitarian underpinning of the indecency policy is that it makes such a policy radically contingent. The case for never prohibiting private offensive actions is based on the significance of individuality, and, more deeply, on the importance of experiments in living for the future well-being of a progressive mankind. But suppose we stop progressing, or at least stop learning from the experience of other people. In that case the indecency policy - and indeed the Liberty Principle more generally²⁷ - is without complete defence. For Mill the possibility of progress is built into human nature. But the position outlined here will not be satisfactory for those who wish to defend the indecency policy but do not share Mill's optimism. Or, of course, for those who reject utilitarian forms of justification.
3. The Liberty Principle Revisited
Where, after all this, are we left with respect to the interpretation of the Liberty Principle? We found that there were four broad interpretative strategies:

i) Naive: harm is a necessary and sufficient condition for intervention.

ii) Necessary condition: harm is a necessary, but not sufficient, condition for intervention.

iii) Unknown inconsistency: the Liberty Principle is inconsistent with the indecency policy but Mill was unaware of this.

iv) Known inconsistency: the Liberty Principle is inconsistent with the indecency policy, but Mill is aware of this.
It is, of course, possible that (iii) is the correct interpretation, but given the availability of (iv) it seems rather uncharitable to attribute (iii) to Mill, so I will not consider it any further. I would hazard the view that (ii) is the orthodox interpretation. Strangely enough, however, it seems to me that it is the only one we can conclusively rule out, as I shall now argue.

The materials in the last section allow us to pursue the naive strategy in the following way. Essentially the strategy requires us to devise a finely contoured definition of harm such that harm is a necessary and sufficient condition for justified intervention. And, indeed, we can supply such a notion. Harm is violation of rights, those rights which have been allocated by the indirect utilitarian strategy outlined in the previous section. On this view we cannot understand the Liberty Principle, and what it permits and prohibits, until the utilitarian calculation is complete. And once the calculation is made, then, on Mill’s assumptions, we find ourselves with rights to be protected from indecent displays, but no rights to be protected from economic competition. Thus we could, if we wish, follow the naive strategy: harm is violation of utilitarian rights. The cost is that there is no longer any quick or intuitive understanding of the critical force of the Liberty Principle.

The ‘known inconsistency’ strategy is equally easy to follow on the basis of the arguments given. Here the idea would be to take a more basic final statement of the Liberty principle, perhaps following Rees or Wollheim, and suggest that Mill thought that it operates as no more than a very good rule of thumb. That is, in almost all cases, harm to interests, or whatever, is a necessary condition of justified intervention. However the details of the application of the principle have to be worked out on utilitarian grounds - hence no prohibition of competition. Furthermore in rare cases we have utilitarian rights (or, indeed, duties) that conflict with the Liberty Principle, so understood, and such rights or duties take priority. The right to be protected from indecency is the most prominent example. Thus on this view the Liberty Principle is not itself to be read in light of utilitarianism, but its conditions of application are determined by utilitarianism. It is a rather rough principle containing a very large intersection with the correct policy, but the principle utility overrides the Liberty Principle where the two conflict. Here we retain the advantages of allowing an intuitive understanding of the Liberty Principle, but must always recognise that it can never be the final authority.

What about the more common ‘necessary condition’ interpretation? Either we take this literally and say that by ‘necessary’ we mean ‘necessary’, or take it in the more lax form as ‘necessary for the most part’. If we do the latter, then the interpretation reduces to the last one, so we must think of necessary as meaning ‘strictly necessary’ to retain an autonomous interpretation. Therefore, on this view, if indecency is to be prohibited then it must involve harm to others, or threat of harm. Therefore, just as was the case with the naive interpretation, we need to fine-tune the concept of harm to extend it in this way. And we know how to do this: we appeal to a utilitarian account of rights. Can we then say that violation of rights is a necessary, but not sufficient, condition of justified intervention, as the necessary condition interpretation requires?

It might be best to approach this question by way of an example. The paradigm of the necessary condition analysis in action is the case of economic competition: we realise that economic competition causes harm, which is enough to place it ion the public domain, but when we come to see its utilitarian advantages we decide not to prohibit it. However economic competition causes harm in the ordinary sense of harm, and from that it does not follow that it causes harm in the technical ‘violation of utilitarian rights’ sense. And recall that, because of the problem of indecency, the necessary condition interpretation must appeal to the technical sense of harm, not the ordinary sense. So does economic competition cause a harm in the technical sense? If it does, then we must prohibit it. If it does not, then there is nothing to enter in the public domain. But in neither case is there need or opportunity for a second utilitarian calculation. For what could be achieved by a second utilitarian calculation that had not already been achieved by the first? What further input could it draw on? As far as I can see, there is none.

To put the argument a different way, the necessary condition interpretation says, first we find an analysis of the Liberty Principle, and then we apply it in practice by means of utilitarian balancing. But the utilitarian balancing has already taken place in the definition of the Liberty Principle in the first place. What further balancing could be required or possible to turn the necessary condition into a sufficient one? In other words, once we fine-tune the concept of harm sufficiently well to make it a necessary condition of justified intervention, it turns out to be fine-tuned enough to provide a sufficient condition too. Or it had better. For there are simply no resources available for us to carry out a second utilitarian balancing.

So, to my surprise, we find that there are only two viable interpretative strategies. The necessary condition analysis must reduce either to the naive (or not so naive) interpretation or to the known inconsistency interpretation. This seems a significant result.

How should we finally settle the interpretative issue? I will have to leave that to others. But we should note that while accommodating the indecency policy is a test for any interpretation, the Millian defence of it that I have given is consistent with both remaining approaches, and so does not force a choice.

1 John Stuart Mill, On Liberty, Essays on Politics and Society, ed. J M. Robson, Toronto, 1977, Collected Works of John Stuart Mill, xviii, 223.

2 J.C. Rees, A Re-Reading of Mill On Liberty, in On Liberty in Focus, ed. John Gray and G.W. Smith, London, 1991, p. 174.

3 Richard Wollheim, 'John Stuart Mill and the Limits of State Action', Social Research 40 (1973), 1-30. The term 'morality dependent harm' is from Ted Honderich, ' "On Liberty" and Morality Dependent Harms', Political Studies 30 (1982), 504-14.

4 David Lloyd Thomas, 'Rights, Consequences, and Mill on Liberty', in Of Liberty ed. A. Phillips Griffiths, Cambridge, 1983, p. 168.

5 Mill, On Liberty, CW xviii. 276.

6 J.S. Mill, On Liberty, CW xviii. 295-6.

7 H.L.A. Hart, Law Liberty and Morality , London, 1963, pp. 38-48. It is not quite clear whether Hart takes himself to be articulating Mill's view, or putting forward a similar view for himself, perhaps on different grounds.

8 Indeed the argument given by Mill trades on an equivocation in the use of the term 'offence'. The sense in which bad-mannered behaviour is 'offensive' is not the same as the sense of 'offence against others' sufficient to bring an action into the sphere of public regulation. The latter sense is illustrated by the legalistic phrase 'offences against the person'.

9 Jonathan Riley, 'One Very Simple Principle', Utilitas 3 (1991), 23.

10 One way of defending the asymmetry is that public actions 'impinge' on us in ways that purely private actions do not. So if the objection to indecency is largely aesthetic then 'impingement' will produce an appropriate distinction. I owe this thought to Jerry Valberg. However as this does not seem to be a view one could attribute to Mill, I shall not consider it here any further.

11 David A. Conway, 'Law, Liberty and Indecency', Philosophy 49, (1974), 137.

12 J.C. Rees, 'A Re-Reading of Mill On Liberty', pp. 185-86.

13 John Skorupski, John Stuart Mill ,London, 1989, p. 342.

14 I’m grateful to Roger Crisp for illuminating correspondence on this point.

15 C.L. Ten, Mill On Liberty , Oxford 1980, p. 105.

16 Ten, Mill On Liberty, p. 105-6. For a sophisticated attempt to balance such considerations, although deliberately in the absence of any specific theoretical framework, see Joel Feinberg, Offense to Others (, Oxford, 1985.

17 Feinberg, Offense to Others.

18 Mill, On Liberty, CW xviii. 224.

19 And, of course, in Chapter 2 of On Liberty Mill makes out in great detail the utilitarian advantages of a right to freedom of thought.

20 J.S. Mill, Chapters on Socialism i, Essays on Economics and Society 1850-1879, ed. J.M. Robson, Toronto, 1967, Collected Works of John Stuart Mill, v.

21 Mill, On Liberty, CW xviii, 261. It is interesting to note that nothing like the pleasure of individuality appears on Bentham's list of several dozen categories and sub-categories of pleasures in An Introduction to the Principles of Morals and Legislation , ed J.H. Burns and H.L.A. Hart, London, 1970 (The Collected Works of Jeremy Bentham), p. 42-5.

22 Mill, On Liberty, CW xviii, 217.

23 Conway, 'Law, Liberty and Indecency', p. 140.

24 Although it does raise a more general worry: how effective at communicating its message will a private experiment in living be? However, literature and direct exposure to volunteers may well be sufficient.

25 Rees, 'A Re-Reading of Mill On Liberty', p. 186, Mill, On Liberty, CW xviii, 284-5.

26 Kurt Baier, 'The Liberal Approach to Pornography', University of Pittsburgh Law Review 40 (1979), 620.

27 The contingency of the Liberty principle on the possibility of progress is clearly implied in Mill's notorious discussion of children and barbarians. Mill writes 'Liberty, as a principle, has no application to any state of things anterior to the time when mankind have become capable of being improved by free and equal discussion.' On Liberty, CW xviii, 224.