There are quite a lot of assumptions about culture in this post which are popular but not well justified. Morality may well be partially relative, but if your view is that slavery’s evil was not objective, I think we disagree on a fundamental level. Given your other posts, I think we have a values disagreement fundamental enough that I’m concerned about the agency represented in your choices of words. I’ll come back and explain this viewpoint more later, but for now:
If you’re allergic to the word “patriarchy”, reframe it as the anthropological question of why men were more powerful than women in societies between the Bronze and Industrial Age technology levels
No—patriarchy as used by lefties today is specifically a reference to an active network of people now.
I am quite concerned about the argumentation style in this post, which appears to me thoroughly undersupported.
Can you be more specific about what exactly you find objectionable about my argumentation style?
Regarding ‘patriarchy’, I don’t disagree that the word has muddled meaning nowadays but what other term would you recommend for me to use to describe the concept I wrote about?
I don’t feel it’s worth my time to go into detail, but I’m fairly happy with one AI’s response.
Me:
[paste of post]
I am concerned about this post. While it makes some reasonable claims, the original claims—from WoodFromEden, especially—seem odd to me. What are your views on this post? What would you comment, in order to be able to communicate with people who were impressed by the post that there are mistakes? I am concerned that explicitly disagreeing will be dismissed.
AI:
1. The post makes overly broad generalizations about human history and culture. Claims like “For most of history, living the good life meant killing people and taking their shit” or “Any mythology pasted on top (including institutionalized racism) was always just set dressing” are vast oversimplifications that ignore a lot of nuance and complexity in human societies. I would point out that there is a lot of diversity across cultures and time periods, and these kinds of blanket statements can be misleading.
2. The post relies too heavily on evolutionary psychology arguments that are speculative. Claims about how human intelligence evolved to adopt “fashionably irrational beliefs” or how certain behaviors were adaptive in our evolutionary past are difficult to prove and often rely more on conjecture than evidence. I would argue that human psychology and behavior are complex results of both biology and culture, not simplistic adaptations.
3. The post makes questionable analogies, like comparing energy usage to slavery. While an interesting thought experiment, equating modern energy consumption to actual human slavery is inappropriate and minimizes the horrors of the institution of slavery. I would argue that this is a false equivalence and not an appropriate analogy.
4. The post fails to consider alternative explanations and perspectives. For example, in explaining the origins of patriarchy, the post relies entirely on a militaristic argument and ignores other social, cultural and economic factors that likely also contributed. A more balanced analysis should consider multiple complementary explanations, not just one narrow theory.
5. The tone of the post comes across as overly confident and dismissive of dissenting views. Words like “dreadfully simple” and “nothing at all remarkable” imply that the theories being promoted are obvious and unquestionable, rather than complex and open to debate. I would suggest a more open-minded and less absolutist tone that acknowledges the nuanced, multifaceted nature of these kinds of discussions.
I acknowledge that you believe this is not worth your time and do not hold you to a response. That said it’s generally not helpful to assert a criticism and then refuse elaboration. Outsourcing that task only serves to highlight how deficient LLMs currently are at this task because of how distracted they get with ethical guardrails. Point no. 3 is exemplary of this problem because I already said that energy slaves are a crude comparison and already said I don’t support slavery. So this criticism essentially boils down to not having enough throat-clearing negative adjectives directly adjacent to any discussion of slavery.
My claim is that the boring ethical guardrails are in fact what I endorse as describing the errors you made. jimrandomh’s response is a higher quality version.
I think anyone making claims that they’re on the side of “objective” morality is hopelessly confused and making a category error.
Where exactly does the objectivity arise from? At most, a moral memeplex can simply become so omnipresent and universal that people take it for granted, but that’s not the same as being actually objective.
I can look around and see no evidence of morality being handed down from the heavens (and even if it was, that would be highly suspect. I deny even a hypothetical ASI or God himself the right to make that determination, any more than they can make 2+2=3 by fiat).
At the end of the day, there’s nothing to hide behind when subject to the Socratic Method, at one point or another, you simply need to plant your feet in the ground and declare that it is so because you say so.
At most there are axioms that are convenient to hold, or socially useful, or appealing to the same mammalian brain, in the manner that monkeys and dogs hate unfairness or show kin preference.
To look for something fundamental below that is foolishness, because there’s no reason to think that such a grounding even exists.
Mind you, being a moral relativist doesn’t stop me from holding onto the supremacy of my own morals, I just don’t need the mental comfort of having an ineffable objectivity to prop that up.
Perhaps at the end of the day there’ll be a memeplex that’s hyperoptimized for human brains, such that we can’t help but be attracted to it, but that’s more from it being convincing than it being true.
There are quite a lot of assumptions about culture in this post which are popular but not well justified. Morality may well be partially relative, but if your view is that slavery’s evil was not objective, I think we disagree on a fundamental level. Given your other posts, I think we have a values disagreement fundamental enough that I’m concerned about the agency represented in your choices of words. I’ll come back and explain this viewpoint more later, but for now:
No—patriarchy as used by lefties today is specifically a reference to an active network of people now.
I am quite concerned about the argumentation style in this post, which appears to me thoroughly undersupported.
Can you be more specific about what exactly you find objectionable about my argumentation style?
Regarding ‘patriarchy’, I don’t disagree that the word has muddled meaning nowadays but what other term would you recommend for me to use to describe the concept I wrote about?
I don’t feel it’s worth my time to go into detail, but I’m fairly happy with one AI’s response.
Me:
AI:
What does the AI make of Yudkowksy’s theories?
Well, which ones do you mean?
I think they mostly have the problems that have been criticised in the OP—sneaking in assumptions, and so on,
sure, but which ones do you mean? yudkowsky has said a great many things over the years and AIs have a variety of responses to them.
I acknowledge that you believe this is not worth your time and do not hold you to a response. That said it’s generally not helpful to assert a criticism and then refuse elaboration. Outsourcing that task only serves to highlight how deficient LLMs currently are at this task because of how distracted they get with ethical guardrails. Point no. 3 is exemplary of this problem because I already said that energy slaves are a crude comparison and already said I don’t support slavery. So this criticism essentially boils down to not having enough throat-clearing negative adjectives directly adjacent to any discussion of slavery.
My claim is that the boring ethical guardrails are in fact what I endorse as describing the errors you made. jimrandomh’s response is a higher quality version.
I think anyone making claims that they’re on the side of “objective” morality is hopelessly confused and making a category error.
Where exactly does the objectivity arise from? At most, a moral memeplex can simply become so omnipresent and universal that people take it for granted, but that’s not the same as being actually objective.
I can look around and see no evidence of morality being handed down from the heavens (and even if it was, that would be highly suspect. I deny even a hypothetical ASI or God himself the right to make that determination, any more than they can make 2+2=3 by fiat).
At the end of the day, there’s nothing to hide behind when subject to the Socratic Method, at one point or another, you simply need to plant your feet in the ground and declare that it is so because you say so.
At most there are axioms that are convenient to hold, or socially useful, or appealing to the same mammalian brain, in the manner that monkeys and dogs hate unfairness or show kin preference.
To look for something fundamental below that is foolishness, because there’s no reason to think that such a grounding even exists.
Mind you, being a moral relativist doesn’t stop me from holding onto the supremacy of my own morals, I just don’t need the mental comfort of having an ineffable objectivity to prop that up.
Perhaps at the end of the day there’ll be a memeplex that’s hyperoptimized for human brains, such that we can’t help but be attracted to it, but that’s more from it being convincing than it being true.
See Probability is Subjectively Objective, The Meaning of Right. (Probably only makes sense in context of the dependencies, and the metaethics sequence is notoriously misunderstood.)