Bryan Caplan  

The Awful Mill

Notes on the Phillips Curve... Neuroscience and Ethics...
I've never been a fan of John Stuart Mill.  Yes, he had a massive IQ and a dreadful Tiger Dad.  But his thinking is shockingly muddled. 

One especially cringeworthy example: In the span of two pages in On Liberty, Mill names one "ultimate" principle and one "absolute" principle.  His Ultimate Principle:
It is proper to state that I forego any advantage which could be derived to my argument from the idea of abstract right, as a thing independent of utility. I regard utility as the ultimate appeal on all ethical questions...
His Absolute Principle:
The object of this Essay is to assert one very simple principle, as entitled to govern absolutely the dealings of society with the individual in the way of compulsion and control, whether the means used be physical force in the form of legal penalties, or the moral coercion of public opinion.
You might think that Mill would argue that his Ultimate Principle implies his Absolute Principle - or at least that that the two principles never conflict.  That would be silly and dogmatic, but consistent.  Instead, Mill temporarily forgets his Ultimate Principle in favor of his Absolute Principle:
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. His own good, either physical or moral, is not a sufficient warrant. He cannot rightfully be compelled to do or forbear because it will be better for him to do so, because it will make him happier, because, in the opinions of others, to do so would be wise, or even right. (emphasis mine)
It gets worse.  Mill then admits big exceptions to his Absolute Principle:
[T]this doctrine is meant to apply only to human beings in the maturity of their faculties. We are not speaking of children, or of young persons below the age which the law may fix as that of manhood or womanhood. Those who are still in a state to require being taken care of by others, must be protected against their own actions as well as against external injury. For the same reason, we may leave out of consideration those backward states of society in which the race itself may be considered as in its nonage... Despotism is a legitimate mode of government in dealing with barbarians, provided the end be their improvement, and the means justified by actually effecting that end.
Two pages, two "ultimate"/"absolute" principles, each with a big exception.  Whatever your views, this is awful philosophy.

The world is full of Mill fans, who will probably complain that I'm missing Mill's fine and subtle distinctions.  But Mill's distinctions just pile confusion on confusion.  Examples:

1. The definition of "utility":
I regard utility as the ultimate appeal on all ethical questions: but it must be utility in the largest sense, grounded on the permanent interests of man as a progressive being.
But a man's "own good, either physical or moral" surely includes his "utility in the largest sense."  And Mill says that's "not a sufficient warrant" for violating his liberty.

2. The role of discussion:
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.
Unfortunately for Mill, neither his Ultimate nor Absolute Principles leaves any role for mere "capability."  You could say, "If free and equal discussion will improve a person, you should respect his liberty."  When words work, there's no reason to resort to beatings.  But after free and equal discussion fails to open the eyes of a person capable of free and equal discussion, why not try coercion?  No matter what a person's "capabilities," Mill's Ultimate Principle commands coercion and his Absolute Principle forbids it.

I freely admit that it would be easy to fix Mill.  Most obviously, he could keep his Ultimate Principle, explicitly demote his "Absolute" Principle to a mere rule of thumb, point out major counter-examples, then argue that people underestimate the negative effect of coercion on utility.  But this just proves my point: Mill wasn't a good enough philosopher to notice and repair elementary flaws on adjacent pages.

P.S. If you think that Mill was a pretty good philosopher for his time, check out his critic James Fitzjames Stephen.  I'm in much closer agreement with Mill's conclusions, but Stephen is a far superior philosopher.

Comments and Sharing

COMMENTS (13 to date)
Nathan Sumrall writes:

One minor quibble: the confusion you point out in the definition of utility disappears if we read (as I always have) "man" as "mankind" here, as opposed to a particular man's own good. This is consistent with other passages, particularly where he attempts to explain how self-sacrifice can be utility-maximizing (but also where he addresses the straw-man of utility as individual desire).

That aside, while my initial reaction is that you are a bit hard on him, I find this line of critique compelling and am interested to read the link.

Steve Waldman writes:

do you think it is incoherent to claim, for example, that jumping in front of a bus to save a stranger's child is "utility maximizing behavior"? i guess i don't. i've been infected by a very modern and tautological definition of utility that simply refers to a hypothetical function whose maximization explains an agent's behavior. therefore, we cannot evaluate whether behavior is "utility maximizing"; we observe behavior and assert, as a theoretical matter, the existence of some function whose maximization would induce that behavior. that assertion is itself not tautological: it is on the contrary rather a heroic leap to suggest that human behavior can be reduced to the maximization of some unknown and unobservable function, and that if this function were adequately characterized its maximization could be relied upon to predict and characterize ongoing behavior. but as heroic (and unlikely) as that assertion is, in modern quasiscientific usage, that's all that "utility" means, right?

my knowledge of and curiosity about the history of utilitarianism are quite limited. if you tell me that it is inconceivable that Mill would have meant this, if you say he would find the modern, cautious form of utilitarianism foreign and incomprehensible, i will gladly take your word.

but it is interesting that under the modern, cautious understanding, the apparent inconsistency between Mill's ultimate and absolute principles disappears. under the modern understanding, the two can be reconciled under something very akin to Star Trek's "prime directive". utility simply means doing what one will, unmolested by the coercion of others. if "what one will" means jumping in front of a bus to save a child, or putting ones hand in a meat grinder for no apparent reason, then that behavior is ipso facto utility maximizing behavior.

could Mill have meant anything like this?

Evan writes:
In the span of two pages in On Liberty, Mill names one "ultimate" principle and one "absolute" principle.
This is a problem right away. Human values are so complex that the only way you can compress them into one ultimate principle is to make that principle really vague and all encompassing. For instance "preference utilitarianism" is primarily an improvement over "pleasure utilitarianism" because preference is a vaguer and more encompassing term.

Moral philosophy will drastically improve once moral philosophers realize that attempting to create an all-encompassing moral theory that can be expressed in a few short sentences is impossible. You may be able to come up with useful one-sentence rules of thumb, but they will always have exceptions. If we ever develop a comprehensive moral theory it will be kilobytes long at the very least.

Snorri Godhi writes:

Thank you for confirming my understanding of what JS Mill said: I thought I might have missed something, but never went back to re-read carefully. It would seem that Mill tried to "prove" by assertion that utilitarianism and natural-rights (Lockean) liberalism are one and the same.

But it gets worse. Jump to chapter 5, "Applications", and you only have to read a couple of pages before Mill introduces an important distinction: economic freedom is motivated by utilitarianism; non-economic freedom (which Mill thought to be distinct) is a good in itself, a natural right if you wish.

The way I understand this is that Mill was an upper class Englishman who thought that making money is vulgar, and nobody could possibly value the freedom as to how to do it.

Chris writes:

Where do you think Fitzjames Stephen went wrong?

Bruce Cleaver writes:

@Evan -

I agree with your conclusion (a realistic, workable moral philosophy cannot be simple). I have noticed the same problem with some strains of libertarianism: they assert one or two principles (principle of non-coercion being among them) and proceed to construct the desired panoply of human behavior from those few principles - as if it were an exercise in geometry theorem proving.

Mike Rulle writes:

While it would be a non-sequitur to say that John Stuart Mill caused me to drop out of a PHD program in political philosophy after 5 years (in favor of getting an MBA in place of writing a thesis), I always use him as a story telling example of why I did drop out.

Richard writes:

Another good reasons to read Fitzjames Stephen is that he is one of the greatest writers of non-fiction prose of the 19th century. Passages of "Liberty, Equality, Fraternity" are so well-written that they almost hurt.

DougT writes:

Mill is inconsistent. I'm shocked. Next, you'll be posting how Aristotle is sexist. I'm shocked.

Mark Little writes:

I have not read Mill, I carry no brief for him, and while I have not much studied philosophy, from what little I've read of philosophical utilitarianism I believe I am not a utilitarian. But based only on reading this post, I do not see how the citations of Mill support your criticisms. Reading these Mill quotes with only ordinary charity in construing the meaning of an author, I read

1) Mill's "ultimate principle" is a meta claim that merely asserts his commitment to utilitarianism as the principle for evaluating purported ethical rules.

2) Mill's "absolute principle" is an ethical rule that Mill proposes, consistent with the ultimate principle. It merely asserts in a broader context the de gustibus assumption commonplace in welfare economics theory. That is, we should presume that you know better than I what will best enhance your utility. (And your personal freedom thus maximize the contribution of you mite to the aggregate utility; but that aggregating part is where Mill and the utilitarians go wrong.)

3) His qualifications are innocuous. We may agree to the presumption that you know best what is best for you, but I can rebut that presumption by correctly pointing out that you are insane, demented, grossly ignorant of your circumstances, or an infant. This is not bad philosophy, it is common sense.

The latter points in the post could be answered similarly. Really, I don't think I'm saying anything essentially different from

I freely admit that it would be easy to fix Mill. Most obviously, he could keep his Ultimate Principle, explicitly demote his "Absolute" Principle to a mere rule of thumb, point out major counter-examples, then argue that people underestimate the negative effect of coercion on utility.

except that you've chosen to give an uncharitable reading to Mill. Maybe more extensive excerpts would show why, and my reading the quoted Mill may be wrong, but as written the post is curious.

(@Steve Waldman: I agree with you.)

Addison Thiel writes:

"I freely admit that it would be easy to fix Mill. Most obviously, he could keep his Ultimate Principle, explicitly demote his "Absolute" Principle to a mere rule of thumb, point out major counter-examples, then argue that people underestimate the negative effect of coercion on utility. "

So your problems are that his absolute principle should not always apply because it sometimes contradicts his ultimate principle. I think he is looking at the big picture, where it must ALWAYS apply to achieve his ultimate principle. This is what his argument boils down to:

principle 1: we want to maximize utility
principle 2: we maximize utility by maximizing personal freedoms and not coercing others

And yes, even though others may think some people are stupid and make bad choices and should not be allowed to make them because they do not seem to maximize their utility, Mill says coercion will cause a decrease in total utility for society, even if some individuals appear to be better off.

Why? Possibly because people will actually be happier making their own bad choices than being powerless to choose (would you rather play a video game yourself and either win or lose, or just watch someone win?). Possibly because Mill thinks that it will be tough to draw the line for where a government is going too far and being counterproductive.

How about this example: In law, we require the prosecution to prove beyond reasonable doubt that a defendant is guilty. Yes, sometimes guilty people will walk free, but fewer innocent people will be convicted. We could remove the reasonable doubt clause, but we believe we would probably have more injustice, so we have it in place.

It is the same with Mill's argument. If you bend the rules sometimes (make it only a rule of thumb), they aren't really rules. Just like you can't have some people that don't apply to the law, you can't allow some to be coerced and others free. It must apply universally for the system to work, even if there are some inefficiencies.

"Two pages, two "ultimate"/"absolute" principles, each with a big exception. Whatever your views, this is awful philosophy."

Is any idea terrible if it has exceptions? Surely an economist wouldn't throw out the whole idea of the free market just because there are externalities that cause market failures...they just note them and recommend ways to correct for them....Mill does not think you can correct the inefficiencies in his ideals without compromising the whole thing and being worse off.

You say, "Mill's Ultimate Principle commands coercion and his Absolute Principle forbids it."
I think it is very clear his ultimate and absolute principles are to be taken together. I can sum it up in one sentence: For society to maximize overall utility, we should avoid coercion.
Of all your "fixes" for Mill, about all that he could improve is to be a bit more explicit about WHY coercion is bad, this would strengthen his argument. But I guess he probably thought that it was obvious and needed no explanation...

Maxim writes:

I believe Addison Thiel is correct.

Also, Mill's book "Utilitarianism" is really his clearest expression of the philosophy. I think you would be hard-pressed to find contradictions there.

"On Liberty" is where he discusses the policies and rules of thumb he believes maximize utility -- which is much easier to quibble with.

Jeremy Bowman writes:

Mill does make mistakes occasionally, but he's still a great philosopher. His two principles apply in entirely difference contexts, and are intended to guide different kinds of decision-making. His "absolute" principle is meant to guide political decision-making, for everyone including non-utilitarians, whereas the "ultimate" principle is constitutive of all morality -- at least as far as Mill is concerned, as a utilitarian. But Mill knows that others are not utilitarians. The first context is "public" and political; the second is "private", and concerns moral matters that differ from one individual to the next.

Since Mill knows most of his readers are non-utilitarians, he has to argue for the "absolute" principle in such a way that he can sell it to non-utilitarians. That often means downplaying "private" morality, including his own utilitarianism and its "ultimate" principle, and instead emphasising that toleration is a virtue for everyone, something we often do nowadays by putting the word 'morality' in scare quotes. If we all adopt that tolerant attitude, a political case can be made for taking the "absolute" principle as the basic rule of thumb for political decision-making, guiding new laws, government policy, etc. Since it is meant to be the basic rule of thumb, it had a different flavour from other rules of thumb.

Analogously, a medical reformer might encourage doctors to adopt as an absolute principle "do no harm". In doing so, he would have to be aware that different doctors have different religions, and different "private" ideas about morality -- some are utilitarians, some are Kantians, and so on.

Comments for this entry have been closed
Return to top