I like the reasoning process you've used, but I don't think it escapes the naturalistic fallacy at all.

I follow you from (1) to (3).

But I don't follow you to (4). Why can't we ask why we have the goal? Why can't we drop the goal if it's counterproductive or impossible?

We do this all the time. "My goal is to get a high-paying job." "Oh? Why?" "Because I want to buy Taylor Swift." "You mean her music?" "No, Taylor. Herself." "You can't do that." "Really? Darn. Well, then I always dreamed of working in a coffee shop."

I already granted that if we're just describing phenomena, we can't escape ourselves--we can't do something we can't do. So in that sense we are indeed constrained by the goals we actually have.

But you haven't actually argued why we can't ask "why why why" over and over again until we run out of reasons. You declare that we have the only sentient willful perspectives we know of, but you don't explain why we can't ask why we have the perspectives we have. Maybe that will give us better perspectives on our perspectives.

Obviously the answers are going to have to be interpreted by us, because there's nobody else to do it, and it's going to have to cohere enough with goals we actually have to motivate us. But this is just a description of what is. There's no ought here.

As I described before, when you dig like a three year old who doesn't want to go to bed, you end up with the realization that the behavioral control system with goals is ultimately there because it aids us in perpetuating our kind. That's as far as the why goes (until it turns into physics and loses its emergence from microscopic happenings). But it does go there.

The universe doesn't care if, now that we can grasp this cognitively, we follow it, or if we reject it, just like it doesn't care whether or not the moth bashes itself into light bulbs. But we might care, because we like purpose, we like generalization, and we like that it gives powerful explanatory power to our otherwise rather random-seeming revealed preferences.

We might not. But we might also eat too many sweets, have bad teeth and diabetes, and argue, "well, the me back then wanted to be happy and they did...the me now can't say that past-me was wrong". I might think, "Well, I want to bash my finger with a hammer." And I do, and that makes me happy for a split-second; then my finger hurts and I am not happy. Who I am I to argue I was wrong in that instant when my goal was to feel good and for that instant it felt good to swing that hammer?

Once you admit that, "Well, okay, look, our goals aren't that myopically stupid. We can in fact formulate goals that last longer than a fraction of a section of ephemeral thrill," there's nothing that rationally stops you from formulating larger-scale goals--and you can keep going all the way out to evolutionary constraints on existence of creatures like us as the goal-generation mechanism, and then you run out.

Now, we might decide that it's too much fun running around with swords trying to cut each other down to stop, even though we'll end up bleeding and disembolwed in a matter of minutes. But I don't think so. I think our goals can generalize beyond a few minutes. We might decide that we should lie, cheat, and steal, as long as we're not caught...and some people do. But most don't, because we can generalize beyond ourselves, especially with the aid of empathy that helps us get direct emotional feedback. And I think, as a practical matter, if we think about it, we can generalize beyond the immediate people we know to just people. Maybe life, maybe not; that's less clear to me.

If you insist that people stop at individual-feeling-good, there's no better logical reason to than there is to insist that you not consider longer than a half-second into the future. It's just an arbitrary threshold, set by you because it does kind of nicely fit a regularity in nature, but it's not the only one. You picked it. I might not. I might pick song lyrics instead: "Give me everything tonight / for all we know we might not get tomorrow." It's not a very good strategy to get people to have a 401k plan for retirement, though.

I also think you have an over-expansive definition of happiness. People do all kinds of things that they think are important which nonetheless cause considerable discomfiture and sadness. People go on near-certain-death missions because they think it's important, not because they're giddy with the idea of what's going to happen to them and their family. Yes, their goal-directed behavior does make them do it, but unless you define "happiness" = "engaging in that goal-directed behavior which you actually engage in", I don't think it's a very good use of the word.

Indeed, I think it is the confusion between the conventional definition of "happiness" (which aligns well with what advertisers try for in rousing desire in you) and "what we judge to be maximally important" that makes it seem like the individual goal-setter has to be the basis of all ought-statements. Once you notice how individuals act, when the situation gets very very very important, you find extremely high levels of self-sacrificial behavior with no realistic prospect of individual happiness, like running into burning houses to save one's children.

"That's dumb; being burned yourself makes you unhappy so let the kids burn," is an ought-statement that seems, to most people, grotesquely horrific. "Wow, she burned to death saving her kids; how stupid to go through all that pain," seems unbearably callous for what most people consider a noble sacrifice. "OMG, she burned to death saving her kids; she must have been so ecstatically joyful in her agony for that to make her happy!" seems shockingly cruel in its misapprehension of what was going on.

So my auxiliary claim is that, actually, the goal isn't pointed at the self, and it's not pointed at good feelings in any conventional sense of the word. It really is pointed at continuation of life. It's just that most of the time we can't tell because as a distributed system, self-care is important, and self-interest works well because we each have maximum information about ourselves; good feelings are the most common and prominent way to guide us in that regard.