A man may be said to be a partizan of the principle of utility, when the approbation or disapprobation he annexes to any action, or to any measure, is determined by and proportioned to the tendency which he conceives it to have to augment or to diminish the happiness of the community: or in other words, to its conformity or unconformity to the laws or dictates of utility. [IPML 1.IX]
This is as close as Bentham comes to defining what it is for someone to be a utilitarian. Yet it is given strictly in terms of approval, not as "belief of correctness" or anything like that. This definition has some obvious advantages. Firstly, this definition at least implies something about the man's character. If a man merely believed that actions which produce sub-optimal levels of happiness were (to this extent) "wrong" - and this "wrong" being a description of the action or its properties, and not involving his own feelings - it would say little about him. What if he doesn't care about whether things are right or wrong? Secondly, a definition in terms of correctness is very hard to resolve: what exactly is the propositional content of "utilitarianism is the correct system of ethics"? [On this, see Dave Pearce: This objectivity doesn't entail that valuable experiences can have, as distinct from simulate, a type of mind-transcendent, truth-evaluable "propositional content" over and above their intrinsic phenomenology which somehow manages to alight on properties of the mind-independent world. HI 2.19 (my emphasis)] Yet problems with the approval definition remain.
Imagine the case of a man who approves of actions in proportion to his conception of their conduciveness to happiness, except (say) when he is being tickled. When this man is being tickled, he approves of the tickles - and anything else he brings to mind - in total disregard to the happiness of others. Shall we say that this man is a utilitarian nonetheless, or that he is a utilitarian except when he is being tickled? It seems awkward. One possible solution is to admit that there are degrees of utilitarian-ness: a numerical measure would be the correlation between the approval by the person, and their conception of the object to promote general happiness. But this opens up another problem: Imagine a man who conceives actions to be conducive to the general happiness, in proportion to his approval of them.
In other cases, of men who are utilitarian by Bentham's definition, it remains to be answered "Why are they so?" (and, if indeed this be a different question, how did they come to be so?). That is, how (and why) are their feelings tied to this standard, the happiness of all? Is it not true that our feelings are either there, or they are not? We approve of something, or we do not? Predicting this line of questioning, Bentham asks whether - assuming that calling something right is expressing approval - whether, by admitting the existence of such feelings without external reference, we imply that
"there are not as many different standards of right and wrong as there are men? and whether even to the same man, the same thing, which is right to-day, may not (without the least change in its nature) be wrong to-morrow? and whether the same thing is not right and wrong in the same place at the same time? and in either case, whether all argument is not at an end? and whether, when two men have said, "I like this", and "I don't like it", they can (upon such a principle) have any thing more to say?"
Now what is here to stop us taking up the gauntlet? What would stop someone from saying that, in as much as "right" expresses approval, there are as many different standards of right and wrong as there are men? That things are both right and wrong in the same place at the same time? The answer would seem to be - dislike. We dislike the idea that these things are true. At the base of it is Bentham's insinuation that such a position involves offering "sentiment or opinion as a reason for itself". Of course nothing could be a reason for itself - or could it? In any case, what need is there of reason? A feeling exists, or does not exist. If it exists, it is sufficient. If not, then there is nothing to argue over. And anyway, how, exactly, does reference to a consistent standard provide a reason? (Ultimately, the question "what is a reason?" has not been satisfactorily answered. And if we cannot say what a reason is, how can we claim to know what it is we have reason to do?) And again, even if there must be reference to some external standard for there to be a "reason", which standard would it be? What ultimately elevates "the amount of happiness in the universe" above "the number of plastic forks in the universe"? Why is one number to be maximized, and the other trivial? Because we dislike the idea of attaching importance to the number of plastic forks, and are more comfortable believing our feelings to be determined by the feelings of others? But why? Eventually, we are so inclined, or we are not. Inasmuch as "why?" asks for a justification, maybe there is no "why".
So this is my ultimate problem with defining "utilitarian" without reference to the "correctness" of the theory: How does someone who does not believe utilitarianism to be correct, come to remain a utilitarian? Co-incidence? "It just so happens, I feel inclined to approve of things in proportion to the tendency I believe them to have to promote the general happiness". Which is of course sufficient, for as long as it lasts. But the first time they notice "It's not conducive to the general happiness, but I approve of it anyway", why (and how) would they ever go back?