Tim Urban on cosmology
Go hug someone pic.twitter.com/I63AaP1xQu
— Tim Urban (@waitbutwhy) September 16, 2021
Robin Hanson on value drift
Value drift is just a generic problem for humans, ems or AI. It’s just what’s always happened so far. It’s the default for what will happen in the future. If you hate it you’re in trouble because it’s really really likely. Some people for some reason think that value drift in humans is bounded well, while value drift in machines is not only unbounded but happens quickly and I don’t really see the grounds for that. […] The main thing is that in the past when value drift happened change was so slow that you didn’t see it in your lifetime so you didn’t worry very much about it. As change gets faster and your lifetimes get longer your life will encompass more value drift. And then whether it’s humans or machines or whatever you will see it, and if you don’t like it then you will see something you don’t like.
John Richardson on Nietzsche’s metaethics
[Nietzsche discusses] three main ways of valuing: the body’s, the moral agent’s, and his own. We could also call these animal, human, and superhuman valuing. Each has its own “semantics” (or “intentionality”); that is, its own way of positing its values as good. So Nietzsche has what might look like three separate and inconsistent metaethical positions but that are really three elements in a unified account of valuing.
[…]
Our drives value simply by using signs to steer by (toward). They “see” or “interpret” their values as good just by using them this way. They don’t posit them as “true” to anything outside them. Instead they judge and adjust these signs as they learn how well they “pay off” in expanding power. As we saw, the drives don’t recognize what they’re doing as they value. They don’t see the “frame” of their valuing around their values; they lack the perspectivist truth. But they refrain from the externalist mistake of thinking their values tasked to match real goods outside.
By contrast Nietzsche thinks that our agential valuing does make that externalist posit. This is one of its main impositions on our drive-valuing. In order to “tame” the latter for social life, the habit of obeying external norms needs to be inculcated. It’s to license this habit of obedience that the conviction is gradually ingrained that there are real values outside one’s valuing that one needs to align it toward.
[…]
The historical character of this posit and the way it is overlaid on a deeper valuing that doesn’t make it suggest the contingency of such externalism. They support Nietzsche’s optimism that human can find a way to grow out of what is only a (deeply settled) bad habit.
[…]
Nietzsche’s frequent expressions of [error theory] are unsurprising given what we’ve just seen: they apply to our agential, moral valuing which does indeed claim its values to be real—which they’re not.
[…]
But this error-theory does not apply to the two other ways of valuing in Nietzsche’s scenario. Bodily valuing makes no truth-claim, and his own valuing does, but a different one that (we’ll see) has a chance to be true. Nietzsche denies that all valuing makes the mistake of positing its values as real. And why indeed would he allow our agential-moral valuing to represent valuing in general? Human is “the sick animal” due precisely to the defective way it values. Nietzsche’s return to “natural” values is his effort to bring our conscious and worded values into healthy alignment with our drive-valuing; this will include undoing that false posit.
[…]
Nietzsche justifies his values by direct appeal to the values we already have. He tries to point out values we have without noticing them. The “ought” is supplied not from outside but by what the person values already. He claims only to offer the means by which that valuing will want to improve itself.
By his perspectivism, Nietzsche gives credit to our existing values as the only determiners of what’s good for us. So his appeal is ultimately to these. But our valuing of these values includes a will and ability to improve them, in the two fundamental respects we noticed in §1.4. We will to improve them as signs for power—a will embedded deeply in us just as living things. We also will to improve our values in how well they face the truth—a will bred into us humans and indeed distinctive of our kind. These deep aims function as second-order or meta-values, criteria by which we will to improve our first-order values.
John Richardson, Nietzsche’s Values, Chapter 1
Elijah Millgram’s pragmatic critique of internalism
That motivations fail to agglomerate is exhibited in the most striking logical feature of internalism (and of its cruder relative, instrumentalism), namely, that one’s bottom-line desires and projects are incorrigible.
[…]
You want what you want, and someone who insists that you are wrong to do so, when mistakes about such things as how to get what you want are not at issue, is just bluffing.
Subjective motivations can change, in all manner of ways, but they cannot be corrected, and this means that nothing could count as the rational investigation, on the part of such a creature, as to whether its bottom-line guidelines and priorities were correct. Since the creatures do not correct their own motivations, the design strategy is reasonable only if they do not need to; in other words, only if, for the most part, the designer can equip them with motivations (or ensure that they pick up motivations from their surroundings) that will not need correction. That in turn is feasible only if the designer can anticipate the practical problems his creatures will face, and only if the guidelines his creature would need to negotiate them are sufficiently compact to be stored and accessed. Given plausible cognitive constraints on processing, memory, and so on, that in turn requires that the environment the creature is anticipated to face be both stable and simple.
Coloring in the line drawing, we see that [Bernard] Williams’s alethic state of nature is something on the order of a tourist-brochure version of a village in the hills of Provence, where life goes on as it has since time immemorial. The villagers work their plots of land, growing the same grains and vegetables they always have; they herd their sheep and goats; they bake rustic bread and knit rustic clothes; they hunt rabbits and deer; they build houses out of the local stone; they marry and raise children; when they get old, they sit outside the village pub and drink pastis; they play boules in the park; eventually, they die, and are buried in the cemetery behind the church. The internalist design solution is satisfactory for this form of life. The designer knows that his peasants will have to work the fields, so when it comes time to own a field and work it, they come to have a desire to do so. They need to be made to reproduce, and thus are built so that, when they get old enough, they will want to have children, or anyway want to do things that as a predictable side effect produce children. Not all of a subjective motivational set need be hardwired, of course; a disposition to mimic others, and to learn and adopt one’s elders’ thick ethical concepts, will keep the games of boules going and the pastis flowing. Because life in the mythical village never changes, there is no need to delegate to the peasants themselves the task of investigating what their motivations ought to be, and no need to equip them to correct their motivations; thus, there is no need to complicate their cognitive or normative systems with the gadgetry that would take.
[…]
Analytic philosophy has done something that is quite peculiar: instead of making sense of humanity, we have been philosophizing for the inhabitants of a romantic fantasy of traditional peasant life.
[…]
Instrumentalist (or “Humean”) theories of practical reasoning are how philosophers talk through the strategy of hardwiring designated objectives into an organism, so that it can execute a life plan suitable to a stable environment. Your environment is no longer stable enough for relying on desires to be a decent strategy. Instrumentalists (“Humeans”) have a view of practical rationality suitable for a cruder, simpler species.
The Great Endarkenment, ‘D’où venons-nous . . . Que sommes nous . . . Où allons-nous?’
quote elijah millgram internalism pragmatism bernard williams
John Richardson on Nietzsche’s naturalism
Nietzsche thinks he cares more about truth than other philosophers do. This is partly because he is not in thrall to a moral bias, but also because he understands better the kind of truth there really can be—the kind humans can and do have. So he rewrites philosophers’ previous idea of truth while still giving it preeminent value.
[…]
In announcing these truths he contributes to what he thinks is a prolonged, ineluctable process by which our modern scientific will to truth finally faces the truth about values—the last and hardest topic for it to face. As these truths are exposed, our culture, and the rest of the world through it, is confronted with a great spiritual crisis and challenge: How can and will we go on to value once we have uncovered these truths about our valuing? How can we value, now for the first time, honestly (i.e., while facing the truth about what we’re doing)?
[…]
It is extremely difficult to do so because this truth tends to undermine our [values] […] insofar as they involve a framing claim that these things (that are valued) are really, independently good. For the truth, Nietzsche holds, is that all values are dependent on valuings—are “perspectival.”
[…]
Recognizing Nietzsche’s idea of values as signs is the key to much of his thought about them. Seeing a value as a sign, we see why he insists that it’s not only humans that value. Animals are clearly responsive to signs in their perceptual discernments. So a predator may employ a certain smell as a sign of prey. And we can see ways that plants are responsive to signs as well. Nietzsche holds that willing (or aiming) is something that all organisms do. It depends not at all on consciousness.
[…]
Our human values, as worded, are distinctive in being held in common, as norms. They are accepted because this is “how one values” in the community to which one belongs. They thus serve a “herding” function, which strengthens the group but at the expense of members’ individuality.
John Richardson, Nietzsche’s Values, Preface