I wasn't considering that, and it does make the question more critical. I'll attempt to write something and it might go poorly. 1. From the subjective standpoint, forgiveness from me. Being a Catholic or at least something approximate to it, I believe in unconditional forgiveness, no questions asked, yes, even for the worst people ever. But forgiveness has this subjective connotation as well as the objective connotation of what we do physically (as opposed to spiritually) with them, and forgiveness in this sense has to be more choosy for the society to work. 2. So how do we choose? We might build a theoretical model, and then apply it practically, or weave in the practical aspects as we go. If I'm being asked theoretically what the state should do, now (in some sense) there is an ethical choice to my response (do I lie and give a partially practical one if I think I'm better at weaving in practical elements than the state?). I honestly don't think the distinction between theoretical and practical can be applied so easily, an order that looks practical from my end looks theoretical from the perspective of the person I order, so it can't have anything to do with the content of the decision. 3. Now I'm thinking about the so-called objective 'perfect' decision. There are a lot of philosophies that so-to-speak posit the existence of this, i.e. some algorithm that gives the right decision for every case. I think this sort of ethics-on-ice is not only the wrong way to literally go but also a bad idea to have as a guiding principle or limit. That ethics sometimes happens like this has been too quickly extrapolated into the notion that it is always like this, and the systems (like Kant) that do this well are useful to consider but it leads almost invariably to hack philosophy (e.g. modern utilitarians quibbling about mildly different formulas). 4. I don't think the question can be phrased from the eye of God, as it were; or in another way, if I am being asked "Do you punish them or do you forgive them etc.", then someone is asking me in a time, in a place, with some motivations and expectations that I have to consider. (There is the sense in which it can be phrased from the eye of God, but only in the thought-experiment, and - I don't know how else to say it - there is a different grammar for the thought-experiment than for the real ethical question.) 5. I think you wouldn't be satisfied if I incorporated too many 'outside' components into the question - the political pressure from different sides, the optics of a too-lenient sentencing, the inexpedience of having the trials be too long, reduced punishments for those who snitch on their comrades, pre-establishes laws or precedents that have to be followed but aren't exactly right for the situation etc.. These feel incidental in some way, we should in some sense pretend they don't exist. And now we have what some might take to be the pure question, but I think something has changed about it, added or removed. To answer it, I would just say I would do my best to be fair. This is not satisfying, but it seems to me that the form of the question forces that answer upon me. If I were to say things, like I expect the book to say, like if the soldiers did x, punish them this way, but if they also did y, punish them less, and if they did z also... If I answered in this way it might give a greater sense of satisfaction, but it wouldn't be a true satisfaction. (Or maybe more precisely it would satisfy the grammar of the thought-experiment, but not of the real decision in question.) 6. This, I think, arises because the ethical decision is really a subjective one, 'subjective' as in 'pertaining to the subject', not as in 'opinion', and the subject does not think in formulas and books, unless he is a philosopher, and then only sometimes. (I want to say the ethical recommending is related to the acting, in the same sense Wittgenstein relates willing to acting, but I'm not able to explain this in a few words.) To recommend an ethical action is closer to recommending a book than it is to recommend a move in chess. I'm not sure if this explained anything, but maybe this in some way answered the question.

Yes, you're right, I was being too pessimistic about everything. It's perhaps better for me to say - while still being mostly speculative - that we're getting incredibly diminishing returns from academia, both because of the passage of time and the increase in degrees, which means wasted energy exists in a lot of places. And that problem of 'wasted energy' is nowhere near as important as many other problems, and it is very hard to properly diagnose, and in some sense it is not really a 'problem' at all, so it won't get solved or even generally recognized for maybe the next 20-30 years. I don't even know how it could begin to be solved personally. I tried to avoid saying things like "all theoretical concepts are meaningless to a certain capacity" - this usually expresses a particular sentiment of the beholder more than it expresses a fact, and I haven't ever seen the fact explained to me more deeply in a way that feels both satisfactory and true to the sentiment. That isn't how I meant to come across.