Ill will? -- Of course not! Those who I consider my closest friends are those who are willing to debate vigorously with me. I'm told I often sound angry when discussing philosophy. Perhaps this proclivity is evident in writing as well. If so, my apologies. More than anything, I'm enjoying myself, and this spat of ours.
About the appeal to consequences:
If you can somehow turn the statement "X action produces good consequences, thus X action is good" into "I think X is good to do, therefore X is good", then by all means do so. I'm having trouble, though.
About happiness:
Happiness is an emotion, and an ineffable one at that. I'm not sure it can be described, per-se, without resorting to synonyms. Science may have something to say about this matter. For example, it might be able to describe brain states that are often present when people say that they're 'happy'.
Are they therefore less moral than other people as well?
This could be an equivocation over the term 'moral', depending on how you meant it. If I am sad, I am not immoral, and similarly if I am happy, I am not necessarily moral.
A moral person would be one who tends to act with the intention of promoting happiness, and avoiding or negating unhappiness. I'm not entirely decided over accountability under a utilitarian scheme. In fact, I'm not even sure if it's rational, under utilitarianism, to hold someone accountable in the traditional sense.
Under utilitarianism, you would lock up criminals for three reasons (that I can think of): to deter others from such consequentially bad behaviour; to stop them personally from doing it again during their time; and to rehabilitate them, if possible.
As for your example of the hammer, you're thinking in terms of accountability, when you need to think in terms of consequences. There's no consequential reason to punish you for selling them the hammer.
The baby example is interesting. Did Hitler actually think he was doing the right thing? Did Lenin and Stalin? But the crux, once again, is the terminology. It matters not whether the mother -- or Hitler -- is guilty, or whether they actually did the crime; what matters, is whether punishing them will produce better consequences than not pushing them.
About your chaos theory point:
I agree that it's impossible to know
precisely the consequences of your actions. However we can still practice a poor-man's utilitarianism, by doing our best to predict, and giving our causal reasoning, with historical examples to back our case.
Your argument that it's "impossible" to practice dubious. You wouldn't say that it's impossible to live life at all, would you? But that's what you do every day of your life: you predict the consequences of your actions, then you act. Utilitarianism may be impossible to practice perfectly, but I fail to see what's wrong with that.
Scarcity and negative utility:
The idea of 'potential utility' being equal to 'utility' is questionable. It's not really the same thing, or equivalent, but I'm not sure how to grade it.
I still think you'r missing something in the example, though. Let's say that the ten people on the island are happy 80% of the time, and unhappy 20% of the time, each. This gives them a utility ratio of 1:4 negative to positive, rather than 1:1, that is, your '0 utility' score. Let's say their children are the same, as are their grandchildren, and any future generations that can survive on the resources as well. Now, every generation born on the island has grown up knowing that there's limited resources, and that at some date, they will die out. This makes them accustomed to the idea that they and their friends might eventually die before their time. So instead of that, they decide on euthenasia -- that when the last of their supplies runs out, they'll spend a night celebrating, then all kill themselves.
Where's the negative utility in this scenario? As long as more people are happy, then the utility value will not be negative. There's nothing 'necessary' about the negative utility when there's finite resources; it's up to the psychologies of the people who exist. Happiness isn't inseparably linked to resources, nor does it depend on them.
About the patients:
In real life, there is
always more context than you've given me. Your final example is amusing though, and quite relevant to an earlier point.
Yes, you would be morally wrong for buying the pretzel if, for example, you know that you don't like pretzels, and that this pretzel will actually make you sad. However only by deontological standards does someone inherently deserve punishment for doing something morally wrong.
Maybe one day we'll create artificially intelligent mind-reading vending machines with an interest in punitive justice and moral theory, so we can ask them their opinion on the matter.