I have the following two definitions:

(i) Act Utilitarianism: an act is right iff it produces at least as much total happiness as would any alternative act open to the actor.

(ii) Rule Utilitarianism: an act is right iff it conforms to one of that set of rules, general adherence to which would produce at least as much total happiness as would adherence to any alternative set of rules.

I fail to see a difference between these two definitions.

I will prove this by showing that, in any given situation, X , if act A is optimal under (i) then it will also be optimal under (ii). I would like someone to tell me why this proof is wrong.

  1. Suppose that, in situation X , Act A is optimal under definition (i).
  2. Consider any set of rules, B, i.e. any possible candidate for the optimal set of rules under (ii). Define B' as:

Rule Set B': Follow Rule Set B in every case except for in situation X , in which case do act A

  1. Then B' will ALWAYS be equal to or better than B.

  2. Thus, it follows that the optimal set of rules will always mandate Act A in situation X .

  3. So, by definition (ii), Rule Utilitarianism mandates Act A in situation X . Thus, the two definitions are equal.

I know there are two places where I was a bit careless. I did not really prove the jump from 3 to 4, but I think we can agree that this is valid. I did not prove the converse, which I don't believe is necessary anyways.

I'm looking for someone to tell me where I am wrong...

I'm looking for rigorous answers that take into account the specific definitions and logic that I claim follows. Not looking for your explanation of what the the two types of utilitarianism are.

  • 1
    On act utilitarian stealing a smoothie from a vendor does not hurt anyone that much and might bring you great pleasure, so it is right. However, if this becomes a habit, or if everyone does it, the consequences might be unhappy. So it can't become a rule, and is wrong on rule utilitarianism. The mistake in your argument is that your "any set of rules" is as equivocal as "any set" or "any integer definable in fewer than twelve words", which integers are so depends on which have already been "defined". The rules have to be fixed once and for all, you don't get to "adjust" them for each act.
    – Conifold
    Commented Apr 4, 2017 at 20:24
  • 2
    Possible duplicate of What is the difference between Rule Utilitarianism and Act Utilitarianism?
    – virmaior
    Commented Apr 4, 2017 at 22:11
  • But to add a comment in addition to a close vote, the key difference is that in act utilitarianism, the agent herself constantly calculates the legitimacy of her actions whereas in rule utilitarianism the agent follows a rule which is expected to in most circumstances maximize utility. / i.e. act agent decides whether to tell the truth based on whether it maximizes utility; rule agent follows a pre-determined rule that (arbitrarily for our purposes) says tell the truth, and the rule was at some non-immediate point decided under the belief it maximize utility across most cases.
    – virmaior
    Commented Apr 4, 2017 at 22:14
  • @virmaior this is pretty clearly not a duplicate. not sure if you even read my post but I create a novel proof and am asking for opinions on it. literally has nothing to do with that other post except for the topic at hand. Commented Apr 5, 2017 at 14:20
  • @Conifold Basically the crux of my proof is that the "optimal" rule set as defined by rule utilitarianism necessarily simply breaks down into a brute force set of rules, listing every possible situation and every possible optimal action--a.k.a. act utilitarianism! Commented Apr 5, 2017 at 14:27

1 Answer 1


Okay, I don't think this question is substantially different from another question which I've already marked, but I'll do my best to give you an answer here anyway. Simply put, you're misunderstanding the difference between Act utilitarianism and rule utilitarianism.

I will define act utilitarianism and rule utilitarianism as follows:

Act utilitarianism:the position that in order to maximize some desirable feature (or minimize some undesirable feature), each agent should for each action pick the course that produces the optimal outcome on the basis of his or her individual knowledge and expectations.

Rule utilitarianism: the position that in order to maximize some desirable feature (or minimize some undesirable feature), rules should be established that produce the optimal outcome across many actions incorporating the limitations and constraints on the information available to those agents.

Thus, both seek to optimize outcomes, but act utilitarianism does so by having agents optimize their own acts and rule utilitarianism does so by optimizing rules for actions (not generally at the agent level).

You propose the following rule:

Rule Set B': Follow Rule Set B in every case except for in situation X , in which case do act A

What's proposed in B' is basically a reduction to the following rule: always be an act utilitarian because this produces maximal happiness. I take it that's your point, viz., that the best type of rule a rule utilitarian could come up with is to be an act utilitarian. And that's probably true on the definition you supply.

But the definition you supply differs from the one I suggest here precisely in that it does not incorporate the problems with calculating utility on the spot that rule utilitarianism seeks to address.

Rule utilitarianism exists primarily because act utilitarianism on the view of many suffers from an epistemic problem. This is inherent in the formulation of Mill (and Bentham) which does not spell out an answer to the following important question:

*Am I optimizing actual or expected consequences? (and do I know with certainty the outcome and effects of my actions? do I have enough knowledge or depth of processing power to calculate correctly?)

In other words, am I doing right if I do what I think produces the most happiness or am I doing right if I do what produces the most happiness?

Both horns are problematic. If I pick the think horn, then Mao killing the sparrows occurs. And Mao is a great guy there for wanting to supply enough food.

Conversely, if I pick the produces horn, then it's not clear that my choices matter for ethics. My accidental choice to not buy the cake on discount may lead to great happiness for a poor family even though I never thought about it. Or my dropping change on the floor might lead to a horrific accident. Here, we'd have a very strong form of moral luck.

The rule utilitarian hedges the question a bit better by coming up with rules that are designed to optimize said goal but that don't depend on the individual agent arriving at the epistemic best path under the stress of the moment. It doesn't completely eliminate the horns, but it dilutes their painfulness, because it can suggest that our duty is to follow rules that seem optimized to produce the ends we want even if in particular cases they don't.

Thus, we can return to your reduction. The rule utilitarian rejects it because it doesn't sufficiently grapple with the agent's epistemic realities. A major point of transitioning from act to rule is to get away from the vagaries and gambling that arise from making judgments at each particular juncture for each particular act.

Thus, for the above reasons, I don't think your reduction truly captures the position of the rule utilitarian. Instead, it just makes him an act utilitarian who doesn't think rules add anything.

  • I think an acceptable answer here needs to break down the language of the definition of Rule Utilitarianism that I provide.. i.e. Rule Utilitarianism: an act is right iff it conforms to one of that set of rules, general adherence to which would produce at least as much total happiness as would adherence to any alternative set of rules. This is what I'm basing the logic my proof on, and to me it seems to be a very straightforward, essentially mathematical definition. If you disagree with it that's fine, but for the purposes of this discussion you need to stick to this. Commented Apr 5, 2017 at 15:44
  • If you would like to, as an aside, provide a definition that you deem to be more accurate, then I would be interested in seeing that as well. Commented Apr 5, 2017 at 16:19
  • "rule utilitarianism" is an existing view, so I've answered according to my understanding of that view. (see for instance: en.wikipedia.org/wiki/Rule_utilitarianism ; plato.stanford.edu/entries/consequentialism-rule ). My answer is based on my understanding of what motivates people to actually have that view. So conversely, I ask, please find a rule utilitarian who uses the definition you supply. My sense is that such a person doesn't exist -- because if they accept your reduction, then they would logically just be act utilitarians
    – virmaior
    Commented Apr 5, 2017 at 22:41
  • But I'll amend my answer to explain the views a little better.
    – virmaior
    Commented Apr 5, 2017 at 23:54

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .