Moral progress and philosophy

https://forum.effectivealtruism.org/posts/YC3Mvw2xNtpKxR5sK/phd-on-moral-progress-bibliography-review

The gods are philosophers, not engineers: As we become more powerful a greater proportion of the cruxes should become ethical since the range of possible facts about what we should do expands in proportion to the amount of things we can do as per should implies can

!!!RIGID is just what rigourous (favourite self-description of analytics) actually means. Rigour means you can push down on one end and the other end goes up.

Philosophy lacks consensus on bottlenecks to progress, game trees, rigid hierarchies + decomp., thus sliw progress low convergence, fashions, dictate which areas are progressed on how do rigidity and a lack of liw hang8ng fruit relate, rigidity like info propagating far?

If secular ethics is so young, itd be an extraordinary coincidence if the easiest to formalisr theory (utilitarianism) was the right one, rather than the just the lowest hanging fruit. Consider historical analogies? Astronony or euclidean geometry? Early forms of logic?

Are we worse at psychology than at physics bc the human brain is optimised for psych but not physics so psych has skipped all the low hanging fruit but physics hasnt. Some evidnece for this is the fact that it took a long time for scientific methods to beat out common sense in psych but not in physics

Does parfits same mountain from 3 sides analogy geometrically support the idea that in bad worlds moral trade/cooperation is easier/more worthwhile?

For tradition to work as a source of moral guidance, the moral environment must be stable. Analogously, water must be still for the sediment to settle and leave it clear.

Check karnofskys responses to beethoven post for philosophical steering.

Summary: Types of Philosophical Progress

Key Definitions

Philosophy: Activity of Xunzi, Nagarjuna, Plato, Hegel, Kripke, Kant
Philosophical Problems: Not fixed historically (e.g., determinism moved from philosophy → physics → partly back)
Philosophical Progress: Decision-relevant work on current philosophical problems

Types of Progress

Scientific Breakaway
- Problems get precise formulation/techniques for scientific/mathematical testing
- Examples: causation, consciousness, intelligence, life, logic
- Original philosophical discussions may not be decision-relevant
Problem Dissolution
- Consensus that problems are decision-irrelevant
- Two subtypes: a. Incoherent Non-Problems
  - Examples: analyzing vague concepts like knowledge (internalist/externalist)
  - responsibility (attributability/accountability)
  - consciousness (phenomenal/access) b. Only Relevant Under Implausible Worldviews
  - Examples: medieval doctrinal debates
  - ethical views based on outdated natural teleology
Direct Consensus
- Two subtypes: a. Negative Consensus
  - Discarding solutions: simplistic analyses of free will, personal identity, knowledge
  - Rejecting theories: substance dualism, strong cultural relativism b. Positive Consensus
  - Examples: impermissibility of harming animals
  - compatibilism
  - secular moral realism
  - complexity of abortion ethics
Proto-Progress
- Extending existing judgments to new situations/technologies
- Examples:
  - Singer’s drowning child (regarding distance/aid)
  - Current debates:
    - personhood and split-brain subjects
    - wild animal suffering
    - gene-editing
    - preference-changing drugs
    - AI moral patienthood
    - population ethics

Important Caveats

Consensus Trustworthiness
- Must show difference from general population
- Not explained by selection effects
- Data example: moral realism correlation with normative ethics (62.5% vs 58.1%)
Untrustworthy Consensuses
- Philosophy demographics: WEIRD, male, elite university
- Some contrarian positions worth updating toward:
  - Examples: bestiality, infanticide, sincere universal anti-natalism
  - Various “weird” conclusions
  - modal realism
  - dialetheism
Special Cases
- TERF views in philosophy journals vs broader academia
- Publication bias toward controversial positions
- Need to control for theism in moral realism stats

Noted Gaps in Framework

Doesn’t capture psychological/evolutionary/historical debunking arguments
Unclear treatment of conceptual therapy (e.g., Wittgenstein, Quine, Dennett)
Question of whether “decision-relevant” is right criterion

Notes So philosophical systems need constraints, they need rough ground to push against. But they don’t have any source of strong constraints. So they need to push against a wide range of desiderata/topics to get anywhere. This encourages system building. Consider sellars on how things hang together. And maybe foundationalism too. So for philosophical systems to succeed/make progress they have to be large and rigid (where rigidity is a bidirectional transfer of force - so being large and rigid is the opposite of being flexible/modular). This means that large non modular systems of philosophy should not last as long as other ones. How accurate is this? Consider Spinoza, Wolff, scholastics, stoics?, Buddhist schools? Indian schools? Chinese schools as less rigid? Hegel and neohegelianism? How might systems interact with different systems of external incentives to determine how philosophical systems last. Modern academic vs monastic vs independent wealthy vs bureaucratic in China. E.g. publish or perish, having to compete with other systems or not, having to actively win converts - the public/private uncompetitive/competitive dichotomy is the analogy here. Young philosophers, old philosophers

A sub question is what counts as progress - see metaphilosophy essay

Another one is whether rigidity or the perception of rigidity is what matters

Make sure you give an objective account of rigidity so you don’t cherry pick who counts as rigid to confirm your thesis.

Read more of Hanson’s stuff on rotting systems

Are philosophical systems just slaves to their institutional hosts? No it doesn’t seem like it - you can ask why some systems fall with their institutions and others don’t.

What kind of system will produce more Q primes?

So the dilemma is between being large and rigid vs flexible or small or both.

This explains the greater tendency towards systematization in philosophy.

Connect with philosophy vs Q primes

‘the unreasonable ineffectiveness of philosophy’

Sequence vs cluster thinking is relevant.

Sociology of Philosophies: A Global Theory of Intellectual Change is very relevant but massive.

Allen’s the institutional revolution is maybe relevant.

Also remember to check Tom moynihans book on extinction.

Lucretius/newton/machiavelli’s anthropic arguments about extinction/length of history are all relevant to moynihan on extinction
When did the idea that the present and the past are the same kind of thing emerge? This lies behind things like evaluating statements about the present and past in basically the same way. Herodotus or Thucydides or Sima Qian maybe? Alternatively, when did people start thinking only in non-legendary time, wherein all figures and events have precise dates/properties, unlike characters in fiction.

Given SSC post the control group is out of control philosophy must be fucked. Or maybe not. If social sciences actually aren’t that much better than philosophy, maybe the correct verdict is just thinking that intellectual progress isn’t that hard, and mostly happens in a very wholesale way, with developments in the hard sciences mostly driving everything. This would line up ok with the early modern period.

Distinction between collaborative splintering (of topics) in science and competitive splintering of positions in philosophy, (and maybe ither humanities fields?)

Saudis/egyptians sponsoring favourable interpretations of scripture
Evangelical churches in US
Mormons with brigham young

Different reference classes of:
Startup ideologies: marxism, existentialism, etc
Now-established ideologies: US evangelicalism
State-sponsored ideologies: Turkey w/ armenian genocide, China
Fraudulent research: e.g. tobacco sponsored smoking studies

Does any of this mean EA shouodnt throw more money at early career researchers though?
Mamy examples of close personal networks leading to influence doesnt mean theyre necessary though, money may not have worked back then but it will now.

Check ea forum post of predictably updating and compare to the peer disagreement/agreement theorem re truth seeking in philosophy. How will it relate to Williamson on metaphilosophy, how does the CA vs CE/truth seeking vs definition making question affect it re philosophical progress. Robin Hanson on moral progress - he argues genuine progress should be a random walk, like any discovery process. Maybe see the less wrong or EA post about why predictable updating isn’t irrational or a signal of something other than truth seeking. Super interesting! More specifically, will fashion cycles be particularly strong evidence against moral progress? Integrate this with less wrong/forum posts on philosophical progress Plus your metaphilosophy paper. Wikipedia on aumanns theorem quite good

MERGING OF OPINIONS AND PROBABILITY KINEMATICS by huttegger is even better on Aumann and related issues Martingale and Aumann

How does eugenics fit in to the moral circle expansion narrative? What is the moral circle exactly and how can it overextend? Circle of care =/= moral circle, moral implies interventionist. Also all the different ideas around who has moral agency. Pedophilia etc. You could distinguish two circles, the circle of moral patients - which mostly expands, barring the dead, communities, etc. and the circle of moral agents, which expands a bit kinda then fluctuates randomly - e.g. hard determinism about people, no for animals, children and elderly, the dead, communities. These are all relevant for considerations of Justice/deontology

It seems like publish or perish incentives for controversy (defend what had a 10% chance of being true) makes philosophical disagreement look more radical than it is e.g. anti-natalism. This may discourage deferring towards any particular philosophical view too much. But at the same time it means people will overweight the treatment effects of philosophy this includes it’s ability to produce rational consensus. So philosophers are just disappointing elite liberals like other arts professors.

How can we justify the continued relevance of ancient philosophers but not ancient scientists? Pre institutional religion is a plus. The fact that the set of low hanging fruit is more constant for philosophy so the ancients got a greater proportion which compensates for the lesser precision. Maybe something else about the low fidelity and stuff fucl around with the quality of founder vs quality of successors ratio - consider kings, companies, etc. Is the proportion of philosophical value Plato found the same as the proportion of mathematical value Archimedes found?

Use of funnel plots to investigate publication bias in philosophy.

Paul Graham’s essays; Age of consent is a random walk roughly. (eh is it really, it seems to have gotten older for a number of reasons, including pregnancy and child rearing impinging on the ability to complete industrial schooling. Maybe a state of perpetual childhood/invention of adolescence too idk. Is adolescence neurologically distinct? The later end especially seems very blurry. puberty marks the start ok, but brain development blurs out until 25/30). Birth, child-rearing, sex and marriage, employment, and death are fertile grounds - it doesn’t seem like these show any trends really. taboos emerge where one group can dominate another, but only just, consider the dogmatism of the church in the counter-reformation vs the high middle ages. Short term drivers of taboos - what ideas are tainted by recent atrocity, (argumentum ad hitlerum) Powerful groups in danger of losing influence? What will soon be cool to assert? Does antisemitism/hitler being the paradigm of evil over Mao or Stalin count? What can’t you criticise - apart from kids with cancer. You can only see fashions as fashions from a distance, and by comparison - the same goes for moral fashions.

How long term are all these trends? Are they all just artefacts of modernity? So something to do with living standards/wealth. One way to see how much we really depend on Aristotle would be to diff European culture with Chinese: what ideas did European culture have in 1800 that Chinese culture didn’t, in virtue of Aristotle’s contribution? How significant is philosophy as a field of study turning away the brightest minds? c.a. Russell Well it’s false for undergraduates right? And great achievers in other fields certainly end up writing philosophy more often than picking up ornithology, but I suspect the two pasttimes are similarly common. cf. sprague camp aristotle and the gun.

Some arguments that it’s not all horseshit - you can’t Sokal hoax analytic philosophy - though even the Sokal hoax was greatly exaggerated. Do tenure committees have to agree with you to get tenure? idk

We should be very suspicious of process-based justifications of philosophy, ‘of sure it turns out it’s useless but the process is intrinsically valuable’. That sounds like giving an inch when you should concede a mile.

As 4 and 5 are incomplete goods, efforts to increase them must avoid decreasing the accuracy of the selector process, (by encouraging biased selection somehow? SJWism maybe? Trans rights stuff?) or reducing the throughput of the selector process, (jargon and technicality?). You should make some more predictions here about what will succeed.

Question of how relevant has this been for ethical progress as a change in ethical outcomes - who cares about the beliefs of people or decision makers. The argument for commonsense declining is that these types of philosophical progress will increasingly cause trajectory changes while previously they’ve only accelerated things if anything (e.g. banning slavery). What exactly is the link here?

Why does the Arthur meme apply in philosophy so much? Memes

Is philosophical impact heavy tailed.

Might philosophy ba harmed more by institutional pressures against interdisciplinary work than other fields? Maybe because of the lack of a defined subject area/lack of internal Q primes?

Qualify anti-natalism with sincere absolute (not climate change dependent) anti-natalism. Maybe add in not thinking privacy, free speech, etc. are intrinsic goods? Also historicist suspicion of romantic love, authenticity, privacy. Maybe virtue ethics generally, but you could say it’s just a different frame of virtue/vice language, (e.g. problematic, curious, etc. Vs classical lists) note that you’d need to control for virtue ethicists being disproportionately religious, so maybe just take proportion of philosophers that are secular VE and compare to left liberals or secular left-liberals generally. Also the issues of philosophers being ahead of the curve on SJW views to an extent - sure compared to non-elite academics, but definitely not for elite academics. (Might have to control for age here?) Also account for virtue theory being more neglected in philosophy than the public during 20th c. Generally consider age-lags with age of survey respondents and deaths of philosophers. Surprising academia is progressive relative to society? What about patrons? Pandering to the rich?

What’s the point of this/who cares: I’m trying to develop an inside view of the impact of philosophy as a career path for me - contrast 80k profile. Also a sub question for moral circle expansion, moral progress, value drift questions - could use this for ea critique comp.

What about the Wittgenstein view of philosophy as conceptual therapy. De re de dicto distinction may be an example of an output here, also Lukeprogs list is a mix of this kind of individualist progress - where it’ll be perpetual but maybe therapeutic techniques can improve - and the precision toolmaker/underlabourer of science view of philosophy. How do these more specific models fit in to my more general one?

Robin Hanson on moral progress - he argues genuine progress should be a random walk, like any discovery process. Maybe see the less wrong or EA post about why predictable updating isn’t irrational or a signal of something other than truth seeking. Super interesting! More specifically, will fashion cycles be particularly strong evidence against moral progress?

PG - As the world progresses, the number of things you can win at by getting the right answer increases

The other kind of progress is the I tellectual tool making progress - it should be obvious to you that by using some technique or concept in your work whether it is fruitful or not. Though there is some risk that tool which are especially fun or easy to us will be used disproportionately. How could you optimize production in philosophy if it really is tool making. You could have trade shows where you demonstrate the usefulness of the tools, you could advertise the tools in a targeted way to those who will get the most use out of them, you could consult with buyers as to what tools they especially need.

If philosophy is low rigidity, how should people cluster in view space differently? They should more arbitrary, or reflect historical factors/selection effects/fashions more.

Programmers, contract lawyers, maybe philosophers should be best at making things explicit. When you don’t make things explicit, how do they still go well? What are the language games we play to keep things on track?

Gricean pragmatics as philosophical progress

CRUCIAL - Buchanan A (2002) Social moral epistemology. Soc Philos Policy 19(2):126–152

Is there a philosophical file drawer? Sort of since null results won’t really get published. But what is a null result in philosophy? It’s kind of like no divergence from common sense. The good thing is philosophical controversialism makes ‘replications’ - really replies, a bit more incentived. But I think the fact you can only publish direct replies in the same journal still lowers the EV of writing replies.

Despite all that stuff about power extensions needing more ethics, it also seems like ethics should become more obvious and uncontroversial as power extends, just not totally fucking up becomes wonderful when holding a nuclear bomb, but not so wonderful when you’re a professor at Oxford.

On decision relevance and conceptual therapy - something is decision relevant if it’s truth/your beliefs about it give you reason to act differently, (check that definition). Eliminating confusions, and hence distractions, is this in an indirect way. It should change your credences in meta-questions like ‘is this a coherent question’/‘is this first order question decision-relevant’, and these are themselves decision relevant. In addition, it reduces confusion, which just directly lets your attention and effort move on to (generally) more helpful things, but you do need to be wary of scholastic intractable hierarchies of conceptual clarification, since they are attentional treadmills.

See Liam Kofi Bright’s page against peer review

Re theory of change are academic philosophers and majors ambitious enough to be worth converting? Is EA IQ or ambition or perspective constrained? Many facets of talent constraint.

Ian Morris on 80k on long term history

Hamming - You and Your research Why so many verbal/unsystematic - written/systematic pairs, (jesus, paul, socrates, plato, kongzi, mengzi) axial age thing? weird selection effect about them being the first to survive because of writing. Previous people like zoroaster couldnt get their ideas transmitted or complexified bc. no writing. And then eveyrone after them just wrote for some reason? Why weren’t there any more pairs in the middle ages pre-mass literacy though?

How do the great achievements of different fields differ with age? How do citations? This informs us of the metrics usefulness. How much is mean reversion? Why are the different fields different? Are the fields still different under different institutional settings? Is it crafty fields and sciency fields that’s the difference?

Aumanns agreement theorem and pragmatic conceptions of truth as the limit of rational inquiry. The two awkward things about the pragmatist/constrictivist conception of moral and abstract truths as the limit of ideal rational inquiry is (1) the euthyphro dilemma if it’s meant as an analysis (2) how small a part of mind space we humans occupy. This leads you to think that either (a) our current prima facie rational beliefs aren’t really guides to the truth at all, or (b) human rationality is just a small artefact of our part of mind space, so there’s many rationalities (MacIntyre)

A solution to 2 might be something like Aumanns agreement theorem, implying we could bootstrap ourselves into general rationality by agreeing with progressively less human-specific agents.

Need to come up with an explanation of why the causes of moral progress/general moral accuracy tend to roughly track the actual right-makers/good-makers. Something about reason/coordination/self interest/effectiveness of the virtuous in propagating their ideas. Idk.

Where is philosophy vs other kinds of intellectual work on the instrumental validity of being remembered as the unifying author of many insights. E.g. vs maths, history, science, art, etc. When and why are there firms in the marketplace of ideas? How does this relate to the epistemic benefits and costs of diversity? How does it vary across different issues? E.g. controversial vs non controversial, personal vs impersonal, niche vs mainstream? The efficient market of ideas

To what extent is the rise of interdisciplinary work a sign of diminishing marginal returns in soft fields. You’d think it’s not simply science aping to do interdisciplinary stuff, but the fact that you can’t make rigid robust inferences up and down from subtopics to topics that makes interdisciplinary work a necessity, since new grads can’t profitably work on sub-sub-sub topics as much as they can in science, so there’s less funding for them to. Idk if this is right, what does it predict about what topics are funded in philosophy and does it somehow assume the fields are rationally planned from above? What does it predict about which fields are engaged with in interdisciplinary work?

Intellectual movements; Comtean positivism, general semantics, libertarian economics, Rand’s objectivism, (maybe) American pragmatism, Georgism, eugenics, May Fourth/New Culture movement, futurists, the Encyclopedistes, FOSS, anti-nuclear, psychoanalysis, Frankfurt school, McLuhan/media studies, Edward Bernays and public relations, Founding and institutionalisation of various social sciences/disciplines including management studies, psychology, economics, and sociology, pan-Africanism, New Confucianism, (not Neo-Confucianism) fair trade movement? BLM? socialist movements? Tea Party? More non-western movements - esp. Indian. Definitely Quakers, as per SSC essay apparently source of modern idea of conscience/conscientious objection

Moral/philosophical change; UDHR, absorption of Buddhism into Chinese philosophy, Anti-slavery, animal rights, changes around age of consent/pedophilia norms - inc. 1977 petition, Renaissance-now changes in the penal system, women’s suffrage, attitudes towards infanticide/abortion, attitudes towards the dead, attitudes towards future generations, growth of nationalism, different lists of cardinal virtues across time and cultures.

Why was rape considered similarly bad in the past while slavery and genocide were not? This was true even though the current rationale of self ownership of our bodies didn’t apply, instead it was a modesty based justification
I think it’s because rape had more quickly diminishing marginal utility, especially for powerful entities like states.
Does this explain why rape by Gods was more acceptable? Does it explain why rapes in conquest of foreign nations was also more acceptable?

See A history of consent in western thought book chapter by Johnston.

We now group philosophers into schools like british empiricists/continental rationalists that didn’t exist as schools; during the 17th c. there was scholasticim, the new philosophy, mechanism/corpuscular philosophy, cartesianism, cambridge platonism, baconianism, scepticism. Later the leibniz/wolff philosophy. All these groups cited each other, agreed about method and substance, and saw themselves as having common enemies/a common project.

Can one apply super forecasting techniques to Phil if the Qs don’t resolve?

I thought the most modular disciplines, with the most levels of specialisation, were the ones with the lowest transfer of learning cost, or perhaps the highest rigidity? But if they have the lowest transfer of learning cost, then the great achievements in those fields should be made younger. One reason this trend may not hold is that figures in empirical disciplines need time to build up the career capital to secure the funding needed for their research. See philosophy and waste-paper basket joke.

Impact and the economics of research What mode of knowledge production, private patronage, public patronage, freelancing, or universities, (either western or Buddhist or Confucian) tend to produce the most impactful research. This can be broken down into truth-conduciveness, targeting under-theorised topics with high marginal utility, amount of high quality research done, and the right kind of spread of knowledge beyond the institution or actor into shaping the actions of other institutions and actors.

https://www.lesswrong.com/posts/B7P97C27rvHPz3s9B/gears-in-understanding The idea of a gears level understanding basically encompasses and expands my ideas of rigidity I believe. Rigid, simple, wrong

Monothematica

Explorer