Mill’s Freedom of Speech Today

Democracy is unthinkable without freedom of speech. But how far should this freedom of speech go, when people can be indirectly harmed by words and their aftermaths?

In order to answer this question, I will outline John Stuart Mill’s arguments for freedom of speech and its limits. After that, I will argue that Mill’s so-called “harm principle” is for several reasons not limiting enough. Instead, I will suggest that the bar must be set higher in accordance with society and its values, which will always be in accordance with time and place.

Although thinkers such as William L. Chenery (1955, p. 65) argue that “freedom is good because there is no better alternative” even when it comes to speech, I will argue that this is not the case and maybe even impossible.

Mill on freedom of speech

In Chapter two of, On Liberty, John Stuart Mill (2010, p.28) presents four arguments for freedom of speech. In his first argument Mill claims that “all silencing of discussion is an assumption of infallibility.” But according to Mill, we are all fallible and therefore this could endanger us to silence an utterance that might be true.  Thus,  in order to avoid this possibility, we should not silence any opinion.

 In the second argument Mill (2010, p.77) argues that a prevailing opinion is seldom the entire truth and that even an erroneous opinion could “contain a portion of the truth.” Therefore, he opines that silencing such an erroneous thought could prevent the “collision” of these opinions and the consequent revelation of the missing truth.

While the first two arguments focus on the value of truth the next two concentrate on the manner in which we believe what is true. The third argument states that an opinion would never be recognised as true, but as prejudice, if it was not “vigorously and earnestly contested.” (Mill, 2010, p.77) For Mill contesting is essential in order to understand why something is true.

In addition, if  an opinion is not disputed, “the meaning of the doctrine itself will be in danger of being lost or enfeebled, and deprived of its vital effect on the character and conduct.” (Mill, 2010, p.77) Again, Mill argues that the dispute is of paramount importance to comprehend our own beliefs  (e.g. democracy has to be debated to show why its values are good).

Although Mill advocates freedom of speech ardently, he argues that the instance where freedom of speech should be limited is for self-protection: “That the only purpose for which power can be rightfully exercised over any member of a civilized community, against his will, is to prevent harm to others.” (Mill, 2010, p.77)

The question that arises here is what does Mill mean by harm? There is a philosophical debate about Mill’s definition of harm, but he most likely meant that harm has to invade directly and in the first instance the rights of a person. Taking this definition into consideration, the limits of freedom of speech seem to be very few.

Setting the bar right

Currently, there is a resuscitated attempt to prohibit the political party NPD (National Democratic Party of Germany) in Germany, which is known for its right-wing extremism. Would Mill support such a prohibition and consequently the intervention of freedom of speech, because the party has racist opinions? Most likely not, as long as the party’s opinion does not lead listeners to hurt people directly, e.g. by forming a mob and lynching Jews and African people after the speech.

But what, if the listeners would harm people in the long run? The recent “Döner-Morde” (Kebab Killings) in Germany, were executed by people with connections to the NPD. But if we decide to ban the party because of potential malicious influence, we would arguably also have to ban some films and video-games. The video-game debate is another issue altogether, suffice to say that the demarcation of long-term malicious influence has to be defined carefully by today’s standards.

Mill might intervene at this point by arguing that freedom of speech would always promote the truth and therefore the good in the long run. Otherwise, we would not be living in a world where morality has ostensibly advanced to the present, better stage.  Nevertheless, if one had had the chance to prohibit the NSDAP in its inchoate state, where no harm had been caused directly yet, it would have been better to do so, merely to prevent the death of millions of innocent people. Simply because the initial indirect harm was the bases for the following direct harm. Even though “truth” and good prevailed after the defeat of the Nazis, “truth” and good could have prevailed without innocent dead. Kant would accuse Mill of permitting lives to be used as means for the greater good, although life, according to Kant, is a means in itself.

Further, what about psychological harm? Mill would probably not incorporate psychological harm, since, firstly, whether Jews were harmed by the party’s speech is difficult to prove and, secondly, truth will prevail eventually. Therefore, Joel Feinberg (Corlett, 2006) argues that harm in Mill’s sense is insufficient and one should add the “offence principle”,  which sets the bar of harm so high that even some cases of psychological and indirect harm could cause the interdiction of some offensive speeches.

The problem with this argument is that some people might judge something to be offensive, while others do not; e.g. the caricatures of Mohammed in Denmark were offensive to the Muslims but not to others. Secondly, other speeches, which are normally regarded to be permissible, should then be prohibited as well. Religious speech might be offensive to other religions and political speech offensive to other parties or nations.  Feinberg writes that many factors have to be considered in order to decide what is offensive.

Mill might argue that offence could be avoided, if one wanted to, and mere knowledge of existing offensive opinions are not enough to prohibit that opinion. For example, the opponents of the extremist could simply not listen to the speech and thus be not offended. This might be valid is some cases, but often, especially in times of globalization and permanent involuntary media exposer, it is difficult to avoid these opinions.

Further, should they be allowed to spread such opinions to other, manipulable people? Again, Mill would argue for the allowance, since in his opinion, truth would prevail eventually. This might be true for a world where everyone listens respectfully, carefully and openly to each other and everyone has the possibility and will to expose himself to different opinions.

But as long as that is not given – and this is clearly the case – it is arguable that restrictions should be imposed. Plus, in modern day philosophy there is a serious debate, whether there is freedom of choice and thus freedom of opinion at all existing, and if it does, to what extent. In case there is no or limited freedom of opinion, we have to be careful what opinions are permissible.

At the same time, Mill did not consider that someone’s freedom of speech could simultaneously undermine the freedom of speech of someone else. The stigmatization of other ethnic minorities by the NDP could make it utterly difficult for these minorities to express their opinion.

Another problem is that Mill thought that truth is the most important value.  Mill (2010, p.34) refutes the argument that “the truth of doctrines, but their usefulness” is more important. On the first glance, this seems strange for the godfather of Utilitarianism, but  looking closer, it is clear that Mill (2010, p.35) thinks “the truth of an opinion is part of its utility.” At least in the long run. I already mentioned above what difficulties the time involved could cause.

In addition, our society has indeed chosen other values over the sublimation of truth. For example: breach of privacy, sedition, obscenity, slander, malicious falsehood, religious and racial hatred, and other acts are considered in many societies, unlike Mill does, as harmful and should, therefore, be considered before freedom of speech. (Banks and Hanna, 2009) If it can be proved that the NPD spreads racial and religious hatred (and thus indirect harm) it should be prohibited.

Another problem with Mill’s theory is that he, like Milton, sees truth as a coherent concept. Unfortunately, I cannot expatiate this matter, but I want to emphasise that truth might indeed be not a coherent concept. Mill, as the epitome of an Utilitarian, believed that truth or right is what promotes eventually most happiness. This theory itself has many imbroglios, such as: when is “eventually”, is happiness the same for everyone, and do the means always justify the end?

Even more important: theories such as cultural relativism argue that truth is relative to society. Another theory called Emotivism would argue that there are no prepositions, but merely emotional attitudes. Although I would argue that there is truth, it is very likely to be more complex than Mill suggested and in a dualistic world almost impossible to define.

Ludwig Wittgenstein (1999) wrote that there is no truth existing outside of mathematics. And since all our premises outside of mathematics (or the noumenal world, the world of ideas, according to Kant) are based in some way or the other on the outside world (the phenomenal world), there can be no absolute knowledge and thus no truth. Ludwig Feuerbach also opined that truth is inconceivable for the human: “The truth does not exist in the thinking, not in the knowledge for itself. The truth is only the totality of the human life and being.” (Feuerbach, 2001, §58)

Thus, if there is no truth in Mill’s definition, his entire argument is extremely weakened. But how is truth defined then? If we use Michel Foucault’s (1994) “regime of truth”, which he defines as mechanisms in history that evokes discourses functioning as true in certain times and places (see: relativism), we could define truth more relativistic and to a certain extent determined by society. Consequently, truth will prevail but constantly alter over time, unless there is a stagnation in society. With the same relativity values which override freedom of speech are chosen.

The problem of prohibition

But Mill might argue that the prohibition of speeches, as Eric Barendt (2007, p.8) writes, “creates a suspicion of authority and destroys tolerance.” In addition, how can the government make a “distinction between good speech which is free, and bad speech which may be banned or restricted.” (Barendt, 2007, p.173)

Indeed, this could give the government too much freedom to abuse their power. But this is always the case and can merely be prevented, if the government is a mirror of society and therefore a conglomerate of citizens forming a benevolent Leviathan, who is uttering their opinions and acting in their will. If this is the case the “regime of truth” is in accordance with the moral Zeitgeist.

Lastly, as long as freedom of speech is merely restricted to an extent where discussion is still feasible, freedom of speech will regulate other values, the same way other values regulate freedom of speech. The spectrum of freedom of speech is so still big enough for the NPD to express their opinion, without causing harm and is even able to contest the other values by which means freedom of speech is limited.

I have argued above that the Mill’s “harm principle” is not limiting enough to freedom of speech, since it is axiomatic that his definition leaves too much space for indirect and psychological harm. Further, I have contested Mill’s arguments with the thought that other values than truth are important and that truth is most likely no coherent system, but working in accordance with society. Then, I stated, that freedom of speech will, in the ideal case, regulate itself in a reciprocal and synergetic way with other values, which are, as much as freedom of speech, defined by society and its standards.

Finally, I want to end this essay with another thought. Stanley Fish (1994, p.102) wrote that “abstract concepts like free speech do not have any “natural” content but are filled with whatever content and direction one can manage to put into them.” This would extend my argument for cultural relativistic values to freedom of speech itself. Consequently, freedom of speech would itself be relativistic and so “free speech, in short, is not an independent value but a political prize”.


Banks, D. and Hanna, M., 2009. McNae’s essential law for journalists. New York: Oxford University Press Inc.

Barendt, E., 2007. Freedom of speech. Oxford: Oxford University Press

Corlett, J. A., 2006. The Philosophy of Joel Feinberg. The Journal of Ethics, [e-journal] Vol. 10, Available through: Jstor database [Accessed 14 December 2011].

Chenery, W. L., 1955. Freedom of the Press. New York: Harcourt, Brace and Company

Feuerbach, L. A., 2001. Grundsätze der Philosophie der Zukunft. Elibron Classics

Fish, S., 1994. There is no such thing as free speech: and it’s a good thing, too. Oxford: Oxford University press

Foucault, M., 1994. The Order of Things: an archaeology of human science. London: Vintage Books

Grayling A. C., 2007. Philosophy 1: a guide through the subject. New York: Oxford University press

Kant,  E., 2008. Grundlegung zur Metaphysik der Sitten. Stuttgat: Reclam Universal-Bilbliothek

Mill, J. S., 2010. On Liberty. 1st ed. London: Penguin Books

Wittgenstein, L., 1999. Tractatus logico-philosophicus. Stuttgart: Suhrkamp Verlag

No Comments Yet

Leave a Reply

Your email address will not be published.