Why J.S. Mill’s ‘The Harm Principle’ is Useless

Introduced in ‘On Liberty’, and conceived by J.S. Mill, the harm principle states that, the only reason for which prohibition can be exercised rightfully over a member of a civilised community is to prevent harm to others. With an impressive liberal pedigree from such supporting academics as Joel Feinberg, it may be hard to understand why the harm principle in the eyes of so many falls so short. Indeed, this essay will suggest that while the harm principle is of use in some respects, such as a brief and effective test of legislature, or a moralised view of harm, it is too weak to stand up to the test of sustained philosophical and legal rigour, failing due to its vagueness of the term ‘harm’, over inclusivity as a result, and its overall illiberal nature, therefore making it useless both legally and in a wider sense.

As Ripstein argues, Mill, and subsequently Feinberg (despite his best efforts) fail to describe the exact status ‘harm’ is given as part of the maxim. He uses the example of a harmless trespass, highlighting how the harm principle cannot provide a cogent account of either the harm committed, or the grounds for criminalisation (Ripstein, 2005: 218), and is forced to extend the harm principle to his ‘sovereignty principle’ to account for its shortcomings (Ripstein, 2005: 219). Vernon furthers this, pointing out Mill’s failure to define ‘harm’ at all throughout ‘On Liberty’ (Vernon, 1996: 625). So, it seems neither of the harm principle’s arguable primary propagators are clear themselves on what exactly ‘harm’ constitutes. Indeed, despite Feinberg’s best efforts, the ambiguity of his definition of ‘harm’, and the wrongs it includes are problematic (Feinberg, 1984: 105), given the amalgamation of Kantian rights, which in themselves would provide a plausible reason for criminalisation, harm or not, and purely interest-based notions of right and wrong, which as previously described “cannot act as a significant limit on the scope of criminalisation” despite often being necessary. (Stewart, 2009: 23) This confusion removes much of the harm principle’s critical force. A result of such ambiguity is understandable speculation. As Epstein points out (Epstein, 1995: 401) the concept of harm has, and continues to greatly expand due to developments in a range of sciences, which expanded causal links, leading to a situation where “a little causal creativity can go a long way towards eviscerating the harm principle.” (Dripps, 1998: 8). Thus, this also leads to a situation in which the principle can be greatly overextended, rendering it useless. How do we, for example deal with harm dealt in a self-defence situation, or harm resulting from competitive business? (Stewart, 2009: 19) Stretching this further, realistically, any action can in some way be attributed to a harm further down the line, which in the eyes of the harm principle theoretically makes it worthy of criminalisation, with any potential benefits of such ‘harms’ being unaccounted for, such as the by-product of competitive business spirit, being economic growth, despite the harm it may cause.

This also leads to an issue surrounding the harm principle, of liberty itself. As Dripps points out (Dripps, 1998: 9) “The harm principle’s application is contingent on the facts”, and given the difficulty of discrediting a logically hypothesised causal link, “the facts are likely to be found adverse to liberty.” Holtug correctly points out the principle’s deficiencies, which offer an “at best inadequate” defence of personal liberties. If some people were to find IVF offensive, for example, they could appeal to both “psychic harm” and “communal harm” and indeed, to those who experience the perceived undermining of their values, such distress will seem harmful. Therefore, IVF should logically be prohibited. It should also hold true therefore that smoking in the UK should be prohibited because of the adverse effects it has on the NHS in terms of costs, and therefore on the UK taxpayer, advancing a more severe kind of state paternalism disliked by Mill, despite his strange commitment to utilitarianism (Gostin, 2008: 215) (‘I regard utility as the ultimate appeal on all ethical questions”). (Mill, 1962: 136).  Thus, the harm principle “underwrites” many of the prohibitions it is supposed to include. Such restrictions are strange given Mill’s disdain for social convention’s stifling characteristics, (“eccentricity of conduct, are shunned equally with crimes”) (Mill, 1962: 110) which under the “psychic” principle for the first time, have a legal basis for preventing society’s eccentricities, which Mill so valued. Such an assault on liberty reveals the true character of the harm principle, which leads us to our second issue. In Stewart’s hypothetical harmless rape scenario (Stewart, 2009: 30), in order to adhere to the harm principle as we have in such previous examples such as IVF, proof of harm would be necessary, which in this case it is not, meaning such an invasion of personal autonomy is not seen as a harm, and is not vulnerable to criminalisation. Therefore, the harm principle “fails to constrain the scope of criminalisation, and fails to take individuals seriously.” (Stewart, 2009: 26).

Nevertheless, it would be wrong to entirely discount the harm principle. As Dripps points out, “the harm principle would prevent wise legislation very rarely, but quite regularly would prevent unwise legislation.” (Dripps, 1998: 11). The overall tone of ‘On Liberty’ celebrates individual excellence and autonomy, which is prima facie correct, along with the fact it is wrong to cause harm to another individual, which is ingrained in our language and culture and “dies a very slow death even in philosophical discourse.” (Epstein, 1995: 371). Epstein points out the fact that proving harm in the case of homosexuality or pornography (beyond offense) cannot be done easily (Epstein, 2004: 41), while moving beyond the point of ease infringes the entire point of the principle, which expresses a clear desire to be clear and simple, or “to offer judgements easily, casually and with apparently unqualified confidence.” (Smith, 2006: 8). Furthermore, Holtug points out the harm principle has its advantages when we moralise the concept of harm, as we begin to view harm more as wrongdoing, therefore explaining the liberal view of not coercing a person to negatively affect others when they consent to it (Holtug, 2002: 378) while the cases of excessive causal links are weakened, considering the lack of wrongdoing to others brought about by homosexual acts or other subjectively ‘harmful’ practices.

Holtug’s remedy moves away from Mill’s original concept, though, while Epstein’s easy-going approach would soon crack under pressure if the principle was to be taken seriously as a legal bed-rock. As previously suggested, the harm principle would be detrimental to liberty, and in the situation of it being salient in jurisprudence, I believe would allow judges to quite easily criminalise standard practices, while also allowing them to “strike down laws that elected bodies and liberal principles may see wise to not condemn.” (Dripps, 1998: 11). To reinforce this point, as Stewart highlights, “No plausible version of the harm principle is going to count direct harms only”, in which case the harm principle is unable to clearly constrain upon what grounds criminalisation stands.

To conclude, the harm principle does indeed sound sensible, but in reality, is of no practical use. It’s failure to fully assign a meaning to ‘harm’, and therefore contain the scope of criminalisation, along with its potential to be overextended with creative use of causal-links and be damaging to liberty makes it impractical, and of no use to a realistic legal framework. While in its original sense, it could be seen as useful, as a simple defence of basic liberties in a debate, its use stretches no further I believe. A practical development for the harm principle would be to moralise the concept of harm, removing many of the ambiguities surrounding the term.














Dripps, D.A., 1998. The liberal critique of the harm principle. Criminal Justice Ethics17(2), pp.3-18.

Epstein, R.A., 1995. The harm principle-and how it grew. The University of Toronto Law Journal45(4), pp.369-417.

Feinberg, J., 1990. Harm to others. Oxford University Press.

Gostin, L.O. and Gostin, K.G., 2009. A broader liberty: JS Mill, paternalism and the public’s health. Public health123(3), pp.214-221.

Holtug, N., 2002. The harm principle. Ethical theory and moral practice5(4), pp.357-389.

Mill, J.S., 1966. On liberty. In A Selection of his Works (pp. 1-147). Macmillan Education UK.

Ripstein, A., 2006. Beyond the harm principle. Philosophy & Public Affairs34(3), pp.215-245.

Smith, S.D., 2006. Is the harm principle illiberal. Am. J. Juris.51, p.1.

Stewart, H., 2010. The limits of the harm principle. Criminal Law and Philosophy4(1), pp.17-35.

Vernon, R., 1996. John Stuart Mill and pornography: Beyond the harm principle. Ethics106(3), pp.621-632.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s