1. Regarding maximizing outcomes: I wonder if you’ve seen Courtney Denning’s article from several days ago (a short personal story) https://medium.com/wholistique/how-to-live-a-more-sustainable-life-and-maintain-your-sanity-edb1715435d7 and my article from December (a longer article, half of which reacts to a Peter Singer book) https://medium.com/books-are-our-superpower/should-you-maximize-your-ethical-outcomes-201e3695a857
2. I detect a meta-question in here: "Should we use the word 'should'?" Yes, obligations, expectations, and desirability are important concepts and real forces in our social world...and, yes, on the other hand, talking about what we "should" do often just generates shame and doesn't lead to improved behavior, leading to a net loss for human happiness and thus frustrating the goals of utilitarianism. There's a level on which (as you mentioned, as Kant mentioned) we *should* do something only if we *can* do it. Perhaps, on the meta-level, we *should* use "should-language" only if speaking that way *can* lead us to change our behavior regarding the thing we actually *should* do? i.e. We still should do the thing, even if we choose to avoid fruitless conversation about the fact that we should do it?
[Related: The question of why we listen to certain voices as "moral experts" and not others. Nikki Stern has a book about this, "Because I Say So: Moral Authority's Dangerous Appeal," reflecting on how, as a 9/11 widow, she was often asked for her view on foreign policy, as if her personal anger and grief could justify whether a nation should go to war; she was aware it could not. https://bookshop.org/a/4480/9780692938317 ]
3. My one point of disagreement is the paragraph about forgiveness. It's not that I disagree with the concept, but I'd use a different term. My immediate intuition is that you're talking about something more like "immediate excusability," not "forgivability." My reasoning is that, when we "forgive" someone, it's usually because they've done something really wrong and thus we can coherently speak of forgiving or not forgiving them. Example: Most people have had at least one very difficult person among their inner circle—family, friends, colleagues—who has lied, stolen, manipulated, assaulted, etc. and generally done all the things a human is definitely not supposed to do. Forgiveness has something to do with letting go of our own resentment toward them and then reassessing (in the absence of our resentment) whether we want or need to re-enter a relationship with them or else just gently let them go. If we have gotten to the point where we are deliberately forgiving them, we already know that we ourselves should *not* act the way they did. We probably believe we *wouldn't* have done it, either. If, by contrast, someone behaves a certain way and we can swiftly excuse them (because we can empathize with their feelings, rationalize their decision, or both), then, yes, arguably, we are saying that we might have behaved the same way if we were in their shoes.
I wonder about this idea of "excusability," too, though: In excusing them, maybe we are just recognizing that we have the same type of character flaw as they do, so we personally would likely commit the same error or regrettable choice, but we imagine that someone else—a hypothetical "better person"—might be able to choose a better path. In that case, we're simply saying that we have no personal moral authority to criticize and weigh in with a "should have"...and/or we're saying that using "should have" language about the situation might be ineffective because the person whose behavior is in question isn't likely to respond to that language.
