A phrase taken from Ziz, it means “this world saving plan is good, but it isn’t enough to save the whole world”. I use it a lot to explain my thinking to normal people; I often get asked why I don’t write things up as academic papers, or do a startup that would predictably-in-advance earn lots of money if I sunk a year into it. The answer is universally that I have more impactful things I can be doing with my time. If the sum total of things you expect to do over your life isn’t enough to save the world, then you should try to figure out something else to do with your life. Writing papers and doing startups is not part of a trajectory that lets me e.g. end aging or solve another problem on that scale.
From here, this is an assertation that even if some action you could spend your time on is commensurate with having a plan that saves the whole world, that you still shouldn’t do it, because it is too morally abhorrent.
There’s a particular failure mode people fall into, where they think that because the world is terrible, then they’re justified in taking actions that directly hurt others if those actions can be justified within either a utilitarian framework or within an existing power structure, e.g. “how things are done”. An example from fiction is, from s1e10 of Avatar the Last Airbender, Jet decides to take out a village filled with enemy soldiers who were abusing and killing the civilians, which would take out the civilians too. But ultimately, the reason soldiers would themselves abuse citizens mostly comes down to trauma. A real life example is saying that your enemies should be imprisoned in horrible conditions, or even tortured, because “maybe it’ll incentivize future people to do what you want, and besides, that’s just how things have to be”.
In the way that trauma is the enemy of life, fighting trauma with trauma just hurts people who haven’t given up on hope yet, it just psychologically breaks them. Utilitarian analyses can become corrupted if the person doing them is traumatized, because trauma wants to spread itself, and that will seep into the utilitarian analysis in perverse ways. Hence, “How can you even consider that? There’s got to be some other way” is not about ideological purity, but about avoiding a certain class of blind spots that systematically bias utilitarian analyses in a certain direction.
The normal definition of the word, with the added understanding that since trauma is present in a physical system (the brain), that getting rid of it is fundamentally an engineering problem.