3 ms·
Good argument. To me the reason for not speeding due to it being a net benefit to society is just a specific case of speeding due to it being the right thing to
by johnsonjo 9y ago
Good argument. To me the reason for not speeding due to it being a net benefit to society is just a specific case of speeding due to it being the right thing to do. One point I got from your argument (though not directly stated) is that it's often important to clarify why the thing you are saying is the right thing to do is in fact right.
I feel your understanding of altruism seems to coincide with a utilitarianism standpoint of it is the right thing if it creates the greatest sum total happiness for society. I feel Utilitarianism is close to defining what is right, but if you have ever studied it Utilitarianism has some flaws (when I say this I don't mean to attack your personal standpoint on altruism/right only Utilitarianism in general.) Often these flaws can be fixed with a more rule based approach to Utilitarianism. I personally like Utilitarianism's approach, but probably take a more Aeristotelian approach which is more along the lines of it is the right thing if it makes me act in a more virtuous manner. Aeristotle believes those of truly virtuous characters wouldn't desire to do wrong things in the first place.
- Jach 9y agoIf you taboo the words "virtue" and "virtuous", what are you left with? Seems like it can be reduced to utilitarianism. If utilitarianism is "what is right is that which maximizes utility according to this utility function", this idea of virtue maximizing can be restated as "what is right is that which maximizes utility according to this utility function, which states its highest utility is that which makes me act more virtuously, with these comparison heuristics to determine 'more' or 'less' virtuous". My own criticism of utilitarianism is that utility functions aren't always objective, and for different people or different contexts can output different values. If you can agree on a function it can be a useful tool in moral reasoning (and especially uncovering inconsistencies if you've agreed to a few functions then find a contradiction with intuition and what the calculations say) but typically we want to know how we should agree on a function in the first place.