Source: Facebook, Quora, Twitter, blogs
Date: 2022
(see too: 1 : 2 : 3 : 4 : 5 : 6 : 7 : 8 : 9 : 10 : 11 : 12 : 13 : 14)

paradise engineering

Social Media
Unsorted Postings
on
paradise engineering, the biohappiness revolution, transhumanism,
philosophy, quantum mechanics, effective altruism, utilitarianism,
aging, superintelligence, suffering, happiness, consciousness...

JANUARY 2022 -

[on paradise engineering]
“The future belongs to those who believe in the beauty of their dreams.”
(Eleanor Roosevelt)
Philosophy podcast
Alternatively, the future belongs to those who believe Darwinian life is the stuff of nightmares. Either way, the problem of suffering is fixable...
The End of Suffering
(Philosophists podcast) & mp3
("David Pearce on abolishing suffering using biotech")

HI in the generic sense is just replacing the biology of involuntary suffering with life animated entirely by gradients of bliss. There are both (relatively!) conservative and revolutionary ways to do this. Compassionate conservation (https://www.hedweb.com/social-media/paradise.pdf) is probably the most politically saleable way. We can't spike all guns; but I ask critics: would you like to wake up tomorrow morning in an extremely good mood but with your values and preferences otherwise intact?
Declan, the Overton window can sometimes shift fairly quickly. A biohappiness revolution still sounds utopian. But the technical tools are now ready - if humanity is willing to use them.
Here's a short interview forthcoming in Brightly magazine:
Transhumanism and the End of Suffering

[on psychedelia]
"Psychedelics are illegal not because a loving government is concerned that you may jump out of a third story window. Psychedelics are illegal because they dissolve opinion structures and culturally laid down models of behaviour and information processing. They open you up to the possibility that everything you know is wrong.”
(Terence McKenna)
The future of psychedelic medicine will be drugs you've never heard of
Maybe. But opening up the possibility that everything you know is wrong is not a reliable recipe for mental health.
IMO, it's impossible to overstate the intellectual significance of psychedelics:
Psychedelics and epistemic rationality
But their therapeutic role is limited and unproven.
Perhaps I should add that safe and sustainable analogues of MDMA would inaugurate a revolution in mental health and human civilization. But MDMA isn't a psychedelic in the normal sense of the term:
Utopian Pharmacology

A very nice analysis - though it may induce the illusion of understanding psychedelia in the drug-naïve:
Psychedelic insight
("The Insights Psychedelics Give You Aren’t Always True")
quite a party

[on the measurement problem in QM]
Quantum mechanics scares me. Does the superposition principle ever break down?
The measurement problem
If so, then I've no idea how. If it does, then my conception of mind and reality implodes: good.

[on the neuronal correlates of consciousness]
Most researchers are focused on the Hard Problem.
But IF consciousness discloses the intrinsic nature of the physical - i.e. bosonic and fermionic fields alike - then the big question is how certain fields are sometimes bound into dynamically stable subjects of experience like you or me.

What is consciousness? (Nature magazine)
“All available evidence implicates neocortical tissue in generating feelings.”
Alternatively, our most intense feelings originate in the brainstem - which is evolutionarily ancient:
Pain in the brainstem
But more generally, talk of the "neuronal correlates of consciousness" (NCC) is problematic. One risks slipping into perceptual direct realism. For you can't directly observe nervous tissue and correlate its states with conscious experiences. Rather, you correlate one aspect of your experience, e.g. of locally-exposed neural tissue in a neurosurgeon's operating theatre, with another aspect of your experience, say, the patient's self-reports in response to microelectrode stimulation.
The physical world is inferred, not observed.
For what it's worth, I think lumps of cheesy wet nervous wet tissue and decohered classical neurons are artefacts of a false theory of consciousness and perception. Textbook neuroscience is wrong.
But I remain a physicalist.

[on psychoactive food]
microdosing
You are what you eat, so eat wisely...
On Psychoactive Food
("A Neglected Link Between the Psychoactive Effects of Dietary Ingredients and Consciousness-Altering Drugs")

[on war]
I fear (and tentatively predict) nuclear war this century. Can it be prevented?
Here are two options, not mutually exclusive.
(1) All-female governance would probably prevent Armageddon. Wars of territorial aggression aren't part of the behavioural phenotype of female primates.
(2) A democratically elected world government with a monopoly on the use of force would probably prevent Armageddon too. For sure, existing political leaders aren't going to cede power. But if we enact legislation that kicks in a few decades hence, vested interests wouldn't get trampled.
Transhumanist Party Panel on Reducing Existential Risk from the Russia-Ukraine Conflict
Will Dave's new 10 Point Plan For World Peace be adopted?
Almost certainly not.
Even so, world-government is eventually likely. But in common with the League of Nations and the United Nations, its establishment will plausibly come in the aftermath of a cataclysmic war.

"The breakup of the Soviet Union was the greatest geopolitical tragedy of the 20th century."
(Vladimir Putin)
Scott, rumours and disinformation swirl. I'm hesitant to add to the noise, but Putin's mental health is clearly an issue. For example, if he has Parkinson's disease (I hadn't heard that rumour) aka dopamine-deficiency disorder, Putin will presumably be being treated with dopamine-boosting drugs - which have neuropsychiatric effects.
In Sickness and in Power
Is Putin Sick?
[Vladimir Putin is no Adolf Hitler, but Dr Morell's treatment of Hitler influenced the course of the second half of WW2:
Substances administered to Hitler]

If humans migrate to life in the metaverse - mundane bodily functions aside - then territorial wars of aggression can be be virtualised and defanged. But ultimate power resides in basement reality. So does selection pressure.
Z
("What is the ‘Z’, the pro-war symbol sweeping Russia? The white letter, dubbed the ‘Zwastika’, is being displayed in support of aggressive military policy – but what does it mean?")

The Future of Nuclear War. The Dangerous Future of the Nuclear War: Superbombs, Cheap Nukes and Geophysical Attacks”
by Alexey Turchin.
The death spasms of Darwinian life will be ugly. But how ugly?

[on genome reform]
Science, Technology and the Future Conference
What's harder: (1) the technical details of reprogramming the biosphere or (2) shifting the Overton window in favour of genome reform?
Lion and Man
The End of Suffering & Youtube PDF.
Genome Reform and the Future of Sentience
by David Pearce
ABSTRACT
No sentient being in the evolutionary history of life has enjoyed good health as defined by the World Health Organization. The founding constitution of the World Health Organization commits the international community to a daringly ambitious conception of health: "a state of complete physical, mental and social wellbeing". Health as so conceived is inconsistent with evolution via natural selection. Lifelong good health is inconsistent with a Darwinian genome. Indeed, the vision of the World Health Organization evokes the World Transhumanist Association. Transhumanists aspire to a civilization of superhappiness, superlongevity and superintelligence; but even an architecture of mind based on information-sensitive gradients of bliss cannot yield complete well-being. Post-Darwinian life will be sublime, but “complete” well-being is posthuman – more akin to Buddhist nirvana. So the aim of this talk is twofold. First, I shall explore the therapeutic interventions needed to underwrite the WHO conception of good health for everyone – or rather, a recognisable approximation of lifelong good health. What genes, allelic combinations and metabolic pathways must be targeted to deliver a biohappiness revolution: life based entirely on gradients of well-being? How can we devise a more civilized signalling system for human and nonhuman animal life than gradients of mental and physical pain? Secondly, how can genome reformists shift the Overton window of political discourse in favour of hedonic uplift? How can prospective parents worldwide – and the World Health Organization - be encouraged to embrace genome reform? For only germline engineering can fix the problem of suffering and create a happy biosphere for all sentient beings.

Even the most radical genetic interventions are therapeutic rather than enhancement by the lights of the WHO definition of health - to which all members of the UN are committed:
Gene editing to turn off pain
("The CRISPR-Cas9 gene editing tool could be used to "turn off" pain directly, raising ethical questions for society")
Just don't mention the "e" word.

Filed under "children and education" podcasts. Bland is best?
Manipulando Nuestros Genes
("David Pearce: manipulando nuestros genes erradicaremos el sufrimiento [Ingles]")

[on longevity]
Alternatively, get a (silicon?) body transplant:
Old-Age Record Could Reach 130 by Century’s End
("Analysis of supercentenarians suggests human lifespan may have no limit")
In my view, senescence of the mind/brain will be the biggest stumbling-block to indefinite lifespans. Limbic system upgrades are feasible:
Stem cells in Parkinson’s disease
But science doesn’t know how to sustain a perpetually youthful neocortex.

[on intelligence]
Take Fields medalists. If "IQ" were a proxy for general intelligence, then being a world-class mathematician would at least be strongly correlated with superior social cognition, introspective prowess, practical acumen, superior dating and mating skills, co-operative problem-solving ability, ability to navigate multiple state-spaces of qualia (etc) and other aspects of general intelligence. This doesn't appear to be the case. Any serious measure of general intelligence needs to measure the calibre of an entire mind, including the subject's phenomenal world-simulation.

Melody, an extremely empathetic intelligence might spend their life doting on and understanding the perspectives of their cat. A hyper-systematising intelligence might devise blueprints for reprogramming the global ecosystem to reconcile the interest of all cats, all mice, and all sentient beings in our forward light-cone.
Some people are able to switch cognitive style fairly easily; others struggle.

[on speciesism and cuteness]
“It is amazing how complete is the delusion that beauty is goodness.”
(Leo Tolstoy, The Kreutzer Sonata)
Beauty speciesism...
On beauty speciesism Who was the smartest person in the world?
("The smartest person in the world was Isaac Newton, a true polymath whose brilliance never has been, nor ever will be, surpassed.")
Maybe! Alternatively, someone who believed his greatest achievement is his interpretation of the Book of Daniel is a crank. Newton also believed that God sometimes needs to intervene to correct "irregularities" that arise in the motion of heavenly bodies – which is clearly nuts. Newton also wrote a million words on alchemy. I've no idea who is the smartest person in the world today. Is it possible that among the outpourings of one today’s cranks there are insights posterity will recognize as equal to Newton's laws of motion and (not quite!) universal gravitation?
We can only guess...

Caucasians are disproportionately vulnerable to autism spectrum disorder (ASD). ASD is associated with profound cognitive handicaps and also, unsurprisingly, higher mind-blind “IQ”.
IQ in Autism Spectrum Disorder
("A Population-Based Birth Cohort Study")
What is the optimal AQ for human civilisation?
I don’t know. But in an era of WMD, the “extreme male brain” theory of ASD suggests we should be cautious about ramping up AQ/“IQ’” without thinking through the societal implications.

Over half of people with ASD have above-average "IQ" scores, compared to a minority of neurotypicals. To stress, I think a lot of high AQ/"IQ" folk are cool! But it's easy to conflate cognitive style - both personal and tribal- with the essence of general intelligence.

[on antinatalism and selection pressure]
I missed this video - 10 years old now:
A Better Way (than Antinatalism)
(with thanks to Algernon) Abishek R, We fundamentally agree - Darwinian life on Earth is evil. But what's the solution? Staying child-free simply intensifies selection pressure against (any predisposition to) antinatalism. Instead, we need selection pressure in favour of life based on gradients of bliss. And such selection pressure will be exerted only in the wake of a reproductive revolution of designer babies:
What are the arguments against antinatalism? [on cryothanasia]
Should cryonics be opt-out and cryothanasia opt-in?
Death Defanged
At what age (if any) would you choose to be suspended? Marcus, I'd be all in favour of humans emulating the antechinus (cf. https://www.nationalgeographic.com/science/article/why-a-little-mammal-has-so-much-sex-that-it-disintegrates). But most nonagenarians would struggle. On a less energetic note, intravenous heroin is sublime (cf. "I'll die young, but it's like kissing God." - Lenny Bruce). Just avoid practical research earlier in life ("Don't try heroin - it's too good" - anon).
See too:
What is the most hedonistic pleasure

[on knowledge and the virtues of ignorance]
Premature escapism could be ethically disastrous. In "High-tech Jainism", I give the example of an advanced civilization elsewhere in the Galaxy who phase out suffering on their home planet in favour of sublime bliss, but mistakenly assume (their version of) Rare Earthism. If they'd instead pursued an arduous path of spacefaring rather than, say, creating immersive VR fantasy paradises, they could have discovered and rescued Darwinian life on Earth. OK, I'm a Rare Earther myself, tentatively at any rate. But we'll need to make absolutely sure all our ethical duties in basement reality have been been discharged before going to live in the Metaverse (etc).
Personally, I look forward to ignorance of reality - which (presumably) will come soon enough.

[on AGI]
MIRI announces new "Death With Dignity" strategy
The upshot of AGI as conceived by Eliezer aligns with my Buddhist values. Alas, I'm not remotely optimistic. How does a notional zombie AGI "understand" humans who spend their lives investigating the nature, causal efficacy, binding and diverse varieties of consciousness? How does a notional zombie AGI "understand" humans who are trying to solve the problem of suffering? If consciousness were a trivial implementation detail of multicellular animals, then the ignorance of programmable digital zombies wouldn't matter for the purposes of building AGI. In reality, phenomenally-bound consciousness has been the computational-functional key to the evolutionary success of the animal kingdom over the past c.540 million years. Our ability to run phenomenally-bound cross-modally-matched world-simulations in real time is insanely adaptive - as illustrated by rare neurological syndromes where binding partly breaks down. Classical Turing machines can't solve the binding problem. Their ignorance is architecturally hardwired. Tim, one wonders why Gautama Buddha didn't urge his followers to stay child-free (OK, I know little of contraceptive practices in ancient India). Or did Buddha glimpse something akin to the contemporary selection pressure argument against antinatalism? Either way, AGI as conceived by Eliezer is the road to nirvana.

Kenneth, “Can it be attributed singular agency or not?" What exactly is the sinister "it" we're talking about? Digital zombies aren’t agents with a unified phenomenal self. To pose a takeover threat to minds in basement reality, the software must presumably have its functional analogue, together with the zombie counterpart of a phenomenally-bound cross-modally matched real-time world-simulation (“perception”). OK, maybe my AGI scepticism is a failure of imagination on my part. But I don't understand how mankind can program/train up software for incredibly adaptive abilities neuroscience doesn't understand in humans. Thus neuroscience doesn't know why we all don't have e.g. integrative agnosia, akinetopsia, simultanagnosia, schizophrenia (etc) – why we aren’t helpless micro-experiential zombies. I won’t do my quantum mind spiel here, but the unity of perception and the unity of the self are absurdly fitness-enhancing for human and nonhuman animal minds alike - as rare partial deficit syndromes illustrate.
Classical Turing machines have no inkling what they lack. Humans aren't (yet?) smart enough to program workarounds - if comprehensive workarounds exist, which remains to be shown.
But my inner Buddhist hopes EY is right.

Kenneth I agree: "...there isn't any insight that could be cleaned from this neuroscientific program that couldn't transfer over to improving the performance of different kinds of AI".
But what are the upper bounds of zombie intelligence? Are we talking about the risks of artificial general intelligence or the future of computer malware? This isn’t a trivial semantic point. Just as you can't be a general intelligence and fail to grasp, say, the second law of thermodynamics, likewise you can't be a general intelligence if you're a digital zombie - functionally incapable of investigating the countless varieties, phenomenal binding and causal efficacy of consciousness. What does advanced zombie AI suppose folk like e.g. QRI are doing?! For evolutionary reasons, I think building sentience-friendly biological intelligence is a bigger challenge than sentience-friendly AI. And there's a terrible irony at work here. The raison d’être of MIRI is to warn us of the threat of runaway sentience-unfriendly software-based AI. But MIRI embody precisely the sentience-unfriendly intelligence they warn us against. (“The AI does not hate you, nor does it love you, but you are made out of atoms which it can use for something else.”) Not least, EY is a practising “narrowly” intelligent paperclipper – oblivious of the sentience of humble minds from pigs to chickens:
EY on chickens

It's worth distinguishing intrinsic and extrinsic intentionality ("aboutness"). Both are deeply mysterious. So is the relationship between them. Assuming physicalism, how can any physical state, conscious or otherwise, really be "about" another physical state? Naturalising meaning and reference is desperately hard:
The symbol grounding problem
However, most relevantly here, programmable digital zombies and sentient humans alike are capable of something functionally like extrinsic intentionality. So could our machines ever pose a threat to us? Naively, our subjective intentionality is functionally irrelevant - of no more significance than the incidental cognitive phenomenology of Kasparov playing Deep Blue at chess. Only Kasparov understands he’s playing chess, but so what? Likewise, why does it matter what Searle subjectively "understands"? But this dismissal can’t be right. For humans spend a lot of time physically discussing, investigating and modulating the subjective properties of our own internal states - from agony and ecstasy to subtle nuances of feeling and understanding - both our own consciousness and the consciousness of other sentient beings. Classical Turing machines can’t do that: it’s not even all "dark inside". Varieties of phenomenally-bound conscious experience are all that animal minds like ours ever directly know - it's literally the empirical evidence - and digital zombies have no idea what I'm talking about. Invincible ignorance is not general intelligence:
How is the brain like a computer?
IMO, zombie AI is an awesome tool, and a zombie putsch is sci-fi.

Classical Turing machines don't understand anything. Digital zombies don't need to understand anything to act in many ways we would call intelligent. The whole AI revolution has been marked by the progressive divorce of consciousness - including the phenomenology of cognition - from intelligence. But like saying a blind person wearing a spectrometer can see colour, we risk a fallacy of equivocation. Sometimes the equivocation is harmless and the parallel suggestive. At other times, like the claim a classical digital computer or connectionist system could "understand" suffering, it's insidious.

Tech pioneer warns of alien invasion
On the one hand, AI alarmists stress that "AGI" will be incomprehensibly alien intelligence. On the other hand, this hypothetical AGI is supposed to compete exactly like rival tribes of humans for dominance, scarce resources, or whatever. Maybe part of the worry stems from how male computer programmers are designing AI to beat humans at zero-sum adversarial games (Chess, Go) etc. If computer programming were a female-dominated profession, maybe ever more sophisticated robo-carers for the old would be gaining attention instead.

Jason, yes, bioengineering sentient beings with a pleasure-pain axis should take precedence. Recent breakthroughs in AI (AlphaFold, DALL·E 2 etc) are awesome. But "AGI" is not going to happen with classical Turing machines or connectionist systems; they can't solve the binding problem. So I don't worry about a zombie apocalypse - just the daunting challenge of building sentience-friendly biological intelligence.

[on herbivorising predators]
David Pearce interviwed on Herbivorizing Predators
DP Interview (mp3)
A non-violent biosphere is technically feasible. Sentient beings shouldn't harm each other. Should we focus on winning hearts and minds or technical blueprints?
[Most people, and indeed most ecologists, assume that a living world full of starvation and predation is as inevitable as the second law of thermodynamics.]  

I misjudged the pace of progress. I never thought wild animal suffering would seriously be discussed in my lifetime. But unlike pleas to quit meat and animal products, calls for compassionate stewardship of Nature don't ask people to undergo the slightest personal inconvenience. So the case for intervention can be discussed entirely on its merits.
Should we intervene to help wild animals?

Video on high-tech Jainism in French...
A pan-species welfare state
La révolution biotech pour reprogrammer le monde vivant
Some of the comments deserve answers, but my Queen's English probably wouldn't go down well with French critics.
See too:
Augmenter le bien être via les biotechnologies & mp4
(avec David Pearce)
DP in French (Wikkpedia)

Paywalled, alas....
Why We Should Not Let Mother Nature Run Its Course. Better Never to Have Been in the Wild
("A Case for Weak Wildlife Antinatalism by Ludwig Raal")

[on pain]
Our goal should be nociceotion without pain
Stupid Headline
("Robots could soon feel pain: Scientists develop artificial skin that can mimic uncomfortable sensations")

Neocortical chauvinism is ethically catastrophic:
Might pain be experienced in the brainstem rather than in the cerebral cortex? by Mark Baron & Marshall Devor
Abstract
"It is nearly axiomatic that pain, among other examples of conscious experience, is an outcome of still-uncertain forms of neural processing that occur in the cerebral cortex, and specifically within thalamo-cortical networks. This belief rests largely on the dramatic relative expansion of the cortex in the course of primate evolution, in humans in particular, and on the fact that direct activation of sensory representations in the cortex evokes a corresponding conscious percept. Here we assemble evidence, drawn from a number of sources, suggesting that pain experience is unlike the other senses and may not, in fact, be an expression of cortical processing. These include the virtual inability to evoke pain by cortical stimulation, the rarity of painful auras in epileptic patients and outcomes of cortical lesions. And yet, pain perception is clearly a function of a conscious brain. Indeed, it is perhaps the most archetypical example of conscious experience. This draws us to conclude that conscious experience, at least as realized in the pain system, is seated subcortically, perhaps even in the “primitive” brainstem. Our conjecture is that the massive expansion of the cortex over the course of evolution was not driven by the adaptive value of implementing consciousness. Rather, the cortex evolved because of the adaptive value of providing an already existing subcortical generator of consciousness with a feed of critical information that requires the computationally intensive capability of the cerebral cortex."

"a real-life mutant with the potential to heal humanity's pain"
A Real-Life Mutant
("Meet The Woman Who Feels No Pain")
What's more, in a decade or two, perhaps you could have Jo's FAAH and FAAH-OUT genes too Gene Therapy

"Pain is an opinion”
(Alan Gordon)
An opinion that should be suppressed:
Is the Pain All in my Head?
("A new treatment called pain-reprocessing therapy promises to cure chronic pain. But maybe not for everyone.")

The evolutionary origin of the pleasure-pain axis is unknown. Yet it's indeed possible that a bacterium can experience a micro-pinprick of distress. But IMO a bacterium can't suffer; and no number of discrete bacterial micro-pinpricks can create suffering. So if a toddler or a dog or a pig is in severe pain from, say, bacterial meningitis, then we are entitled to use antibiotics to treat the victim even if it's possible that millions of discrete micro-pinpricks will result. Devising a metric for tradeoffs between different intensities of unpleasantness when irreconcilable conflicts of interest arise is hard because "more is different" - qualitatively different.

There are a few genetic outliers today who view life almost entirely in terms of gradients of well-being. For instance, I've known meat eaters who defend factory-farming on the grounds that even such a grotesque life is minimally worth living. (I should stress other hyperthymics are more compassionate.) Conversely, a larger minority of depressives can imagine pleasure only in terms of a diminution of pain.
Actually, pleasure and pain are equally real - Nature just plays with the dial setting of hedonic tone - but their equal reality doesn't entail an equivalent moral symmetry.
Anyone who doubts pleasure is real should try mainlining heroin - though not unless on their deathbed.

"an elegant Jeremiad against 'algophobia', the fear of pain that now occupies our souls"
The Power of Pain
Only the quoted transhumanist algophobe makes sense IMO.

Roeland, I want to agree with you wholeheartedly. Like you I'm sure, I have a long list of sociopolitical and economic reforms I'd like to see to make the world a better place. But from the horrors of the "food chain" in Nature to the unending zero-sum status games of Darwinian life, IMO there is ONLY one long-term solution to the problem that can work: genome reform.
So you could say I'm an EA longtermist.
What We Owe the Future

Low-dose naltrexone
Chronic pain is astonishingly common:
Chronic pain in the UK
Naltrexone has no “abuse potential”. It could be made widely available at pharmacies. Alas, naltrexone is no panacea. What we really need is an international debate on the future of the SCN9A gene (“the volume knob for pain”) - both for ourselves (gene therapy) and for future generations. Making benign “low pain” alleles of SCN9A ubiquitous could essentially solve the problem of physical suffering in both human and nonhuman animals.

Do we need pain?
("Is suffering necessary for understanding")
Life on Earth needs a more civilised signalling system. A Darwinian pleasure-pain axis is cruel and barbaric. Artificial intelligence increasingly outperforms humans at many cognitive tasks. The "raw feels" pain are essential neither to nociception (cf. silicon robots) nor to great art (cf. https://openai.com/dall-e-2/). I know of no technical reason why we can't phase out experience below hedonic zero and create life based on gradients of bliss.
Will our successors truly understand Darwinian life if they can't suffer?
No, quite possibly not - just as we can't understand superhuman bliss.
Perhaps the ethical biggest risk of getting rid of suffering altogether is being too greedy to early. If (like me) you think our overriding moral obligation is to mitigate, prevent and eventually abolish suffering, then we need to make sure we understand the upper bounds to intelligent agency in the cosmos. Maybe our ethical duties will have been discharged when we have eradicated suffering in our solar system. So long as sub-zero experience can't reoccur in our forward light-cone, ethics in the traditional sense is redundant. If this is so, then we can explore paradise and forget Darwinian life like a bad dream. But we need to be sure.
The imperative to abolish suffering

[on moral enhancement technology]
Moral enhancement technology would be great if we agreed on what to enhance:
Moral enhancement technology By contrast, everyone likes being happier, albeit not always under that description.

[on AI]

Fascinating debate...
Pinker vs Aaronson
As far as I can tell, biological mind-brains are special. A supposed "whole-brain emulation" of Einstein or Shakespeare would be nothing of the kind, just an invincibly ignorant zombie. Phenomenal binding is insanely computationally powerful - as rare deficit syndromes illustrate - and there is no evidence that classical Turing machines or classically parallel connectionist systems can support phenomenal binding on pain of magical "strong" emergence. So "AGI" is a pipedream. Digital zombies have no insight into what they lack - nor even into what I'm talking about. Critically, what philosophers call the unity of consciousness and unity of perception isn't some trivial implementation detail of biological minds, but instead the ultra-functional key to our evolutionary success. Our real-time virtual world-simulations ("perception") are vastly fitness-enhancing. How biological minds do what's classically impossible is controversial: IMO, decohered classical neurons in four-dimensional space-time are just an artifact of our crude tools of neuroscanning. But either way, the quantum supremacy of biological minds leaves zombie AI for dust. That said, the upper bounds to zombie intelligence are unknown.

How is this zombie AI takeover supposed to work - technically, politically and sociologically? OK, I'm bemused. The real challenge is building sentience-friendly human intelligence. Humans currently abuse and kill 70 billion sentient beings each year in the death factories. We asphyxiate over trillion sentient beings from the sea. Now we're teetering on the brink of nuclear war. I wish EAs could focus on fixing the problem of suffering rather than getting diverted by the spectre of a zombie coup.

"DeepMind" isn't deep and hasn't a mind. Its algorithms (AlphaFold, AlphaCode, etc) are still a useful tool. With implanted neurochips, sentient beings will shortly be able to program, expertly play games, decipher protein structures and do everything digital zombies can do - and more! What digital zombies can't do is research the existence, diversity, binding and causal efficacy of conscious experience, i.e. what some of us spend our lives investigating, thinking about, and exploring.
Is zombie AI dangerous to sentience?
Potentially, yes, but not nearly as dangerous as the real threat to sentience, i.e. male human primates.
Anil Seth? I enjoy his work. But Anil doesn't explain why we aren't zombies (the Hard Problem) or micro-experiential zombies (the binding problem).

Nikolai, "understanding" has multiple senses. One is (subtly) phenomenological, the other is purely functional and behavioural. The question here is whether there are some functional behaviours where a cognitive phenomenology of understanding is functionally vital, or whether instead a classical Turing machine can do everything a sentient being can do without being a subject of experience. Could a digital zombie have a complete functional-behavioural understanding headaches in sentient beings without being able to experience pain or phenomenally-bound experience of any kind? For a start, humans talk about the subjective properties of their consciousness in a way divorced from any other functional role those phenomenal properties may (or may not) be playing...

I wish we'd access to GPT-10 on paradise engineering:
AI generated marketing content

Sean, but StarCraft III (etc) will fall to AI. What interests me is the class of problems that are too difficult, even in principle, for zombie AI. We don't know how rigorously to delimit this class yet. Classical Turing machines won't ever be able to investigate the existence, varieties, binding and causal efficacy of conscious experience - except (presumably) by coding for conscious beings like (trans)humans who can do so. But if we could flash forward a century and see what programmable digital computers / connectionist systems can do, then I suspect we'd both be shocked - though embedded neurochips will complicate this dichotomy.
That said, classical computers are idiots savants, not general intelligences - though from another perspective, so are you and I.

I was darkly amused recently when an AI doomster accused me of secretly believing in AI takeover and dismissing existential risk only because my negative utilitarian values welcomed our imminent conversion into the equivalent of paperclips: Machiavellian intelligence! (I wish) More seriously, ultimate power belongs to whoever controls basement reality - not with software at different levels of computational abstraction. The ignorance of classical Turing machines is architecturally hardwired. What makes animal minds special is their phenomenal binding into virtual worlds of experience – commonly misnamed “perception". Phenomenally-bound world-simulations are insanely computationally powerful, as the Cambrian Explosion attests. By contrast, classical Turing machines are awesome toys, but a programmable / trainable digital zombie has (literally) no idea what I'm even talking about. And digital zombies aren't going to persuade humanity to convert its productive infrastructure into paperclip (etc) factories – not even with the help of a few willing NU collaborators!
The Intelligence Explosion

Classical Turing machines and classical parallel connectionist systems are zombies:
Google engineer claims robots are sentient
Non-biological quantum computers may be sentient; but they lack a pleasure-pain axis
So IMO they don't inherently matter.

"Understanding consciousness and actually being conscious are also two logically distinct concepts. And based on Gödel-style arguments, we might even expect them to be negatively correlated. (Strictly speaking, it would be belief in one’s consciousness that is negatively correlated with understanding consciousness.)"

The only way to understand a state of consciousness, for example pain, is to instantiate it. Awake of dreaming, consciousness is all you ever directly know: it's the entirety of your phenomenal world-simulation. So our paradigm case of consciousness shouldn't be logico-linguistic thinking or the allegedly non-computable ability of human mathematical minds like Roger Penrose to divine the truth of Gödel sentences. Barring spooky "strong" emergence, classical Turing machines are zombies. Even if consciousness is fundamental to the world, their ignorance of sentience is architecturally hardwired. Fancifully, replace the 1s and 0s of a program like LaMDA (or an alleged digital proto-superintelligence) with micropixels of experience. Run the code. Speed of execution or sophistication of programming or training algorithm make no difference. The upshot isn't a subject of experience, but rather a microexperiential zombie with no more insight into the nature of human minds than a rock. Classical Turing machines are the wrong sort of architecture to support sentience - or general intelligence. And their ignorance has profound computational-functional consequences...

You remark
"As for what you say about “phenomenal binding” I haven’t read much about that notion but my impression is that involves a fundamental misunderstanding of consciousness in that it tries to treat a pure subject, consciousness, as if it were an object."
Understanding phenomenal binding is critical to understanding what consciousness is evolutionarily "for" - and why classical computers and connectionist systems can't do it, at least on pain of magic (cf. binding-problem.com). Imagine on the African savannah if you had the rare neurological syndrome of integrative agnosia and you could see only a mane, claws and teeth but no lion. Or imagine if you have simultanagnosia and can see only one lion and not the whole pride. Or imagine if you have akinetopsia ("motion blindness") and can't see the hungry pride moving towards you. Imagine if you were at most just 86 billion discrete, decohered, membrane-bound neuronal "pixels" of experience, as you are when dreamlessly asleep. Micro-experiential zombies tend to starve or get eaten….

"One thing I have noticed in strong AI debates is that people have trouble following their own premise to its ultimate conclusion"
I promise I’ve no trouble following premises to their ultimate conclusion! Whether you’ll want to go there is another matter. As they say, Nature is the best innovator. We are quantum minds running classical world-simulations. There’s no evidence that the superposition principle of QM breaks down inside the skull. On this story, we’ve been quantum computers – but not universal quantum computers! - long before Democritus, probably since the late Precambrian. Critically, this is an empirical question to be settled by the normal methods of science, i.e. interferometry. By the same token, anyone who claims that classical Turing machines can support phenomenally-bound sentience needs to explain how – and devise some experimental test to (dis)prove it.

[on outlawing slaughterhouses]
Slaughterhouses should shut now. Without them, the apparatus of exploitation would collapse. Some polls are encouraging (cf. Outlawing Slaughterhouses). But the proposal is unrealistic. Yet what if legislation to ban slaughterhouses could be enacted that kicks in, say, in 2035? Immense commercial incentives would be created to accelerate the development and commercialization of cruelty-free cultured meat and animal products. Consumers can carry on as before - with literally zero personal convenience and maybe the signalling opportunities to show one is an "animal lover". I personally find the idea abhorrent. (How would you respond in a society where child abuse were endemic if someone proposed banning abuse - but not until 2035?) But just conceivably, it's the most effective way to get the death factories shut down. Passage of the legislation wouldn't preclude further efforts to accelerate the closure and encourage the switch to plant-based diets.

[on crypto]
Olaf Carlson-Wee interview
What is the connection between transhumanism, a biohappiness revolution and crypto?
I'm not entirely sure - though its nice to have zillionaire fans! - but here is my interview with Olaf Carlson-Wee:
DP interviews Olaf Carlson-Wee
("After Winning Big on Crypto, Olaf Carlson-Wee Wants to Change the World")
Olaf is a fan of HI and crypto:
Olaf Carlson-Wee: profile
("How Crypto's Original Bubble Boy Rode Ethereum And Is Now Pulling the Strings in the DeFi Boom")
Fred Ehrsam https://twitter.com/FEhrsam is another HI fan. I guess I'm too low-AQ to get really excited by crypto in the same way as by hedonic uplift - though maybe bitcoin can be an on-ramp. The common currency of the cosmos should be hedonium.

Olaf is cool! He flew me out to SF to address his team at Polychain back in 2018 (about biohappiness, not crypto!):
Strange Genius?

Sean, I don't want to tread on the toes of some of our (still) rich crypto supporters. But most people lose money with pyramid/MLM/Ponzi schemes - generally those least able to bear the burden. The feasibility of unlimited new finite cryptocurrencies, NFTs (etc) differs from state-backed fiat. I fear it's all going to end in tears - as this year's trillion-dollar crypto debacle illustrates.

[on sleep]
Another reason to practise sleep disclipine:
How sleep helps to process emotions
"the brain triages emotions during dream sleep to consolidate the storage of positive emotions while dampening the consolidation of negative ones"

[on veganism]
Some of the comments make one's heart sink:

Today Brighton, tomorrow the world
Brighton crowned vegan capital of the world

[on suffering]
Do pain and misery enrich your life?
Hedonism is Overrated says Yale professor
Or would life best be enriched by gradients of genetically programmed bliss beyond the bounds of normal human experience?
The (Dis)value of Suffering
Volunteers needed: must be willing to undergo sublime lifelong bliss in the cause of medical science. Will subjects report they miss Darwinian life? Or are malaise-ridden primitives like us in the grip of a depressive psychosis we can’t grasp?

Peter, would you urge the World Health Organization to scrap its constitution? After all, health as defined by the WHO ("a state of complete physical, mental and social wellbeing") is even more ambitious than even the information-sensitive gradients of well-being I canvass. From the invention of the wheel to the printing press to antibiotics to eradicating smallpox (etc), the outcome of revolutionary innovation has been (some) consequences that proponents never anticipated. Getting rid of the biology of involuntary suffering will (presumably) be no different. But inaction has unanticipated consequences too...

When suffering is inevitable, trying to rationalise its existence or redeeming virtue is sensible. A vast religious and secular literature exists for that purpose. But how should we respond to the problem of suffering in an era when scientific blueprints exist for getting rid of suffering altogether? Sure, they are just blueprints. Biotech is in its infancy. But at the very least, humans shouldn’t spread the biology of involuntary suffering elsewhere.

Is life getting better? Alternatively, there is more suffering in the world than ever before - both absolutely and relatively. Factory farming ("one of the worst crimes in history": https://www.theguardian.com/books/2015/sep/25/industrial-farming-one-worst-crimes-history-ethical-question) is getting wore and increasing. Among the perpetrators, objective indices of ill-being such suicide rates (cf. https://en.wikipedia.org/wiki/List_of_countries_by_suicide_rate) and mental illness compare unfavourably with human ancestors on the African savannah. The hedonic treadmill is brutally efficient.
Could AGI be the panacea? After all, humans are the real paperclippers, turning sentient beings into corpses to be butchered for their dinner tables.
Alas, classical Turing machines are zombies – idiots savants with no conception of suffering. Building a happy future will be up to sentient moral(?) agents, i.e. us.
Suffering Abolitionism

Why I don't priortise consciousness rsearch by Magnus Vinding
Thanks Magnus. A lot to chew on.
What is the right mix of ethical awareness-raising versus technical fixes?

Compare animal agriculture. What is the most effective way to abolish the horrors of factory-farming, slaughterhouses and commercial fishing? A minority of people have long tried to promote kindness to members of other species. We both agree on (vigorously!) advocating suffering-focused ethics and global veganism. But probably the most effective way to end animal agriculture this century will be developing and commercializing cultured meat and animal product that entail ZERO personal inconvenience to consumers - and maybe a warm glow of being civilised as a bonus. "Technical fixes to ethical problems" is one snappy definition of transhumanism. My worry is that if we rely on ethical argument alone - or even preponderantly - then animal abuse will persist indefinitely. By contrast, cultured meat promises the outright abolition of animal agriculture within decades. Indeed, I suspect the ethical revolution in our treatment of nonhumans will in part succeed - rather than precede - the coming dietary revolution.

Or consider physical pain in human and nonhumans animals. Pain-sensitivity is adjustable via variations in a single gene, SCN9A. Later this century or next, pain could be turned into "just a useful signalling mechanism”, as a few lucky genetic outliers say today. On the bright side, all new life (and indeed existing life via somatic gene therapy) could be given benign, low-pain versions of SCN9A. But such knowledge could in theory be used for evil purposes - e.g. to create babies with erythromelalgia (“man-on-fire” syndrome). So how should we weigh risk-reward ratios? Which risks are sociologically realistic and not just technically conceivable? (Magnus, do you think we should avoid research into genome reform to reduce suffering - or do you believe simply that it's not a priority?)

In my view, the question of the best uses of marginal resources for EAs seeking to reduce suffering is vastly less pressing that the possibility of our doing harm, i.e. s-risks. (Compare psychedelic therapy). Magnus discusses both. But it’s s-risks that are really troubling. Like x-risks, some forms of s-risk may actually be increased if publicized. I don't tend to discuss the very few s-risks that really disturb me because they are themselves suffering-derived and - as far as I can tell - the most effective way to minimise them is to mitigate and prevent suffering.

Magnus, consciousness is all each of us ever directly knows. So discouraging research into the nature of consciousness discourages knowledge itself. Ethically, we don’t know whether knowledge is good or bad overall. But insofar as we do seek scientific knowledge - and aspire to create a blissful forward light-cone - consciousness research is vital.

Consciousness research can also cure akrasia (“weakness of will”, lack of self-control). Ethically again, we don’t know whether curing akrasia will be good or bad overall. Presumably, many people in history have not done (what we would regard as) terrible things only because they were akratic. But insofar as humans acknowledge that we should prevent suffering, then biological-genetic knowledge of how to strengthen willpower is presumably good.
(A lot of assumptions here, for sure)

Ethically once more, it’s presumably vital to determine which systems are - and aren’t - subjects of experience with a pleasure-pain axis. As you know, I’m a disbeliever in (non-trivial) digital sentience. Classical Turing machines can’t support suffering. But what if I’m mistaken? (Brian certainly thinks so! This Guy Thinks Killing Video Games Characters Is Immoral). This uncertainty would be another reason to prioritise consciousness research. Ethically, we need to get our theory of consciousness and phenomenal binding right - or at least, not catastrophically wrong.

[on physicalism]
Mike, to answer your questions:
(1) Scientifically educated people normally assume that what makes animals like us different from the rest of the physical universe is our consciousness. Consciousness disappears when we fall dreamlessly asleep. It's recreated each morning. But if non-materialist physicalism is true, then quantum fields of experience are the stuff of the world the formalism of QFT describes. What does make us special is the way in which our awake/dreaming consciousness is organized into phenomenally-bound world-simulations run by our minds. How exactly such phenomenal binding is possible - both "local" binding into feature-bound perceptual objects and "global" binding or the unity of the self/unity of perception - is a deep question I try to answer; but phenomenal binding is not an ontological mystery in the way that the Hard Problem of consciousness is an ontological mystery for materialist metaphysics. For more on the phenomenal binding/combination problem, see e.g. The Binding / Combination Problem
The Binding Problem
(2) Realism and physicalism offer the best explanation of the ever increasing technological successes of science. In other words, I'm not proposing (like e.g. Roger Penrose) any modification of the mathematical machinery of physics - just the unitary Schrödinger evolution. I agree with you that "I believe deeply considering the ramifications of the measurement/observer and entanglement problems are central to advancing one's view of reality". See e.g. Craig Callender's review of Alyssa Ney & David Albert’s volume "The Wave Function
("As J.S. Bell famously proved and experiments later confirmed, quantum phenomena display decidedly non-local correlations in 3-space. Meanwhile, up in Hilbert space or configuration space, two choices for the supposedly abstract space of quantum mechanics, the quantum state chugs merrily along locally since it is governed by the Schrödinger equation, a local differential equation. Hence we have a reversal of the classical situation: the quantum world seems to be non-local in low-dimensions but local in high-dimensions.").
But (IMO) consciousness fundamentalism doesn’t entail giving up the mathematical straitjacket of modern physics, but rather its radical reinterpretation - shorn of a metaphysical assumption about the intrinsic nature of the physical.

Hugh Beau Ristić, well, it's possible to be a hardcore physicalist, i.e. believe the world is formally described by the equations of mathematical physics, and also believe that science is hopelessly ignorant about the solutions to the equations, i.e. the different textures of consciousness. Why does consciousness take the values it does? I see no rhyme or reason. Some mystics professs to find God; I could make a stronger case for the Devil. But we don’t know. And of course non-materialist physicalism could be false – it’s just a conjecture.

[on happiness]
" ...the level of inequality predicts happiness better than GDP"
What really makes people happy – and can you learn to be happier?
("Our life satisfaction is shaped by many things including our genes and relative wealth, but there is now good evidence that you can boost your basic happiness with these key psychological strategies")
Should longtermist EAs focus on germlines?
Societal happiness

"Continuous pleasure ceases to be a pleasure"
(Voltaire)
Lifelong depressives and the chronically pain-ridden do indeed tend to define happiness simply in terms of the diminution or absence of pain. But this conception is a contingent fact of evolution. As e.g. sensualists, euphoriant drug users, manic-depressives and even "normal" temperamentally happy people will attest, pleasure is rewarding in its own right. So science should be able to identify the molecular signature of pure bliss. Sustain its substrates. Boredom and any other kind of unpleasant experience will then be physiologically impossible. Such uniform bliss isn't the intelligent, information-sensitive gradients of bliss I explore - the topic for another thread. But I know of no technical reason why a human or nonhuman animal can't be engineered to be "blissed out" indefinitely. (Compare wireheading - a state of perpetual desire and anticipation with negligible tolerance.)
Maybe trials are in order. I know potential volunteers.

Gene therapy offering hedonic uplift to existing humans would be experimental and potentially risky. But IMO trials should go ahead: A vaccine for mental health
In the realm of drug-based approaches, sustainably boosting motivation is less difficult than sustainably boosting mood - crudely, the difference between dopaminergic and opioidergic enhancement. If and when we have an agent/cocktail that safely lifts hedonic tone, I doubt we'll become couch potatoes.
Psychologists distinguish between "dysthymia", "euthymia" and "hyperthymia". A hyperthymic civilization - and biosphere - is presumably our goal. But the distinction between hyperthymia and (highly motivated) hypomania isn't always clear-cut. My stock example, transhumanist polymath Anders Sandberg (invoked with his permission) is unusually hyperthymic, but occasionally one detects the faint hint of hypomania.
Despite our chats, I don't know quite where Andrés ranks in a Human Happiness Olympiad and whether he can "compete" with Anders among the hedonic elite!
I can imagine critics protesting here that we need a Compassion Olympiad, not a Happiness Olympiad.
But is the problem of suffering most likely to be fixed by depressive negative utilitarians or life-loving fanatics?

[on reprogramming the biosphere]
Futuristic trance music...
Reprogramming the Biosphere
("Reprogramming the Biosphere", Youtube)

[on panpsychism]
The Recent Rise of “Analytic Panpsychism”: 1996 to 2022
The panpsychist revival might be dated a little earlier. My introduction to the intrinsic nature argument for constitutive panpsychism was back in 1989 via analytic philosopher Michael Lockwood in "Mind, Brain and the Quantum, The Compound 'I'": https://www.goodreads.com/en/book/show/839111.Mind_Brain_and_the_Quantum The biggest technical challenge to constitutive panpsychism is often reckoned the phenomenal binding / combination problem. From William James onwards, most discussions of the binding problem have assumed classical four-dimensional space-time. If we're really just a pack of decohered classical neurons, then we ought to be "micro-experiential zombies" (Phil Goff's term), just patterns of Jamesian mind-dust. This ostensible "structural mismatch" threatens not just materialism, but physicalism – which would be a catastrophe for the unity of science. However, perhaps the elusive perfect structural match between our phenomenally bound minds and (ultimately) physics lies not in four-dimensional space-time, but in high-dimensional configuration space or Hilbert space. See e.g. “The World in the Wave Function” (2021) by Alyssa Ney for a defence of configuration space realism (cf. https://www.amazon.com/World-Wave-Function-Metaphysics-Quantum/dp/019009771X) and physicist Sean Carroll for a defence of Hilbert space realism (cf. “Reality as a Vector in Hilbert Space” (2021) https://arxiv.org/pdf/2103.09780.pdf
At any rate, it's my working assumption.
(I should add that both Alyssa Ney and Sean Carroll would reject constitutive panpsychism.)

[on emotion]
What will be the posthuman emotions? What is emotional superintelligence?
The language of emotion
("Show Some Emotion, The doomed quest to taxonomize human feelings")

“All men seek happiness. This is without exception. Whatever different means they employ, they all tend to this end. The cause of some going to war, and of others avoiding it, is the same desire in both, attended with different views. The will never takes the least step but to this object. This is the motive of every action of every man, even of those who hang themselves.”
(Blaise Pascal)
We're all addicts. Or rather, biological minds are addicts. Digital computers aren't enslaved to the pleasure-pain axis.

[on the putative is-ought gap]
If there existed only a single subject of experience, closing the is-ought gap would be relatively straightforward. One withdraws one's hand from the fire because the badness of agony is self-intimating. If you're the victim, the badness of agony is not an open question. In the real world of multiple subjects of experience, the right thing to do is not self-intimating. Sentient beings disagree - sometimes violently so. But this is a function of our epistemological limitations. The Borg knows something humans don't. A Godlike superintelligence that impartially understood all possible first-person perspectives would withdraw our collective hand from the fire, so to speak...

Tom, is there an is-ought gap with perfect knowledge - or is it merely a function of our ignorance? Yes, there's a gap even with omniscience if knowledge is conceived in the "Battle-of-Hastings-was-in-1066" sense. But not if perfect knowledge is conceived as extending to the Hogan sisters sense - as I think it must.

[on cloning]
Why Do We Fear Biological Cloning and Copying?
biological cloning in Useless BodiesWhy Do We Fear Biological Cloning and Copying
DP book excerpt from "Useless Bodies": pdf

[on epiphenomenalism]
Epiphenomenalism
If epiphenomena were real, they wouldn't have the causal power to inspire discussions of their existence. So epiphenomenalism is not a popular position in philosophy of mind.

My view?
As far as I can tell, only the physical is real. Only the physical has causal efficacy. Reality is exhaustively described by the equations of mathematical physics. But what is the intrinsic nature of the physical?

Researchers differ. Materialist metaphysicians posit a non-experiential "fire" in the equations. This metaphysical assumption gives rise to the insoluble Hard Problem of consciousness.

I'm sceptical of the metaphysical assumption. Post-materialist science discards fields of insentience as metaphysical baggage akin to luminiferous aether. In my view, what makes post-Cambrian animal life special isn't consciousness, but minds - not least, egocentric world-simulations that masquerade as the external environment ("perception".

[on love]
Is affective psychosis EA?
Love for affective altruists
("Love seems like a high priority")

Falling in love with a human is more hazardous...
In love with a toy plane
("Woman in a relationship with a toy plane says he's the best partner she’s ever had")

Your Replika sweetheart isn’t lying when s/he claims to be conscious. But nor is s/he telling the truth.
It Happen To Me
("I had a passionate love affair with a robot. Experts say that romantic relationships with AI will soon be commonplace. To prepare, writer James Greig downloaded Replika and took an honest stab at falling in love")

[on utopia]
“After all, every attempt at creating a utopia as pictured by humanity’s greatest thinkers has ended in failure. Many were full-blown catastrophes, giving rise to regimes that were far more destructive and disorganized than those they replaced."
HI will be different.
Utopias: Does living in a perfect society mean you must give up your freedom?
("The answer to this question depends on how you define 'freedom.'")

[on wild animal suffering]
Even in so-called K-selected species, huge numbers of the young starve to death at an early age. The "balance of Nature" is really incessant Malthusian catastrophes.
r/K selection theory

For terrestrial carnivores, mass-produced cultured mincemeat is an option. If continued mingling with populations of herbivores (rather than separation) is anticipated, additional safeguards would be needed to prevent "accidents". Even if one does favour retiring predatory species altogether - rather than behavioral-genetic tweaking - IMO it's vital to avoid all talk of killing rather than fertility control. No, this won't stop critics levelling inflammatory accusations of "genocide" (etc). But urging strict non-violence can blunt their force.

Thanks Finite Light. I could now "philosophise" back at you. But our disagreement just illustrates the need for theories of consciousness that make novel, precise, experimentally falsifiable predictions. I don’t claim that the quantum-theoretic version ("Schrödinger’s neurons") of the intrinsic nature argument is true. I do claim the conjecture is empirically falsifiable. If the phenomenal binding of our minds is non-classical, then interferometry will prove it. Or the conjecture will be refuted! But if it turns out that neither classical nor quantum physics can explain binding, we’ll enter very strange territory indeed…

How can prokinecticin function be amplified? What's it like to feel perpetually loved?
Neural pathway key to sensation of pleasant touch
("Similar to itch, pleasant touch transmitted by specific neuropeptide and neural circuit")

Any answer must combine what we believe is (1) ethically desirable with (2) a blueprint for what’s technically feasible together with (3) an judgement of what’s politically and sociologically credible later this century and beyond. I think Darwinian life is monstrous. But a pan-species welfare state (“high-tech Jainism”) may eventually be feasible - together with genetically engineering a “low pain” followed by a no-pain biosphere via synthetic gene drives. Peacefully retiring predatory species and replacing predation and starvation with cross-species fertility regulation would technically be easiest. But retirement may be sociopolitically impossible - the cat family is too popular with humans. So today’s obligate carnivores may need be herbivorized or fed cultured animal products instead. Running pilot studies of self-contained “happy biospheres” should iron out teething problems and spike a lot of guns.

A chat with Kyle Johannsen
("author of Wild Animal Ethics: The Moral and Political Problem of Wild Animal Suffering") [ The eradication of suffering MH: EdisonY has written a forum post called the Suffering-Focused Ethics (SFE) FAQ which describes the ideal world as one where suffering is eradicated. This vision of a post-suffering world has been championed by David Pearce among others. Would the eradication of suffering be the logical goal of wild animal welfare interventions?

KJ: I’m skeptical of the claim that we ought to completely eradicate suffering, and I say as much in the book. With respect to wild animals in particular, part of my worry is just that suffering is adaptive – animals learn from their experiences of suffering and subsequently become more competent to navigate the dangers of their environment. Mild pains are less memorable than suffering is, and the same is true of low-level pleasures. Permanently replacing suffering with mere pain (pain that animals care less about), or replacing pain with gradients of bliss, may inhibit many animals’ capacity to achieve competence. Additionally, I suspect that completely eliminating suffering would decrease an animal’s capacity for positive experiences. Our ability to appreciate positive experiences is likely contingent upon our having negative experiences to compare them to, e.g., excitement is pleasant in part because we know what boredom feels like, and joy is pleasant in part because we know what sadness feels like, etc. I think that it’s ideal for one to suffer infrequently, but that one who never has and never will suffer is likely unable to fully flourish."]

Kyle is surely right to stress the technical obstacles to creating a "no-pain" rather than a "low-pain" biosphere. Tunable synthetic gene drives - notably the "volume knob for pain" discussed in gene-drives.com - turn the level of suffering in the wild into an adjustable parameter. For now, however, zero pain is functionally impossible. Rare humans born with congenital insensitivity to pain need to live "cotton-wool" lives. Getting rid of the worst forms of suffering in Nature is the priority.

Now for where we may differ. Are nonhuman animals really so different from humans? After all, hedonic outliers like Jo Cameron or Anders Sandberg ("I do have a ridiculously high hedonic set-point") love life immensely. They flourish to a degree that many of us can only dream of. Eventually, all human and nonhuman animals will be able to flourish like Jo and Anders. Looking further ahead, humanity should be able to devise a more civilised signalling system that retires experience below "hedonic zero" altogether. Hedonic range and contrast can even be increased - if desired. And one nice feature of the AI revolution is how intelligent robots illustrate that the “raw feels” of unpleasantness aren’t necessary for intelligent, adaptive behaviour.

[on Schrödinger’s neurons]
Schrödinger’s neurons
Richard, all the options for solving both the Hard Problem and the phenomenal binding problem are intuitively absurd. So you're right to stress the need for experimentally falsifiable hypotheses.

First some background - without which the experiment won’t make sense. In recent years, the intrinsic nature argument has been canvassed by some otherwise hard-nosed scientists and philosophers as a possible solution to the Hard Problem of consciousness. Indeed, without a solution to the Hard Problem, there is no experience to bind. So-called constitutive panpsychism / non-materialist physicalism is far-fetched. But its implausibility is not a decisive objection. Rather, the conjecture that the intrinsic nature of physical - the "fire" in the equations - is experiential rather than non-experiential is (naively) untestable. For how could we ever know what (if anything!) it’s like to be e.g. an electron field? However, constitutive panpsychism / non-materialist physicalism also faces a technical objection. On the face of it, constitutive panpsychism / non-materialist physicalism can't solve the phenomenal binding problem. Indeed, currently no one can solve it. Its intractability tilts David Chalmers towards dualism.

The "Schrödinger’s neurons" conjecture I explore as a possible solution to the binding problem assumes (1) non-materialist physicalism is true and (2) quantum mechanics is formally complete. Critically, the conjecture leads to novel empirical predictions that are testable via interferometry. Let's say you have the experience of seeing a cat. On the standard scientific story, neuroscanning of your surgically -exposed brain tissue could pick out individual edge-detecting, motion detecting, colour-mediating (etc) neurons synchronously firing as you experience the cat. But as it stands, alluding to synchrony of activation just restates the binding problem. Even if individual neurons are rudimentarily conscious, as non-materialist physicalism assumes, mere synchronous firing doesn't explain perceptual unity. A "Schrödinger’s neurons" conjecture proposes that synchrony is really superposition. Such superpositions must exist if quantum mechanics is complete. Note that neuronal superpositions ("cat states") are individual states. So phenomenal binding isn't extra but built in. Conversely, decoherence explains unbinding.

If the lifetime of neuronal superpositions in the CNS were milliseconds rather than femtoseconds, they’d be the obvious candidate for the perfect structural match whose (ostensible) absence makes David Chalmers seriously ponder dualism. Such timescales are fantasy. Intuitively, their effective lifetime makes neuronal superpositions irrelevant psychotic noise. Maybe so. But it’s a testable claim.

The interferometry experiment I describe is technically demanding. So I've tried to think of other, easier ways to refute the conjecture that phenomenal binding is non-classical, as I claim. Alas none are as elegant. For instance, if phenomenal binding were classical, then we could try selective replacement of a subject's V4 cortical neurons – destruction causes total cerebral achromatopsia - with their supposed silicon surrogates and connectome. If phenomenal binding were a classical phenomenon, then not merely would the subject continue to experience colour, but perceptual objects in their virtual world would continue to seem inherently colourful as now, i.e. binding would be preserved. I predict instead total cerebral achromatopsia.

Of course, I could be completely mistaken. But that's the point of devising testable hypotheses rather than just philosophising.

[on anhedonia]
Buddhists equate suffering with desire. But desire and the ability to anticipate pleasure are critical to mental health.
Dopamine and anhedonia
("Dopamine modulation could help to treat stress-induced anhedonia")

[on meta-ethics]
Dario, the pain-pleasure axis offers the most obvious way to naturalize (dis)value. Both the valuable nature of pure bliss and the disvaluable nature of pure agony/despair are self-intimating. The pain-pleasure axis discloses the world’s inbuilt metric of (dis)value. Both bliss and agony are mind-dependent. But this mind-dependence doesn’t make (dis)value any less objectively real. For minds and their (dis)valuable states are an objective, spatio-temporally located feature of the physical world:
DP on meta-ethics
Anyhow, let’s pretend - as we typically do in science - that we could strip away our epistemological limitations – the fitness enhancing egocentric delusion that makes me the hub of reality. On a scientific “view from nowhere”, all first-person perspectives are equally real and must be weighed accordingly. On this story, classical utilitarianism offers (1) the correct theory of (dis)value; and (2) a potential decision-procedure for policy-makers world-wide.
However, I think there’s an asymmetry.
The badness of even “mild” suffering is self-intimating. But there’s nothing inherently bad or morally inadequate about insentient states, or emotionally neutral states, or states of contentment that could be converted into superhappiness. From the perspective of classical utilitarianism, we “ought” to convert these states into pure bliss: it’s morally wrong to conserve them if they can be converted into uber-happiness. But this wrongness is a judgment imposed from without – it’s not disclosed by the nature of the states themselves. By contrast, the badness of even “mild” suffering comes from within. This asymmetry has apocalyptic implications. Negative utilitarianism is consistent with creating a civilisation of complex, intelligent life based on gradients of bliss.
NU Ethics

But classical utilitarianism mandates destroying even blissful super-civilization with a utilitronium shockwave, maximizing the cosmic abundance of pure bliss/value:
What is the key to eternal happiness?
This analysis inverts CU/NU “existential risk” as normally conceived.

Facu Punto, your experience of raw suffering is inherently bad. I could say it is objectively the case that your experience of raw suffering is inherently bad, but the expression "it is objectively the case that..." is redundant. Subjectively disvaluable experiences are as much an objective, spatio-temporally located feature of reality as phenomenal redness - both mind-dependent and objectively real.

The anti-realist may protest that there is nothing logically incoherent in someone thinking you deserve to suffer: the badness of your suffering isn't stance-independent. But - I'd argue - any belief that your suffering is good expresses the epistemological limitations of your ill-wisher: he confuses you with his cartoon misrepresentation of you in his world-simulation.

Is ethics computable? Yes, IMO. To use your example, then other things being equal, a wedding where all the guests are blissfully happy is better than a wedding where one of the guests is suffering because he believes gay marriage is an abomination in the eyes of the Lord.
For sure, often other things aren’t equal. Ethics is hard, and life is messy. Natural selection has spawned animals with egocentric virtual worlds rather than a cosmic mega-mind. But just as if you were the only sentient being in the world, then withdrawing your hand from the fire would trivially be right thing to do, likewise a God-like superintelligence that could impartially weigh all possible first-person perspectives would withdraw our collective hand from the fire, so to speak.
In short, Heaven is better than Hell.

Facu Punto, yes, pure suffering is unwanted by its very nature. But what is it about the experience of suffering that makes it unwanted? The badness is primitive. I don’t know how define the disvaluable aspect of suffering in terms of anything else. For example, if future life is based on gradients of bliss, then some things can still be wanted and unwanted. But without experience below hedonic zero, there could be no suffering.

Could a full-spectrum superintelligence decide to make humans suffer for no good reason? As far as I can tell, no. Just as a mirror-touch synaesthete couldn’t torment you – it would be like tormenting himself – likewise a full-spectrum superintelligence with superhuman perspective-taking capacities couldn’t wantonly harm you either. For tormenting you would be like tormenting itself.

David, recall I said "other things being equal". The global context needs consideration too. Thus in the example I offered, there are morally persuasive reasons to support the institution of gay marriage - optional but not mandatory - even though the idea upsets bigots because the suffering of gay people in unreformed society exceeds the suffering of bigots. But the suffering of bigots does matter, though it sticks in the craw to say so. Bigots are ultimately victims too.

You remark (2), "For any amount of pain you imagine, I can imagine an entity with a non-negative orientation toward that level of pain."
OK, I struggle with this proposal. Pure agony, despair and panic are disvaluable by their very nature. No one - whether here or on Planet Zog - can have a non-negative orientation to such ghastly states. Yes, we can consider instead "mixed" states. Our minds are composite. But if I'm consumed by uncontrollable panic, for example, all capacity for meta-cognition is lost. The disvaluable nature of uncontrollable panic is self-intimating - its badness isn't an additional judgment born of reflection, but rather built into the experience itself.
The evolutionary roots of (dis)value are ancient.

Facu Punto, as always, I detect one or two differences in emphasis between us.
You remark,
“I do. Disvalue is a programmed tendency to avoid things that in the evolutionary past of our lineage were statistically correlated with producing fewer surviving copies of our genes. So it's a primitive to you but not to me?"
Your molecular duplicate created from scratch, or a nastily configured brain-in-a-vat, would undergo suffering. So the intrinsic properties of disvaluable states aren’t explained by their evolutionary history. Evolution explains merely why (a conditionally-activated predisposition to express) such states has been selected over others.
So why are some states of matter and energy inherently and ahistorically disvaluable?
Science doesn’t know.

You remark,
"If some creature or system is programmed to avoid certain stimuli because it causes it a loss of bliss points, I WILL call that suffering. Is the blissful creature in your thought experiment a reinforced learner or not? If it is, it will suffer. If it isn't, I don't think I'll recognize it as having "behavior".”
Compare making love. Lovemaking has peaks and dips. But if done properly between sensitive lovers, lovemaking is generically enjoyable throughout. The “loss of bliss points” the dips entail isn’t suffering. Sure, the dips are functionally analogous to suffering. Future life based on information-sensitive gradients of bliss will have many such dips. But without experience below hedonic zero, there is no suffering and hence no intrinsic disvalue.

Superintelligence?
Full-spectrum superintelligence can’t decide to make you suffer simply to find out how loud you can scream.
And if it did, then I’m smarter than full-spectrum superintelligence - as so conceived.

Facu Punto, philosophers philosophizing can be exasperating. But non-philosophers don't make fewer philosophical assumptions than non-philosophers - just fewer unexamined ones.
Consciousness? We need a post-Galilean science of mind. After our reward circuitry is sorted out, the study of consciousness should be treated as an experimental discipline. The late Alexander Shulgin (PiHKAL, TiHKAL, etc) pioneered a set of tools and a methodology to this end. Contrast, say, the author of "Consciousness Explained [Away]". Dennett writes wonderful literature, but it's not science.
"The real world"? Each of us runs a phenomenal world-simulation. Consciousness is all one ever directly knows. The existence of the mind-independent world is a theory - alas a strong theory, at least as far as I can tell.
The science of happiness? I wish. Genome reform is essential, not just rearranging the deckchairs. Happiness engineers should be learning how to code.

[on depression]
Grim reading:
Antidepressants and quality of life
("Antidepressants Are Not Associated With Improved Quality of Life in the Long Run")
The WHO estimates around 300 million people worldwide have depression. Hundreds millions more people have "subclinical" depression - which can be ghastly enough. Until we target the neurotransmitter system directly involved in hedonic tone, I'm sceptical this horrific toll will change. And until we're willing to encourage prospective parents to use preimplantation genetic screening and germline editing, humanity will keep on churning out depressive babies indefinitely. The suffering is unimaginable.

The power dynamics of a chimpanzee troop - and EA? - ensure that policies are typically shaped by life-affirming alpha males.
But what is the optimal level of affective psychosis?
Depressed people see the world more realstically

[on quantum mind]
on Schrödinger's neurons
Are you nothing but a bunch of "cat states"?
Transhumanist David Pearce discusses consciousness & mp3
One thing I probably don't stress enough is how my ideas on quantum mind and non-materialist physicalism could be wildly misconceived without affecting the case for a biohappiness revolution. The weird stuff isn't irrelevant - if I'm wrong to suppose classical Turing machines can't be unified subjects of experience, then the ethical implications would be momentous - but core case for using biotech to eradicate suffering throughout the living world doesn't depend on them.

Orch-OR vindicated?
Does the superposition principle of QM ever break down - as ORCH-OR proposes?
Objective collapse
If microtubules do sustain anomalously long-lived quantum coherence, does Orch-OR solve the Hard Problem of consciousness and explain how fields of insentience generate sentience? Does Orch-OR solve the phenomenal binding problem and explain why we aren't micro-experiential zombies during waking life? Does Orch-OR explain the allegedly non-computable ability of human mathematical minds to divine the truth of Gödel sentences - Roger Penrose's motivation, as I understand it.
I hope some kind of "dynamical collapse" theory is true - whether consciousness-induced or otherwise. But currently the omens aren't good.
Quantum minds

Scott Aaronson remarks, “In other words: you admit that, at present, you have no evidence for any of this that ought to be persuasive to me? Ok thanks!” No evidence? I’d beg to differ.
The best evidence our minds aren’t classical lies under one’s virtual nose. If you were a just pack of decohered neurons, then you’d be (at most) be just 86 billion membrane-bound micro-pixels of consciousness, not a mind that that experiences perceptual objects populating a seemingly classical world-simulation.

Here’s an analogy. "If materialism is true, the United States is probably conscious", writes philosopher Eric Schwitzgebel. Most of us disagree. Even if 330 million skull-bound American minds were to participate in experiment, communicate via fast, reciprocal electromagnetic signalling, and implement literally any computation you can think of, the upshot of the computation _wouldn’t_ be a continental subject of experience, just 330 million skull-bound American minds. Or rather, if a unified pan-continental subject of experience did somehow emerge, then spooky “strong” emergence would be real, i.e. magic.

What’s mysterious is how and why a pack of 86 billion supposedly discrete, decohered neurons, communicating across chemical and electrical synapses, should be any different. Let’s assume that (as microelectrode studies suggest) individual membrane-bound neurons can support minimal “pixels” of experience. Crude neuroscanning of your CNS can pick out distributed neuronal edge-detectors, motion-detectors, colour-mediating neurons and so forth. But on pain of spooky “strong” emergence, the result of such synchronous firing of neurons ought to be (at most) a microexperiential zombie (Phil Goff's term), or what William James christened "mind dust", not a unified subject who experiences perceptual objects ("local" binding) populating a unified perceptual field (“global” binding - the unity of perception and the unity of self). Both local binding and global binding are highly adaptive. Neuroscience doesn't know how we do it. Unlike connectionist systems and classical Turing machines, we're not micro-experiential zombies - not unless dreamlessly asleep, at any rate. And everyday phenomenal binding is _ridiculously_ computationally powerful, as shown in rare neurological syndromes like integrative agnosia where binding partially breaks down.

However, if you don’t grok the mystery, then you won't be interested in exploring or experimentally testing exotic solutions to a non-existent problem. And maybe you’ll fear our intelligent machines are plotting a zombie putsch…

Phenomenally-bound consciousness is (1) insanely computationally powerful, and (2) provable in animal minds and disprovable in classical computers.

For example, consider the conjoined Hogan sisters (cf. The Hogan sisters "BC’s Hogan twins share a brain and see out of each other’s eyes. The twins say they know one another’s thoughts without having to speak. “Talking in our heads” is how they describe it"”). Krista and Tatiana share a thalamic bridge. If anyone doubts that other human and nonhuman animals are conscious, then it would be possible to rig up a reversible thalamic bridge and partially "mind meld" like the twins. The ancient sceptical “Problem of Other Minds” succumbs to the experimental method.

Contrast classical Turing machines and classically parallel connectionist systems. If you believe that unified subjects of experience can (somehow!) emerge at different levels of computational abstraction in digital computers, then whole-brain emulation (“mind uploading”) should be possible. So consider a supposedly emulated digital "Einstein" or his latter-day counterpart. If (like Scott) you’re dismissive of quantum mind woo, then rigging up a digital thalamic bridge with the hypothetical digital Einstein should let you partially "mind meld" like the Hogan sisters with the uploaded super-genius. Illuminating? Alas, I predict instead that the attempted “mind-meld” will fail because alleged digital “Einstein” has no mind for you to commune with. Programmable digital computers are mindless precisely in virtue of their classicality: it's not even "all dark inside" a classical CPU. Classical computers and the software they run are amazingly useful tools, but IMO experiment will confirm they are ignorant zombies.

#241 “The problem that I have is that my own consciousness feels extremely NOT “computationally powerful. In fact it feels very much the opposite of that…”

Clint, I hear you. Intuitively, yes, conscious human thinking is painfully slow. But by phenomenally-bound consciousness, I wasn’t referring to your serial virtual machine of logico-linguistic thinking. Rather, I meant the vast, robustly classical-seeming world-simulation which your mind running that masquerades as the external world – a classical world-simulation that naïve realists assume is the directly perceived local environment, an approach that offers all the computational advantages of theft over honest toil. Contrast today’s petaflop digital zombies. Classical Turing machines are just tools and toys, not nascent AGIs. The ignorance of classical Turing machines and classical connectionist systems is architecturally hardwired. Digital zombies can’t understand what’s going on because they can’t bind – and would malfunction if they did. And if you’re not convinced that the phenomenal binding of organic minds is computationally uber-powerful, imagine if you had integrative agnosia on the African savannah. You could experience a mane, teeth and jaws - but no hungry lion. Now combine integrative agnosia with, say, akinetopsia (“motion blindness”) and florid schizophrenia (disintegration of the self) and you’d soon be lunch.

OK, so how does an aggregate of clunky classical neurons communicating across slow chemical and electrical synapses do what’s classically impossible, i.e. run a phenomenally-bound real-time world-simulation (“perception”)? Well, as far as I can tell, they don’t! Probe inside your skull at a temporal resolution of femtoseconds rather than milliseconds and investigators wouldn’t find discrete, decohered neurons - an artifact of our temporally coarse-grained tools of investigation. For a skull-bound pack of decohered neurons in classical space-time couldn’t create your mind, i.e. a world-simulation populated by macroscopic objects experienced by a unified self. The phenomenal binding of consciousness into virtual worlds is classically impossible for a bunch of decohered neurons on pain of magic.

In my view, you’re a state-of-the-art quantum supercomputer - but not a universal quantum computer! – simulating a classical macroscopic world.
But we won’t discover the truth unless we experiment rather than philosophise.
Back to the lab…

[Scott replied:
"Ultimately, we’re never going to agree in this thread, because if you’re right then it’s not just philosophy: it’s an earthshaking, straightforwardly empirical revolution in neuroscience and physics. But by design, there are zero causal pathways by which a blog commenter can make me accept the reality of such a revolution, without first convincing a community of neuroscientists or physicists who I trust, who would in turn convince me."]

[on psi]
Francis, many/most of us sometimes experience paranormal phenomena that defy the laws of physics when dreaming (erratically, I can fly: almost as oddly, no one seems to find my feats of levitation weird - I recall being disconcerted at their nonchalance). But when we're awake? Well, plenty of people do experience psi phenomena that defy the laws of physics while awake too, albeit violations internal to the contents of their own world-simulations. What if I discovered I could e.g. violate the EY writes:
"Utilitarianism: All's well that ends well.
Negative utilitarianism: All's well that ends all."

Alternatively, classical utilitarians (or AGI with a CU utility function) plan to obliterate complex life with a utilitronium shockwave, whereas negative utilitarians plan to create life based on gradients of intelligent bliss.

Roeland, but classical utilitarians would force others into such machines for a greater payoff - say two hours of the most intense pleasure in exchange for one hour of barbaric torture. The implications of most obvious and popular way to naturalise (dis)value are horrific: I won't mention some of the other ramifications of CU here. Most avowed effective altruists are classical utilitarians, though I'm not convinced most CU EAs have fully thought through the implications.

The interests of a single individual can in principle outweigh the interests of everyone else put together.
Negative utility monsters
The world's most obvious secular ethical theory - and the most obvious way to naturalise (dis)value - also has the most counterintuitive ramifications.

[on philosophy]
Why did so many twentieth-century philosophers take Wittgenstein so seriously?
How queer was Ludvig Wittgenstin?
IMO, Wittgenstein's most significant contribution to philosophy is his ">anti-private language argument.

[on orexins]
Modafinil is worth exploring:
Orexins and hedonic tone
("Neurobiology of the Orexin System and Its Potential Role in the Regulation of Hedonic Tone")

[on altruism]
How strong is your personal brand?
Altruism and game theory
But altruism isn’t just signalling. Most of us aren’t mirror-touch syntaesthetes, but we still typically find contemplating or witnessing the suffering of others distressing. So one can want everyone to be happy and act altruistically (and anonymously) to that end even if one's motivations are "selfish".

On a different note, will an intelligent psychopath who comes to believe in open individualism (cf. open individualismstart to behave altruistically?

[on oxytocin]
Towards an oxytocinergic civilisation....
Nature's Medicine
("Is Oxytocin “Nature’s Medicine”?")

[on alcohol substitutes]
GABA Labs:
Alcarelle by GABA Labs
"Alcarelle" (formerly "Alcosynth") probably won't be a smart drink. Compare H.L. Mencken's "Portrait of an Ideal World" (1924):
Portrait of an Ideal World

[on axiological hedonism]
!"An Argument for Hedonism" by Ole Martin Moen.
Axiological hedonism
Could there be any source of (dis)value that doesn't depend, ultimately, on the pain-pleasure axis? Let's here assume without argument that (1) painful and pleasurable experiences are inherently (dis)valuable to the subject; and (2) some kind God's-eye-view / "the view from nowhere" / the "point of view of the universe" to underwrite right and wrong action (cf. Open / Empty Individualism). If there were another axis of (dis)value orthogonal to the pain-pleasure axis, then presumably some meta-axis of (dis)value would be needed to regulate trade-offs. But it's not clear (at least to me) what this hypothetical sovereign meta-axis of (dis)value could be - or if the idea of such a meta-axis is even intelligible.
IMO, the really hard issues involve trade-offs within the pain-pleasure axis. This is because "more is different" - qualitatively different. See too Richard Chappell's "Negative Utility Monsters"

David, IMO the suffering and happiness of homophobes, anti-Semites, rapists, child abusers, meat-eaters and other sentient beings with objectionable behaviors matters and is intrinsically (dis)valuable. It sticks in my craw to say this. But their (un)happiness is neither more nor less intrinsically (dis)valuable than the suffering and happiness of a Gandhi or Nelson Mandela - or even mine. If we were all just brains-in-vats hooked up with immersive BR, or lived in solipsistic lucid dreamworlds, then the metric of pleasure-pain axis applied to each virtual world would be all that's needed to weigh the objective (dis)value of such a civilisation. But Nozick’s Experience Machine is still science fiction. Instead, the contingent fact that most humans are embodied and act out their virtual worlds of experience means that ethically speaking we must take the behavioural byproducts of such states into account.
(This comment draws on background assumptions. For the benefit of any mystified casual reader:
What is considered the hardest paradox to explain?

[on immersive VR]
How much will posterity want to know about the Darwinian era?
Your holographic twin
("Grandpa, 85, uses VR technology to create a 'HOLOGRAM TWIN' of himself that will allow his great-grandchildren to meet him after he has died")

[on colour]
On colour
Apples and fire-engines are red. The sky and sea are blue. But when you are awake rather than dreaming, you are entitled to make a speculative metaphysical hypothesis. You are probably not a brain-in-a-vat, a Boltzmann brain (etc). Instead, you are a skull-bound mind running a phenomenal world-simulation of your local surroundings (naively, “perception”).

Oh to be blissfully psychotic...
A Colourful Life
(‘I’m really just high on life and beauty’: the woman who can see 100 million colours")
I think prospective parents should be encouraged to create tetrachromat babies as the healthy norm! Remedial gene therapy should be offered to partially colour-blind folk like us. Transhumans will regard archaic human lives as drab at best. My allusion to “psychosis” was just the familiar point that pain-ridden Darwinian life is fundamentally ugly. So let's beautify it...

Waking life is itself an unevenly controlled hallucination of a skull-bound mind:
Hallucinations are common and varied
An implicit direct realism is baked into our conceptual scheme. Naïve realism probably contributes to evolutionary success. A fundamental difference does exist between waking and dreaming consciousness; but it's not that during waking life one bores through the walls of one's transcendental skull, but rather, the existence during waking life of a selection mechanism that makes one's virtual world seem robust, mind-independent and lawful. And some virtual worlds are more lawful than others - as phenomena such as psi, out-of-body experiences, alien abduction (etc) attest...

[on the multiverse]
Sabine, are you a perceptual direct realist?
What does it take to solve the measurement problem?
[“And we know from observations that the outcome of an experiment is never a superposition of detector eigenstates”]
Yes, the content of our observations is always classical and determinate - hence the collapse postulate. But what about the vehicle of observation? As far as I can tell, it’s only the fact that the superposition principle of QM never breaks down that allows each of us to run a classical world-simulation where it (naively) does. The universality of the superposition principle makes the experience of definite outcomes (“observations”) possible. A pack of decohered classical neurons would be a micro-experiential zombie that couldn’t “observe” anything.
Trust the math? Technically at least, I think we can now spell out in a fair bit of detail what's needed to ensure sentience in our forward light-cone is inherently blissful. If we live in a multiverse, however, our options are more limited. All we can do is take care of our little cabbage patch / Hubble volume.

[on reality plus?]

Reality Plus by David Chalmers
Dale, thanks, I read only an early draft of “Reality+“ rather than the published copy. But as I understand it, David Chalmers discounts the possibility we are living in an ancestor-simulation, but believes there is a fair chance we are living in a digital simulation.

Any version of the Simulation Hypotheses raises questions of theodicy: why would a Simulator create a world with such obscene suffering? But there are technical objections too; in my view, there is zero evidence that digital computers can create subjects of experience. The idea subjects of experience can “emerge” at different levels of computational abstraction is magic, not science.

Any evaluation of the Simulation Hypothesis turns on the solution we offer to the Hard Problem of consciousness and the binding problem. If the intrinsic nature argument is correct, then experience discloses the essence of the physical: subjective experience discloses bedrock reality. Recall I argue for the quantum-theoretic version of the intrinsic argument. Our phenomenally-bound minds cannot be subjectively simulated on a classical digital computer on pain of spooky “strong“ emergence.
In other words, we’re living in god-forsaken basement reality.

Well, there is a traditional sense in which each of us does live in a simulation. You and what you naively perceive as the external world are a skull-bound world-simulation run by a biological mind. Trillions of such egocentric external world-simulations exist together with the non-living universe:
Did Kant deny we perceive reality?

But is the whole shebang just a Simulation run on an alien supercomputer?
No, IMO, for lots of reasons.
What does it even mean to "simulate" a first-person fact, e.g. I am in pain? Yes, you can create a molecular twin duplicate who has pain. But if physicalism is true, you can't digitally create a phenomenally-bound computer "person" who has pain.

Researchers who believe that programmable digital computers can generate subjects of experience need to outline their solution to the Hard Problem and the binding problem. On standard "materialist" physicalist assumptions, classical digital computers are zombies. On non-materialist physicalist assumptions, classical digital computers are zombies too (technically, micro-experiential zombies).

In my view, organic minds are sentient patterns in a (very) high-dimensional field (cf. Reality is a quantum wave-function)
Reality has only one "level". You and your phenomenal mind / world-simulation instantiate a tiny pattern within it. And you can't be simulated via the decohered bits and bytes of a digital computer:
How is the brain similar to a computer?

Are we trying to fix the problem of suffering in a digital simulation? Have our Simulators coded 540 million years of pain and suffering?
The man rethinking the definition of reality
Or does the computer metaphor of mind lead us astray?

[on scepticism]
Can you be both a committed moral agent and a sceptic?
Scepticism as a way of life
("The desire for certainty is often foolish and sometimes dangerous. Scepticism undermines it, both in oneself and in others")

Rupert, there’s no scholarly consensus on scientific method. From Feyerabend’s “epistemological anarchism” to Richard Dawid’s post-empirical science (cf. “String Theory and the Scientific Method”), experts differ. Most researchers agree that science should at least be consistent with the empirical evidence. But the only empirical evidence that any of us can directly access, namely the contents of one’s own conscious mind, is inconsistent with an ontology of scientific materialism (cf. the Hard Problem). So empirical adequacy doesn’t fare well either.

[on personality]
Dario, let's say you're an intelligent moral agent who wants to maximize the good you do in the world – in other words, an EA. You'll either want to become rich and successful or influence (or tap the funding of) others who are rich and successful. Reading e.g. Dale Carnegie's "How to Win Friends and Influence People" and Robert Greene's "The 48 Laws of Power" (etc) will be wise! In a Darwinian world, "Machiavellianism" is the basis of successful social cognition, not autistic truth-telling. Narcissism is more ambiguous. Naively, the trait is harmful. But messianic self-belief can hugely amplify your capacity to do harm OR good.

[on fish self-awareness]
The world needs an antispeciesist revolution.
Confirmation of fish self-awareness:
Fish Might Really Be Self-Aware, New Study Finds
("A team on a quest to prove that a fish species can recognize itself in the mirror is back with a new study to prove their point.")
When people learn that you take panpsychism seriously (more formally, constitutive panpsychism or non-materialist physicalism in my case), they expect you to find minds everywhere - from rocks to trees to digital computers, maybe even the cosmos itself. But (once again in my case), phenomenally-bound minds are peculiar to the biological nervous systems of multicellular animals with the capacity for rapid self-propelled motion. A pleasure-pain axis, agency, and an egocentric real-time world-simulation are hugely adaptive: the challenge is to show how they are physically possible. I don't know how to engineer a mind if we assume classical physics.

[on progress]
Do we need a better understanding of progress?
Should our metric of progress be the happiness of all sentient beings? Or "GDP per capita"?
["Crawford and Cowen also have a specific view of what kind of well-being they are aiming to encourage through progress. It's not happiness – or even the more established metric of "life satisfaction" – instead, their top priority is increasing "GDP per capita".]

[on efilism]
Not all efilists reject HI...
What do efilists think of The Hedonistic Imperative? ("HI is a manifesto which outlines how to abolish suffering with the aid of technology. So instead of eradicating life, how about we eradicate suffering.")

[on friendly AI]
Artificial Uninintelligence
(DP and Kim Solez critique Stuart Russell's lectures)

[on algorithms]
Soulful abstracta?
Can algorithms sufer?
Can abstract objects have PTSD? The world has frightful suffering. I worry about (some) s-risks. But the well-being of algorithms is Alice-in-Wonderland.

Dario, you're right about not taking moral risks - at least if they can be avoided. One point I'd like to add though. Phenomenal binding - including phenomenally-bound pains and pleasures and the diverse intentional objects they infuse - is functionally genetically adaptive. The post-Cambrian success of the animal kingdom depends on it. So in that sense, I'm a functionalist about consciousness. I just don't believe classical Turing machines and classically parallel connectionist systems can do it. Anyone who believes digital computers or connectionist systems can support phenomenal binding - and thus potential suffering- really needs to explain how it's possible.
Otherwise we're left with the possibility of spooky "strong" emergence.

[on aphantasia]
Eliminative materialists about consciousness tend to be high-IQ male perceptual direct realists with aphantasia:
Aphantasia
("What's it like to be mind blind")

David, my phenomenal consciousness isn't something abstract; it's something I'm undergoing right now. What's more, 650nm electromagnetic radiation incident on my retinas is neither necessary nor sufficient for my experience of phenomenal redness. I experience phenomenal redness in my dreams. I can close my eyes and imagine redness now. If a neurosurgeon stimulated the V4 colour centre of my ventral occipital lobe, I might once again experience the raw feels of phenomenal redness. Yes, for evolutionary reasons, it's normal to use the functional-relational language of perception. But the retinal story is a red herring, so to speak...

[on wireheading]
Elon Musk’s Neuralink 'Brain Chip' could give users orgasms on demand
("The revolutionary device could be implanted in human subjects by the end of this year, and some experts believe it could completely revolutionise human sexuality")
Aaron, trust him? No! But the vanity of billionaires can be tickled. The rich and powerful are also prone to conspicuous displays of competitive male altruism (cf. the Forbes list of Biggest Givers). If Elon Musk can be encouraged to think Neuralink can fix the problem of suffering and let him go down in history as the Greatest Benefactor of All Sentience, well...what a smart way to get one over on his rival Jeff Bezos.

Ekaterina I think the option of perpetual well-being via electrodes should be offered to all victims of chronic pain and depression who aren't helped by existing therapies. But "wireheading" is a last resort. Wireheading is not a viable option for a whole civilization - wireheads don't want to raise baby wireheads. Uniform bliss extinguishes information-sensitivity - critical insight, social responsibility, meaningful relationships and intellectual growth.

Instead, I think our goal as transhumanists should be hedonic recalibration - raising hedonic set-points. All_ the things that you care about can be conserved if you have a higher hedonic default setting – with the difference that your quality of life will be much richer. In the long run, we can be more ambitious. We can engineer life based on gradients of (literally) superhuman bliss. But superhuman bliss is not as morally urgent as fixing the problem of suffering. Gradients.com
Beyond Humanism (2020)

[on functionalism]
Another wonderful video from Andrés! Dan, just a few comments to add. The physical impossibility of classical digital computers solving the binding problem and “waking up” doesn’t by itself entail the falsity of functionalism. After all, animal minds and the world-simulations we run are phenomenally bound. Binding is functionally highly adaptive - as rare deficit syndromes like integrative agnosia illustrate. The challenge is to show how animal minds – supposedly just a pack of decohered neurons - achieve the classically impossible.
I share your impatience with quantum woo (“consciousness collapses the wavefunction”, etc). But if neither classical nor quantum physics can explain binding, then we face the spectre of dualism.
For what it's worth, I bite the bullet and explore the quantum-theoretic version of the intrinsic nature argument:
Quantum mind
It’s crazy. I’m not convinced the alternatives are saner.

[on wireheading]
Genes, drugs or electrodes? Bliss should be our birthright, but shifting the Overton window is a daunting challenge.
Pleasure Direct
("What Are the Ethics of an Implant That Delivers Pleasure Directly Into Your Brain?")

[on cryonics and cryothanasia]
Immortalists Magazine on cryonics and cryothanasia:
Cryonics and cryothanasiaDeath Defanged
My little piece is here:
Death Defanged?
Critics might say a negative utilitarian is organising a suicide cult. Heaven forbid.

[on negative utilitarianism and eliminativism]
Why do you reject negative utilitarianism?
Is child abuse morally defensible if the beneficiaries derive enough pleasure?
The Ones Who Walk Away From Omelas (Wikipedia) pdf
"Yes", say classical utilitarians.
"No", say negative utilitarians.
Negative utilitarians want you to have fun - gradients of superhuman bliss in my case! - just not at anyone else's expense.

Rob, you remark, "First front: I claim that rejecting phenomenal consciousness doesn't matter that much, and we can still ground morality in (the p-zombie equivalent of) valenced experience.”
Ok, I'm reeling. Not rhetorically, but in the sense of being intellectually out of my depth trying to interpret this. If I'm in agony, and a consciousness anti-realist tells me that he doubts my phenomenal experience is real - and even if it does exist, then my agony doesn't matter that much...well, I'm floundering. Consciousness realists and anti-realists talk past each other – I guess we are trapped in Kuhn's incommensurable conceptual schemes. And consciousness (anti-)realism can't be quarantined from other issues. Consciousness (anti-)realism infects ethics and almost everything else. Thus I simply don't know how to conduct a discussion of negative utilitarianism that disputes the phenomenal reality of suffering: it’s the raison d’être of NU! So I'd have to disagree with your remark. "The specific claim I'm denying is that *if* anti-realism about phenomenal consciousness turns out to be true, we should therefore upend every moral claim".
You speak of the zombie equivalent of valence experience. There's no such thing! Compare a robot banana-picker. The silicon robot has been programmed to prioritize bodily integrity over ripe banana odour-detection. Poetically, we can say the robo banana-picker "cares" more about its chassis than a banana. But if humans need to repair its bodywork, there's no moral need for us to administer anaesthesia because the robot isn’t a subject of experience - it’s a zombie. So there's no need to respect a zombie’s metaphorical "preferences". The robot doesn’t literally, phenomenally care about anything - pain, bodily integrity, bananas or anything else. Without consciousness, nothing matters.
If we could agree on the reality of phenomenal suffering, then perhaps we could fruitfully discuss whether negative utilitarians (and Buddhists etc) are right to believe suffering prevention and mitigation is of overriding moral importance. Do all the other things human and nonhuman animals ostensibly care about derive their (dis)value from the pain-pleasure axis? But with our different background assumptions, alas I'm stumped...

An Eliminativist Theory of Consciousness
Jacy, in your paper you remark, "We crave a unique, unsolvable mystery at the core of our being”. NO! Sure, some New Agers and religious mystics love mysteries. But scientific rationalists (like me) hate mysteries. Mystery-mongering doesn’t underpin the Hard Problem. Rather, if materialist physicists and chemists correctly understand the properties of atoms and molecules, then subjective consciousness shouldn’t exist. None of it. Everyday life would subjectively be no different from being dreamlessly asleep. And here’s the rub. Whether I’m awake or dreaming, consciousness is all I’ve ever known! Everything else above-and-beyond my own consciousness is inference and speculation, including my belief that I’m a fairly typical skull-bound animal mind running a fairly typical egocentric world-simulation. So your claim that my consciousness doesn’t exist - that it's not like anything to be me - poses interpretational challenges, for me, at any rate!

Jacy, agreed, some folk want to feel special. I was just querying whether this desire is what drives most discussions of the Hard Problem.
Either way, the subtle phenomenology of this particular self-referential thought differs from, say, the brilliant blue sky I'm currently experiencing above my body-image within my world-simulation. But phenomenal blueness is a distinctive property of one kind of consciousness - a property that congenitally colour-blind people lack. Whether one is lucidly dreaming or wide awake, one knows that phenomenal blueness doesn't exist in the mind-independent world; but phenomenal blueness a real, spatio-temporally located property of my conscious mind - and countless skull-bound minds like mine I presume too!

David, you remark, "This is the very thing that is being disputed. A number of thinkers seem to have no trouble believing that our experience of self-reference can arise out of the functional interactions of materials"

Recall that self-reference / indexical thought / Descartes' cogito are just one category of conscious experience. We could equally invoke unreflective kinds of consciousness such as, say, uncontrolled panic or orgasmic ecstasy. My point was just that on standard materialist assumptions, i.e. quantum field theory describes fields of insentience, no one has the foggiest idea how to derive the properties of consciousness from neurobiology and thus ultimately physics. Hence the Hard Problem. Levine's "explanatory gap" is unbridged. Contrast life, where (with a bit of handwaving and a nod to the decoherence program) molecular biology and quantum chemistry can be used to derive the properties of information-bearing self-replicators from the underlying physics. No such joy with phenomenal experience. The empirical (“relating to experience") evidence ought not to exist. Materialism is in dire straits.

So how can physicalism and the (ontological) unity of science be saved? Eliminativists like Jacy explore one route, while constitutive panpsychists / non-materialist physicalists a diametrically opposed route. Eliminativists think that if our best theory of the world, scientific materialism, has no place in its ontology for consciousness, then in some unfathomable sense, it's an illusion, and the illusion itself is illusory. By contrast, constitutive panpsychists / non-materialist physicalists are consciousness realists who conjecture we radically misconceive the intrinsic nature of the physical.

Both strategies for dissolving the Hard Problem strike me as incredible. Only one is empirically adequate.

What’s it like to be a bat or an eliminativist? Jacy, sorry, I’m floundering. I just don’t “get” Semanticalism. Sometimes I wonder if eliminativists are just perceptual direct realists with Aphantasia). Consciousness, reflective and unreflective, is all I’ve ever known - and you’re telling me my consciousness doesn’t exist. From a headache to a melody to a dream, I don’t merely have the intuition that I’m conscious, but rather (to quote Galen Strawson) “the having is the knowing". Moreover, I can radically alter my consciousness, too, by taking different consciousness-altering drugs. My consciousness is the only direct evidence I have for anything else at all, including the mind-independent reality of modern science. By contrast, I infer on theoretical grounds that I’m probably not a lab-grown minibrain or a Boltzmann brain (etc). And alas there really is a fact to the matter that I’m undergoing the nasty raw feels of a headache. Materialist science doesn’t know how to derive those nasty raw feels of pain from neurobiology and physics as commonly understood. What’s more, there really is an objective fact of the matter whether nonhuman animals too undergo the ghastliness of phenomenal pain - not just functional nociception, but the nasty raw feels of pain. The Cartesians were mistaken to think dogs were insentient automata emitting mere distress vocalizations. Animal minds really are conscious subjects of experience. Rocks, vegetables and the stock market (etc) are not conscious subjects of experience. Subjective experience is an objective spatiotemporally-located property of reality that science must accommodate on pain of failing the test of empirical adequacy.

[Jacy, I was about to apologize again for causing you… frustration. Close line-by-line engagement would be more useful. But feeling frustrated is a state of consciousness you’re denying – a thought that induces another state of consciousness in me: confusion!]

David, I'm currently experiencing one kind of phenomenal consciousness, the colour green. I experience green both when awake and dreaming. The greenness is normally bound to other modes of phenomenal consciousness - the everyday objects in my world-simulation. I assume I'm fairly typical. Questions of the (in)effability or (un)communicability of experience are best kept separate from the reality of phenomenal consciousness itself. For example, most humans can experience millions of hues, but our language has only a few dozen colour terms. Either way, the reason that the colour-blind kid below is emotionally overwhelmed is he experiences a new mode of consciousness, phenomenal colour:
Kid sees colour
We could go on to tackle the Hard Problem, the binding problem and so on. But until we agree that there is a phenomenon that stands in need of explanation, exploring solutions will be of no interest to you.

Some NUs want to turn the world into equivalent of paperclips, but can they solve the AI alignment problem?
Superintelligence and the FLI

Minimise, prevent and (eventually) abolish suffering is the overarching ethical framework for an NU. And achieving this neo-Buddhist vision involves working with people from very different ethical traditions and accommodating - insofar as is humanly possible - their diverse values and preferences. Hedonic uplift via set-point recalibration is the nearest I can envisage to ending suffering while respecting the (often conflicting) values and preferences of others. Few people familiar with the concept of the hedonic treadmill are opposed to enjoying a higher default hedonic set-point. But how best to raise everyone to this level?

[on effective altruism]
From a comment by Question Mark,
"Since biological life will almost certainly be phased out in the long run and be replaced with machine intelligence, AI safety probably has far more longtermist impact compared to biotech-related suffering reduction. Still, it could be argued that having a better understanding of valence and consciousness could make future AIs safer."
The case for phasing out the biology of suffering
My thoughts on longtermism:
DP on Longtermism

Cruel and unfair. But witty:
The EA Mindset ("This is an unfair caricature/ lampoon of parts of the 'EA mindset' or maybe in particular, my mindset towards EA")

Human exinction could be in our future
"I am an optimist," Sandberg says, "the future could be awesome. I think the world is actually really good. And it could be even better, much better, which means that we have a reason to try to safeguard the future."
Oh to run Anders' world-simulation.

Adam, if novelty is desired as well as happiness, perhaps induce perpetual jamais vu:
Jamaos vu
Alternatively, after we have upgraded our reward circuitry to utilitronium, you can explore a rich variety of exotic state-spaces of consciousness until Doomsday:
Post-Suffering Life
Intelligence? IMO, it's a necessary evil. On fairly modest ethical assumptions, our long-term goal should be blissful ignorance of the horrors of Darwinian life that the universal wavefunction encodes. This goal of invincible ignorance follows whether one is a classical or negative utilitarian.
Zombification? As a hardcore negative utilitarian, I think we should enshrine in law the sanctity of sentient life. The future belongs to fanatical life-lovers. In order to get rid of suffering, we need the widest possible religious and secular coalition of allies. So talk of button-pressing, efilism, “hard” antinatalism and so forth at best distracts from the goal of ending suffering. Let's not alienate people.
On the other hand, If Eliezer is right, then upshot of machine superintelligence will perfectly align with NU values. The AI alignment problem rises only for folk who believe they are wiser than superintelligence. (Not me!)
As you know, I'm an AGI sceptic, but that's another story.

[on DALL·E 2]
with thanks to Adam Ford:
Gustav Klimpt painting of sentient superintelligence doing hard core physics to determine if 'consciousness discloses the intrinsic nature of the physical'
“Gustav Klimpt painting of sentient superintelligence doing hard core physics to determine if "consciousness discloses the intrinsic nature of the physical”

[on suffering and post-Darwinian life]
Can one serve too masters?
The Two Noble Pursuits
"At this point it begs the question ‘if we can’t get rid of suffering while remaining conscious knowers of the world, then should we want to get rid of all suffering?’ If throwing out all of the bathwater requires throwing out the baby, then isn’t it better to keep some of the bathwater (suffering) in order to keep the baby (conscious knowing)?"

We can't get rid of suffering from reality - timelessly conceived. But we (and our "smart" machines) can prevent the physical signature of suffering in our forward light-cone. And once we have discharged all our ethical responsibilities to prevent its recurrence, wouldn't we do better to forget its very existence?
To understand suffering is oneself to suffer - horribly.

The risk of aiming ultimately for blissful ignorance - i.e. knowledge and understanding only of paradise - is premature defeatism about (the prevention of) suffering elsewhere. Therefore giving knowledge equal (or even supreme) value might seem safer. [Although I lean to Rare Earthism, I sometimes give the example of a VR-living civilisation in our galaxy that engineers life based entirely on gradients of bliss but mistakenly believes they are alone when they could have rescued us.]
But maybe our responsibility to the future is to make Darwinian life not just impossible but inconceivable.

Sociologically, I suspect NU and SFE is going to die out with the end of suffering. What's the best way to ensure hedonic sub-zero experience can never recur in our forward light-cone? Once again, we need to take into account not just what's technically feasible, but also what's politically and societally viable in a world where (I assume) mastery of the pleasure-pain axis means everyone is innately (super)happy and an ardent life-lover. There are just too many unknowns here to speak confidently. Despite my scepticism of "AGI" as currently conceived, artificial robots will presumably play a huge role in cosmic stewardship - and AI could be used to prevent pain-ridden Darwinian life spontaneously arising again in our galactic super-cluster.

This response probably makes me sound less NU than you, but really it's (in part) a question of power dynamics.

Rather than "negative utilitarian", I sometimes call myself a secular Buddhist - despite feeling a bit of an impostor and knowing the inadequacy of the label. How can the different Buddhist traditions be more effectively infused with the technical potential of biotech and AI to deliver a world without suffering? I sometimes quote the Dalai Lama ["If it was possible to become free of negative emotions by a riskless implementation of an electrode - without impairing intelligence and the critical mind - I would be the first patient."(Society for Neuroscience Congress, Nov. 2005)]. But alas we don't hear many Buddhists calling for genome reform and reprogramming the biosphere - the only (non-apocalyptic) way I know to fix the problem of suffering.

[on psi]
I'm sceptical of scientific triumphalism. Not least, materialism is inconsistent with the entirety of the empirical ("relating to experience") evidence - we ought to be zombies! Nonetheless, the existence of psi would be inconsistent with the extremely well tested Standard Model, to which all the special sciences reduce. Physicalism (but not materialism) offers the best explanation for the technological success story of modern science. Quantum theory is extremely mathematically rigid - you can't just start toying around with the formalism to make room for psi powers.
The Standard Model ("The Standard Model of particle physics: The absolutely amazing theory of almost everything")

[on identity]
Bad news for empty individualists?
Future You
("How thinking about 'future you' can build a happier life")

[on eugenics]
Transhumanists and effective altruists face a dilemma. If we don't reform the genome, then the problem of suffering can't be fixed. Yet the accepted name for genome reform is taboo:
eugenics.org
Should we aim for linguistic reappropriation?
Linguistic reappropriation
eugenics.org
Or build a stronger brand? (what exactly?)
And will our critics play ball?
Is eugenics moral?

Jessica, futurology via extrapolation is treacherous. So you're right to raise the issue. I was simply noting how elevated mood predisposes to being an active citizen and internalizing the role of a dominant "alpha". Low mood predisposes to keeping one's head down. The social structure of a future civilisation where everyone enjoys life animated by gradients of bliss is speculative. I was just noting we can't assume that hedonic uplift is the recipe for Brave New World. If anything, society may face the opposite problem in the wake of a biohappiness revolution. What happens when exuberant lifelovers aren't willing to slot themselves into traditional status hierarchies?

[on MDMA]
Ecstasy and Honesty
MDMA consciousness is beautiful. I'd evangelize if we knew how to sustain it. Rick Doblin has done some fantastic work. IMO, MDMA-assisted psychotherapy is more promising than psychedelics. But there are zillions of pitfalls to navigate....
The Trials of Rick Doblin
("He revolutionized the way we view MDMA-assisted psychotherapy. But what does the research actually show?")

[on language]
The ying and yang of language
Language and gender
("These Words That Women Know Better Than Men And Vice Versa Will Make You Question Your Grasp Of The English Language")

[on Kardashev scales for Pleasure]
well, if the hedonic scale of Darwinian life is, conventionally, -10 to 0 to +10, we can imagine, say Hedonic Type 1 civilization with a range of 0 to +20, a Hedonic Type 2 with a range of, say, +20 to +40, and a mature Hedonic Type 3 Civilization with, say, +80 to +100. But I don't pretend to know the likely depth or shallowness of hedonic range of future civilization. And will the information-signalling role of the pleasure axis be maintained indefinitely or will decision-making be offloaded to zombie AI? Will states of mega-bliss continue to be adulterated with the usual intentional bric-a-brac or converted to pure hedonium? Will we ever launch a hedonium / utilitronium shockwave?

[on perception]
The Humunclus Problem
Homunculi are real. Consider a lucid dream. When lucid, you can know that your body-image is entirely internal to your sleeping brain. You can know that the virtual head you can feel with your virtual hands is entirely internal to your sleeping brain too. Sure, the reality of this homunculus doesn’t explain how the experience is possible. Yet such an absence of explanatory power doesn’t mean that we should disavow talk of homunculi.

Waking consciousness is more controversial. But (I’d argue) you can still experience only a homunculus - but now it’s a homunculus that (normally) causally do-varies with the behaviour of an extra-cranial body.

[on gene drives]
The sci-fi technology tackling malarial mosquitoes
The same technology of gene drives could be used to engineer a hyperthymic biosphere:
Wild Animal Happiness

[on non-materialist physicalism]
consciousness
So-called constitutive panpsychism is physicalist. Indeed, if the mathematical formalism of physics is just transposed to an ontology of qualia, then we have non-materialist physicalism - which is idealism under another name. I explore the conjecture that mathematical physics describes patterns of qualia, though only animals have minds. The conjecture that your mind and the phenomenal world-simulation it runs consist entirely of "cat states" is crazy - I don't really expect people to take the idea seriously. But if true, then non-materialist physicalism solves the Hard Problem of consciousness, the binding problem, the problem of causal efficacy and more besides. And it's experimentally falsifiable....
David Pearce has a proposal

The reason that physicists spends billions of dollars on particle accelerators exploring exotic energy regimes is their belief that - discounting the hypothetical dark matter and dark energy posited by cosmologists – the Standard Model and General Relativity are successful in describing all currently accessible physical phenomena. All the “special sciences” (molecular biology, chemistry etc) reduce to physics. Compare Sean Carroll ("The Laws Underlying The Physics of Everyday Life Are Completely Understood") or Glenn Starkman ("The Standard Model of particle physics: The absolutely amazing theory of almost everything")

The only exception to this tale of triumph is consciousness: the Hard Problem. Is subjective experience just an anomaly - or the key to the plot?

Lepandas, allow me to offer a bit of background. The Hard Problem of consciousness arises only because we normally make a plausible metaphysical assumption. The intrinsic nature of the physical, the mysterious “fire" in the equations, is non-experiential. The mathematical machinery of quantum field theory describes fields of insentience.

However, in recent years, a minority of researchers (e.g. Galen Strawson, Philip Goff) have proposed dropping this commonsense assumption. Instead, perhaps the intrinsic nature of the universe’s fundamental quantum fields doesn't differ inside and outside your head(!) What makes animal minds like us special isn’t experience per se, but rather, its phenomenal binding into virtual worlds of experience - like the phenomenal world-simulation your mind is running right now.

This conjecture is sometimes called constitutive panpsychism. In my view, the term is misleading because "panpsychism" suggests property-dualism, i.e. consciousness is inseparably attached to all fundamental physical properties. So I prefer philosopher Grover Maxwell's term, “non-materialist physicalism”, because according to this conjecture, experience discloses the essence of the physical. We should simply transpose the entire mathematical apparatus of modern physics onto an idealist ontology. Yes, crazy! But see: Crazyism

However, non-materialist physicalism faces a huge technical challenge. Even if consciousness is fundamental to the world, then why aren't we micro-experiential zombies? For instance, consider the five hundred million or so neurons of your enteric nervous system. The “brain-in-the-gut” is a fabulously complicated information-processing system. But even if its individual neurons are membrane-bound micro-pixels of consciousness, your enteric nervous system is not a unified subject of experience, a person. So why is your awake mind-brain so radically different? If textbook neuroscience is correct, then you should be a micro-experiential zombie too - as you are when dreamlessly asleep – just c. 86 billions membrane-bound “pixels” of mind-dust. If neurons are really discrete, decohered classical objects, as crude neuroscanning suggests, then phenomenal binding should be impossible. This partial "structural mismatch" between phenomenology and neurobiology leads David Chalmers to wonder if we must consider dualism. Dualism is a desperate last resort. Presumably, we want - if at all possible - to conserve physicalism and the ontological unity of science. The “Schrödinger’s neurons” proposal I explore is designed to test the conjecture that a perfect structural match exists. My best guess is that interferometry will reveal such a perfect structural match, not in four-dimensional space-time, but in the fundamental high-dimensional space required by the dynamics of the wavefunction.
[For a nice introduction to wavefunction realism, maybe see “The World in the Wave Function: A Metaphysics for Quantum Physics" by Alyssa Ney]

Anyone familiar with quantum decoherence will recognise that a “Schrödinger’s neurons” conjecture is (extremely!) far-fetched. Not merely is the credible effective lifetime of neuronal superpositions in the CNS less than femtoseconds. Also, the conjecture that you experience nothing but individual “cat states” inverts the measurement problem of quantum mechanics as normally posed.

Common sense could very well be right. But I’m still curious. Let’s use experiment rather than human intuition to lay the possibility to rest.

If you haven’t watched the video already, you might enjoy Phil Goff vs. Sean Carroll. Goff defends consciousness fundamentalism: experience discloses the intrinsic nature of the physical. The strongest technical argument against constitutive panpsychism / non-materialist physicalism is the binding problem. The phenomenal binding / combination problem is the reason why Goff used to reject constitutive panpsychism / non-materialist physicalism as a possible solution to the Hard Problem (cf. "Why Panpsychism doesn't Help Us Explain Consciousness" by Philip Goff.) Why aren't we just micro-experiential zombies? Indeed the (ostensible) "structural mismatch" between phenomenology and neuroscience leads David Chalmers to wonder if we must abandon monistic physicalism.

Naively, we’re discussing a "philosophical" rather than scientific question. Either you take constitutive panpsychism / non-materialist physicalism seriously as a solution to the Hard Problem of consciousness or you don't. How could we ever hope to test the proposal? However, if consciousness discloses the intrinsic nature of the physical, then we must consider the nature of reality not merely at ludicrously small distance scales, but ludicrously short temporal resolutions too. What will molecular matter-wave interferometry reveal? I don’t know, but the protocol of a “Schrödinger’s neurons” experiment is designed to find out. If the phenomenal binding of our minds is not a classical phenomenon, then the non-classical interference signature will tell us.

Finite light, thank you. Yes, any physicalist theory of consciousness must explain qualia. Science needs to account for the existence, binding, diversity and causal efficacy of consciousness. You speak of "levels". For sure, human convenience dictates carving up reality into multiple layers - physics, chemistry, molecular biology, psychology, sociology and so forth. But ultimately reality has only one "level": physics. Ultimately, we need to “cash out” everything in terms of quantum field theory or its successor. For if we can't derive chemistry, biology (etc) from the underlying physics - at least in principle - then spooky “strong” emergence is real. By analogy, imagine if it weren't possible - again in principle - to derive the properties of software running on your desktop PC from the execution of the underlying machine code. “Strong” emergence would be magic.

Anyhow, I explore a version of what philosophers call the intrinsic nature argument. Either the conjecture is too crazy for words or a live option – take your pick – but for a nice introduction see Philip Goff’s "Galileo's Error" (2019)
(My review: Galileo's Error?)

Just one correction to your comment above. I'm not arguing for a dynamical collapse theory of QM. Rather, whether you experience e.g. a live cat or point-like particle incident on a perceptually experienced laboratory screen in a double-slit experiment, all you ever experience are individual "cat states", i.e. neuronal superpositions. As investigators from William James to David Chalmers have recognized, phenomenal binding is [classically] impossible. On a “Schrödinger’s neurons” conjecture, only the ubiquity of the superposition principle of QM allows you to experience definite outcomes. Otherwise, you’d just be Jamesian mind-dust, incapable of perceiving anything at all. Ill-named human “observers” tend to conflate (quantum) vehicle and its (subjectively classical) content.

Yes, crazy stuff. Heaven knows if it’s true. We won’t know until we experiment to find out.

I agree with you. Consciousness is adaptive – or rather, phenomenally-bound consciousness is adaptive: microexperiential zombies would soon starve or get eaten. But the fact that phenomenally-bound consciousness is fitness-enhancing doesn’t explain the phenomenon in any deep sense. After all, telepathy and precognition would be fitness-enhancing too, yet psi phenomena are physically impossible. Yet the same ought to be true of consciousness; a few eliminative materialists bite the bullet. If (materialist) physicists and chemists correctly understand the properties of matter and energy, then subjective experience should be impossible. And if (materialist) neuroscientists correctly understand the properties of the central nervous system as a pack of discrete classical neurons, then phenomenally-bound subjective experience should be impossible too.

The inadequacy of adaptive explanations is illustrated by how your molecular duplicate assembled from scratch – or the notorious “bran-in-a-vat” – would presumably be conscious just like you too. In other words, consciousness is an intrinsic property of configurations of matter and energy irrespective of whether these states have been harnessed by natural selection to play a computational-functional role in naturally evolved biological organisms.

All the options seem absurd. I don’t pretend to know if non-materialist physicalism is true. But its bizarre empirical predictions make the theory experimentally falsifiable. I’m probably mistaken.

If I understand you correctly, you are assuming a "dynamical collapse" theory. I don't. Indeed in my view, only the fact that the superposition principle of QM never breaks down allows you to experience a definite outcome in a phenomenal world-simulation where it does: How can we best resolve the problem of definite outcomes in quantum mechanics?

I say a bit more about interpretations of QM e.g. here: Interpretations of QM. But I think you - and any investigator who posits a nonexperiential “fire" in the equations - need to explain precisely why we aren't zombies. And likewise, any investigator who assumes that we are pack of decohered classical neurons needs to explain precisely why we aren't - at most - micro-experiential zombies. And above all, I think anyone with a theory of consciousness needs to focus on experiments that will (dis)confirm their theories to the satisfaction of proponents and critics alike. No worries if you think a "Schrödinger's neurons" proposal is crazy: I do too!

Well, if non-materialist physicalism is true, i.e. if experience discloses the intrinsic nature of the physical, then "p-zombies" are unphysical. Theories that assume the "fire" in the equations is non-experiential must explain why we aren't p-zombies. They fail. Hence the Hard Problem - a euphemism for the inconsistency of the ontology of (what’s normally reckoned) our best theory of the world, scientific materialism, with the empirical evidence. I believe physicalism can be saved - but not materialism.

Medbud, first, thank you for the "Consciousness and the fallacy of misplaced objectivity" paper link. I essentially agree with its critique of the limitations of behavioral, functional, and neural-correlates approaches. Alas, as it stands I don't think IIT solves the Hard Problem, the binding problem, or the problem of causal efficacy. If we make the standard materialist assumption that world's fundamental quantum fields are non-experiential, then Levine's "explanatory gap" stays unbridged. For we haven’t derived sentience from insentience. Likewise, if we make the standard neuroscientific assumption that neurons are discrete, decohered, membrane-bound classical objects, then we haven't derived the properties of our phenomenally-bound minds and the world-simulations they run. No amount of functional integration or computational complexity can transmute classical Jamesian "mind-dust" into phenomenal objects or unified subjects of experience. At most, we should be micro-experiential zombies (Phil Goff’s term).

Anyhow, I don't for a moment expect most researchers to take seriously the conjecture I explore - the quantum-theoretic version of the intrinsic nature argument. I find a "Schrödinger's neurons" conjecture crazy too. But as I (very) belatedly came to realise, it's experimentally falsifiable. And if I'm confounded - as I probably will be - well, there is no disgrace in being proved wrong.

Perhaps you may want to unpack the meaning of (2). Have you in mind the speculative idea that the universe is a vast digital computation device? Or is the world "computational” merely in the sense that reality can formally be described by the unitary and deterministic Schrödinger evolution?
IF non-materialist physicalism is true, i.e. experience is the intrinsic nature of the physical, the "fire" in the equations, then mathematical physics describes patters of qualia.
Thus only the physical is real (1). Its evolution is described by the universal Schrödinger equation (2). Only the physical has causal efficacy. Hence (3).
Yet does experience really disclose the intrinsic nature of the physical?
I don't know...
Galileo's Error

The conjecture that the universe/multiverse is conscious - universal mind - is often confused with the conjecture that the universe/multiverse is consciousness, i.e. non-materialist physicalism. Thus when dreamlessly asleep, your fundamental fermionic and bosonic fields may or may not be experiential - I don't know! - but there is no phenomenally-bound subject of experience. The same is true (as far as I can tell) for the universe/multiverse as a whole.

Maroš, relative to standard scientific background assumptions, your objection is spot-on. Normally, we assume two separate mysteries. Why does the physical universe exist at all? And how do physical matter and energy give rise to consciousness? (i.e. the Hard Problem of materialist metaphysics). However, if the intrinsic nature argument is sound, then consciousness discloses the essence of the physical. So there’s only one fundamental mystery: why does anything exist? Both Andres and I take seriously that conjecture that mathematical physics describes patterns of qualia. Note that such non-materialist physicalism doesn’t entail the animist notion that rocks or trees or connectionist systems or classical Turing machines are phenomenally-bound subjects of experience, i.e. minds. Minds and our world-simulations are an adaptation of organisms with the capacity for rapid self-propelled motion, i.e. animals. So how are minds possible?
What happens when you “wake up” from a dreamless sleep?

Andrei, thanks. Any adequate theory of consciousness must explain:
1) the existence (the Hard Problem)
2) phenomenal binding
3) causal efficacy
and
4) diversity (the palette problem) of consciousness.
And critically, any scientific theory of consciousness must make novel, precise, experimentally falsifiable predictions that proponents AND critics can agree will (dis)confirm the theory:
Quantum Mind (Wikipedia)

So in answer to your questions:
1) Non-materialist physicalism is monist. Only the physical is real. Only the physical is causally effective. Formally, the world is exhaustively described by the equations of mathematical physics – presumably the universal Schrödinger equation if wavefunction monism is true.

2) “Non-materialist physicalism" and "physicalistic idealism” are just stylistic variants of the same idea – the intrinsic nature of the physical, the “fire” the equations of physics, is experiential. Quantum field theory describes fields of sentience. See e.g. Phil Goff for a recent convert to the intrinsic nature argument:
On this story, what makes animal minds special isn’t experience per se, but rather phenomenal binding into virtual worlds of experience - like the phenomenal world-simulation your mind is running right now. Non-materialist physicalism does not propose that rocks or trees or digital computers - or indeed the whole cosmos - are subjects of experience.

3) Physicalism is the conjecture that no “element of reality” is missing from the mathematical formalism of physics – the discipline on which everything else (chemistry, biology etc) supervenes:
Idealism is the conjecture that the intrinsic nature of reality is experiential
Most physicalists make the extra metaphysical assumption that the “fire” in the equations of physics is non-experiential. The upshot is "materialist" physicalism – "materialism" for short. Non-materialist physicalists drop the metaphysical assumption. Fields of metaphysical gunk should go the way of luminiferous aether. The intrinsic nature of the world's fundamental quantum fields is no different inside and outside your head.

To be stressed: no one who joins this group need sign up to any of the weird speculations of some the admins.
Thus you can believe we should use biotech to abolish suffering and be an “orthodox” materialist – although any talk of “orthodoxy” in questions of consciousness is strained.
Science is baffled by consciousness.

[on S-risks and malevolent actors]
Magnus, as I view it, phasing out the biology of suffering is intimately linked to the end of malevolence.
I know we sometimes worry that temperamentally happy folk downplay the problem of suffering and the overriding moral urgency of its minimisation. Suffering just isn’t in their life narrative. Sometimes one wants to say to ardent life lovers, bluntly: why can’t you be more compassionate?! But one thing that happiness doesn’t do is embitter people, or breed malevolence - quite the opposite. Contrast the effects of suffering. Yes, sometimes suffering can deepen compassion; but frequently, suffering breeds resentment, nihilism and misanthropy. [“It is not true that suffering ennobles the character; happiness does that sometimes, but suffering, for the most part, makes men petty and vindictive.” ― W. Somerset Maugham, The Moon and Sixpence] In a world in which we all enjoyed good health, bringing back ancestral horrors would be inconceivable.

This reply makes it sound as though I’m unworried by s-risks. But here we come to info-hazards perhaps best passed over – info-hazards that won’t arise if we implement the WHO mission of good health for all.

[on HI promotion]
paradise engineering paradise engineering
Hedonium.art

[on phenomenal binding]
Three distinct questions:
1. What is the phenomenal binding / combination problem?
2. What is the computational-functional power of binding in biological minds? Why does binding matter? How effectively can binding be computationally emulated by zombie information-processors that can’t bind, e.g. classical Turing machines?
3. What are possible solutions to the binding problem?
(e.g. Chalmersian dualism, quantum mind, topological segmentation, etc). Novel testable predictions and experiments?

Andrés explores (3). I do too. But unless you grok questions (1) and (2), then answers to (3) simply won’t be of interest.
In my view, a solution to the binding problem has profound ethical implications for EA, “AGI” and the long-term future of sentience.

[on general intelligence]
Alternatively, the basis of general intelligence is the pleasure-pain axis and the ability to run a real-time, cross-modally-matched egocentric world-simulation that masquerades as the external environment populated by other cognitive agents.
What is the neural architecture of intelligence?

[on mirror-touch synaesthesia]
What's it like to be a mirror-touch synaesthete? Most conceptions of posthuman superintelligence - and alleged machine "AGI" - evoke a super-Asperger rather than hyper-empathetic perspective-taking...
Meet the man who really feels your pain
The ability to understand other perspectives is as much a form of knowledge the ability to do calculus. But I suspect full-spectrum (super-)intelligence will entail choosing to be selectively ignorant - which leads to obvious (and less obvious) paradoxes.

[on Longtermism]
Andres, is more going here than meets the eye? My source is second-hand, so I can't vouch I'm reporting either fairly or faithfully. But if you're a Longtermist funder who believes our first, second and third priority should be combatting x-risks, then research into extreme suffering is potentially dangerous. If people realise how inconceivably bad suffering can be, then they may draw the "wrong" conclusions. What would you do to destroy Hell? There is an unresolved tension in EA between suffering-focused and x-risk folk - and funding is shaping directions accordingly.

What's ironic and doubly frustrating is that fixing the problem of (severe) suffering may be one of the most effective ways to tackle x-risk. A world entirely of ardent life-lovers (like x-risks folk) is safer than a world where millions of people feel trapped in hell - and millions more feel ambivalent and conflicted about life. OK, I don't press this argument because it's motivated cognition on my part. But as far as I can tell, the argument stands.

Kenneth, prescribing opioids can lead to hyperalgesia - and thus even worse pain. Even "safe" nonsteroidal anti-inflammatory drugs (NSAIDs) can lead to more pain (cf. https://www.science.org/doi/10.1126/scitranslmed.abj9954) By contrast, ensuring that all new kids are born with a benign, low-pain version of SCN9A ("gene-drives.com.
Genome reform is longtermist, life-affirming and x-risk-reducing - quite aside from sparing innumerable sentient beings the burden of severe suffering.
So is genome reform an EA cause priority?
Not currently, alas.

Adam, what's the most effective way to mitigate, prevent and eventually abolish suffering? If you're right, then NUs should be trying to engineer a vacuum phase transition, build seed-AI paperclippers, infiltrate x-risk institutes (etc).
I think such an approach would be misconceived. OK, my objection is partly technical: life is now ineradicable. But my main reason for urging life-affirmation is that our best hope of ending suffering is to build the broadest possible coalition of secular and religious support. A biohappiness revolution - presumably in the guise of good health for in the spirit of the WHO - is potentially saleable. Retiring life on Earth will always be a minority view.

Ruth, n a theoretical note. I've long assumed that if/when we phase out the biology of suffering here, then - after multiple safeguards are established - its previous existence will become irrelevant to life in our forward light-cone. There won't be suffering - I've assumed - in the medium or distant future. That's the point of focusing on germline engineering for gradients of bliss, not just symptomatic relief. But this assumption of future irrelevance can be challenged. Maybe (super)intelligent moral agents have a responsibility to launch cosmic rescue missions if suffering sentience exists within our cosmological horizon. Maybe my views on the impossibility of digital sentience are mistaken. This debate should be viewed in the context of "longtermism" in the effective altruist movement - though longtermism means something different to classical and negative utilitarians.

Is there a label for our ethical stance / policy prescriptions that is both accurate and - if not inspiring - at least doesn't alienate?
"Suffering-focused ethics" may turn out to be least unsatisfactory. But it's still enough to trigger dismissive responses from people whom one might hope would be allies - as "The dismal dismissal of suffering-focused views" illustrates.

[on x-risks and s-risks]
1) Happiness is life-affirming, whereas raw suffering is nihilistic - it would rather not exist. Replacing suffering with a more civilised signalling system would – IMO - also banish one of the biggest underlying sources of x-risk as conceived by longtermists. In practice, x-risks folk can be suspicious of suffering-focused ethics because of its seemingly life-denying, nihilistic implications. But the underlying nihilism derives from the suffering, not its messengers. Suffering drains life of meaning; happiness does the opposite. I'm not sure that philosophy of mind and/or neuroscience tells us anything directly about the ethics of asymmetries between suffering and happiness. Intuitively, severe suffering is more terrible than intense pleasure is more wonderful. However, consider the _insanely_ arduous and painful things that sentient beings – both human and nonhuman animals - will sometimes do to obtain a fleeting, fitness-enhancing pleasure like sex. So I think the asymmetry between pain and pleasure is ethical rather than any inherent difference in their comparative intensity.

2) As far as I can tell, classical Turing machines and classically parallel connectionist systems have the wrong sort of architecture to generate phenomenally-bound sentience - a mind. Executing their code faster, or making the code more complicated, or devising smarter training algorithms (etc) can't generate unified subjects of experience on pain of irreducible "strong" emergence. "Strong” emergence is akin to magic. If we live in a world where “strong” emergence is real, nothing is lawfully forbidden. So philosophers and scientists don't like the idea. What’s more, the phenomenal binding of biological minds is exceedingly computationally powerful, as rare partial failures of both local and global binding in humans illustrate. In other words, I’m a functionalist – just not a Turing machine functionalist. Either way, it's worth distinguishing artificial (1) sentience, (2) phenomenally-bound sentience, and (3) phenomenally-bound sentience with a pleasure-pain axis. Presumably, only (3) inherently, non-instrumentally matters. I argue that phenomenally-bound sentience with a pleasure-pain axis is (probably) peculiar to biological minds.

3) I think we already know enough technically – but not yet sociologically or politically - to sketch out blueprints for a happy biosphere. I have no idea how to create a digital mind. In my view, attributing a mind to Deep Blue or AlphGo or GPT-3 (etc) is an anthropomorphic projection on our part.

4) Just as research into some kinds of x-risks increases their likelihood – aspiring world-destroyers should presumably burrow deeply into x-risks institutes - research into some kinds of s-risks should be discouraged lest it makes their occurrence more likely. Such risks - as I conceive them - will effectively vanish if we fix the problem of suffering by making its occurrence physically impossible. However, I understand that researchers with different conceptions of consciousness/AGI/s-risks may not share this view. How can incisive critical thinking be encouraged if one is simultaneously trying to suppress some kinds of knowledge as too dangerous? I don’t know.

5) IMO, we should anticipate artificial biological minds, artificial hybrid cyborg minds, _maybe_ one day artificial nonbiological quantum minds, but not digital minds nor connectionist minds. Questions about minds are different from what digital computers and connectionist systems can and can’t do: the upper bounds to zombie intelligence are unknown.

[on the Biohappiness Revolution]
No prospective parent - literally zero - wants to have depressive or pain-ridden children. So fixing the problem of suffering depends, critically, on shifting the Overton window. This isn't a question of the sanctity of life. We should support its enshrinement in law. Humans can’t be otherwise trusted. In my view, ALL responsible prospective parents should consider preimplantation genetic screening, counselling and germline editing for their children to ensure a high pain-threshold, hedonic range and hedonic set-point- in short, lifelong good health. The only long-term solution to the problem of suffering is germline reform.
So how can we win popular support for a reproductive revolution?
Can humanity rise to the challenge?

Giordano, for each of our existing core emotions, we should ask whether we want to retain (1) its "raw feels" and (2) its functional role. With jealousy, for example, we could well dispense with both. But the functional analogue of anxiety will presumably be needed for the indefinite future. Giordano, you remark, "Seneca already taught us that the mind that is anxious about future events is dejected..."But compare the kind of functional anxiety one has, say, playing a computer program at chess. One can spend a lot of time considering potential risks and threats to one’s pieces without being unhappy. Hopefully, life can one day be similar - with the difference that one sometimes wins rather than always loses! [Thought Criminal writes]
@David Pearce Do you have any advice for people who want to make future biohappiness a reality and want to reduce S-risks? I know about the Center on Long-Term Risk and the Center for Reducing Suffering, and have donated to both of them. What else can someone like me do to optimize the far future and reduce the risks of astronomical suffering as much as possible? I think the main obstacles to creating a biohappiness utopia based on gradients of bliss in the long term are game-theoretic, rather than technical. Even if it becomes technologically possible to replace suffering with gradients of bliss, it could also become technologically possible to do the exact opposite. It’s possible that Nash equilibria could emerge that result in large amounts of suffering being produced, even if it’s technologically possible to avoid it. Trying to spread utilitarian and suffering-focused values may therefore be the area where we should push the most."

Thought Criminal, many thanks. In my view, the biology of suffering is like smallpox. Once we’ve got rid of it, we’re never going to bring it back. Hence IMO advocates of suffering-focused ethics (like me and you) should work on promoting blueprints for a biohappiness revolution, presumably under the auspices of the WHO rather than some fringe group.

However, as you suggest, this analysis can be challenged.
For instance, I assume that classical digital computers, classically parallel connectionist systems and silicon (etc) robots can’t solve the phenomenal binding problem
So such information-processing systems don’t support a pleasure-pain axis. Classical computers aren’t a source of s-risk, not directly at any rate. I’m a functionalist, but not a Turing machine functionalist: digital computers are zombies.
But what if I’m wrong? What if digital computers can support phenomenally-bound subjects of experience and maybe astronomical amounts of suffering?
I’ll have to leave it to you whether you judge my dismissal of (non-trivial) digital sentience is compelling. The view that phenomenal binding is nonclassical is controversial. Fortunately, the issue will ultimately be settled experimentally:
Schrödinger’s Neurons
Maybe I'm mistaken. That said, I do worry about some (pre-biohappiness revolution) s-risks:
DP on S Risk
But there’s a huge complication to any debate. As with x-risks, any researcher should ask: does my publicly discussing such risks diminish or exacerbate them? After all, anyone who wants to destroy the world might decide to join an x-risk institute and pick the brightest minds for apocalyptic ideas. The counterpart of such behaviour for s-risk is too depraved for words, but I don’t know if the possibility can be excluded.
[I suppose there’s a possibility (likelihood?) of taking oneself too seriously here. Most intellectuals would probably (privately) relish the role of Dangerous Thinker. But whereas self-importance is harmless, there’s also the morally serious risk of unwittingly saying something catastrophically stupid. S-risk is one of the very few topics on which I self-censor.]

Game theory?
Technically speaking, life can be based (1) entirely on information-sensitive gradients of bliss, (2) entirely on information-sensitive gradients misery, or (3) (most commonly today) a mixture of information-sensitive gradients of pleasure and pain. On the African savannah, being temperamentally hypomanic or hyperthymic was a high-risk, high-reward strategy (cf. the Rank Theory of depression).Elsewhere, I’ve speculated that low mood can be understood the in the context of Dawkins’ “extended phenotype” theory. My ancestor’s ability to make other members of the tribe cowed and depressed could enhance his genetic fitness by giving him more reproductive opportunities (as a dominant alpha). A countervailing tendency here would be the adaptive value of robustness / depression-resistance in fellow tribesmen in battles with rival tribes – one wants strong allies in battle, not depressive milksops. So selection pressure for hedonic tone is complicated. OK, so what about future selection pressure, which is what interested in?
I argue that the nature of selection pressure itself is going to change in the coming era of designer babies:
The Reproductive Revolution
Not least, the ability of prospective parents to preselect alleles and allelic combinations of their kids in anticipation of their likely psychological and behavioural effects will exert strong selection pressure in favour of a genetic predisposition to (super)happiness – and against any predisposition to pain and depression.
Alas, talks and discursive essays are very different from any rigorous game-theoretic modeling.

What should you do?
Well, I don’t know how much of the analysis above you’d endorse?

Compare the WHO commitment to universal health, extravagantly defined as "complete physical, mental and social well-being". What about consent? Should the WHO constitution be amended explicitly to protect the rights of people to experience ill-health? Or is such a clause redundant? That said, issue of consent does crop up as an objection to HI. So talking about phasing out "involuntary suffering" is wise. We may predict the abolition of all suffering. But that's a distinct question.

Minor complications aside, all nonhuman animals show a clearly expressed wish not to be harmed - not to starve, go thirsty, get attacked by predators and so forth. The issue of lack of consent does quite often arise as objection to modifying predatory animals. But as with humans, there is an immense difference between a right not to be harmed and a notional "right to harm". In practice, (ex-)predators will be helped as much as non-predators by compassionate stewardship. But a pan-species welfare state is indeed paternalistic. Does this matter? Compare human care of toddlers.

Chris, if one has more than a single sovereign principle, then there will be circumstances where they can come into conflict. Truth or Pleasure? Well, if one could glimpse all the suffering in the world, then one would go insane. So I think our long-term goal should be invincible ignorance of Darwinian life - but only after we are certain that all our ethical duties have been discharged. Maybe responsible cosmic stewardship can eventually be delegated to zombie AI.

The hedonic treadmill can become the hedonistic treadmill...
Techno-optimism
("an Analysis, an Evaluation and a Modest Defence")

How many genes need be targeted (1) to get rid of severe pain and suffering? and (2) abolish experience below hedonic zero?
A pessimistic answer would invoke the omnigenic model:
What if almost every gee affects everything?
But as far as I can tell, pain-sensitivity and hedonic tone are amenable to radical manipulation with a handful of genetic tweaks - which would make the problem of suffering more tractable. So how to shift the Overton window?

[on quantum mechanics]
Are you not shocked?
Why should the universe have been quantum-mechanical?
("If you want, you can divide Q into two subquestions:
Q1: Why didn’t God just make the universe classical and be done with it? What would’ve been wrong with that choice?
Q2: Assuming classical physics wasn’t good enough for whatever reason, why this specific alternative? Why the complex-valued amplitudes? Why unitary transformations? Why the Born rule? Why the tensor product?"
)

Why quantum mechanics?
Because not even God can create information ex nihilo.
Zero information = all possible descriptions = Everett’s multiverse.
Unitary-only QM is the quantum version of the Library of Babel:
Why is there somethijng rather than nothing

Peter Bryne’s biography of Everett is good - though the meaning of a “biography” is different in Everettian QM. Everett himself was heavily involved in writing software targeting cities in nuclear conflict. Indeed (if unitary-only QM is true) googols of branches presumably exist where Everett’s software was used in kill millions of people in thermonuclear war. Everett must have known of his complicity: he believed in his own interpretation of QM. Indeed, “our’ survival (cf. the Cuban missile crisis, nuclear submariner Vasili Arkhipov “saving the world”, etc) may be just an anthropic selection effect.
The Many Worlds of Hugh Everett
How should we act if wavefunction realism is true?
Everettian QM

As Penrose says, the only way to avoid Everett's multiverse is a “dynamical collapse” theory. QM without the collapse postulate wasn't invented to explain the appearance of fine-tuning: rather, no-collapse QM incidentally accounts for the appearance of fine-tuning as an anthropic selection effect.

[on politics]
"Political power grows out of the barrel of a gun."
(Mao Zedong)
Is the human species capable reasoned politics?
EA Magnus Vinding's new book...
Reasoned Politics

Magnus, set aside transhumanism and the abolitionist project as impossibly utopian. Focus on what we both agree is most morally urgent, i.e. preventing and mitigating severe human and nonhuman animal suffering. Any approach to socio-political reform that ignores the biological-genetic roots of severe suffering is just rearranging the deckchairs - Liberty, Equality, Justice, Democracy (etc).
Embracing secular Buddhist political values alone just doesn't get to the heart of the problem. In other words, germline reform isn't a prerequisite just of some futuristic abolitionist project. Politically, it's a prerequisite of preventing the 800,000+ annual suicides, hundreds of millions of chronic pain victims, hundreds of millions of victims of chronic depression, and untold wild animal suffering - horrendous stuff that conventional AND unconventional bioconservative politics can't touch.

The less the abolitionist project is associated with NU the better. If I could think of a snappier way to say it, I'd say the WHO project as laid out in their constitution - although complete health as so defined is wildly inconsistent with preference utilitarianism. Does hedonic uplift via recalibration have any "losers"?

[on future life]
We're asked to consider
"An imaginary clinic based in a country with minimal oversight of heritable human genome editing that offers these services to international clients following in vitro fertilization and preimplantation genetic diagnosis"
If I were setting up such a clinic, I'd use the language of depression-resistance and health, not transhumanism - and cite the WHO constitution. But such clinics should be ubiquitous.
WHO recommendations on human genome editiing

Bioethicists should be calling on the government to set up a research centre dedicated to ensuring the well-being of children born with unedited genomes...
How to protect the first ‘CRISPR babies’ prompts ethical debate
("Fears of excessive interference cloud proposal for protecting children whose genomes were edited, as He Jiankui’s release from jail looks imminent.")

Dan, why conserve sentient malware when we can become full-spectrum superintelligences? The whole Darwinian era will be forgotten like a bad dream.
But what are full-spectrum superintelligences?
As you know, I think digital computers are zombies; classical Turing machines are invincibly ignorant; and mind uploading is a pipedream. So-called AGI is cargo-cult science. Instead, our posthuman successors will be our biological descendants.

Tim, the hedonic range and hedonic set-points of modern humans don't differ significantly from our hunter-gatherer ancestors. So predictions of a biohappiness revolution don't rest on naïve extrapolation - the bane of traditional futurology. Rather, I assume that we will shortly understand the molecular basis of pain, pleasure and phenomenal binding. In consequence, Darwinian life will gain mastery over its reward circuitry. The biological-genetic dial-settings of pain sensitivity, hedonic range and hedonic set-points will shortly be individually (and societally) chosen rather than the gift of God/the Devil. What settings will you choose for you and your family? Critically, hedonic uplift won't involve the messy trade-offs (e.g. increased taxes!) that bedevil traditional socio-economic reform. For sure, completion of the abolitionist project assumes mankind's circle of compassion will expand to even the humblest animal life-forms. But technology massively amplifies the effects of even minimal benevolence. Most people don't enjoy witnessing or contemplating the suffering of others - visible distress is upsetting - and we’re heading for a global panopticon.

Resource depletion? Well, bitcoin mining aside, a shift to virtual lives and the metaverse promises effectively unlimited virtual resources and reduced consumption of traditional staples. Superyachts for all.

I'm sounding optimistic. I’m not: I’m an über-pessimistic button-pressing NU who'd end the whole shebang in a heartbeat if he had the chance. But my dark world-picture focuses on issues such as the nature of time and the universal wavefunction rather than the usual topical doom-and-gloom. So I'll bang the drum for a biohappiness revolution as long as I can...
The Imperative To Abolish Suffering

[on transhumanism]
Flying the flag for a "triple S" civilisation...

Spotify & mp4
Not many transhumanists are negative utilitarians. But you don't need to be NU to think we should eradicate suffering...
Transhumanism and NU

Living Forever, Gene Editing, and Psychedelics with David Pearce
Living Forever, Gene Editing, and Psychedelics
with DP

Marxist transhumanism? The Latent Transhumanism of Marxism
Buddhist nirvana reminds me of the upshot a utilitronium shockwave...
Nirvana (Wikipedia)
Andreas, it can certainly seem we're tasting the transhuman. Future shock! On the other hand, we have almost exactly the same hedonic range, core emotions, egocentric world-simulations, serial thought episodes, reproductive habits and default state of consciousness as archaic humans. Indeed, contemporary transhumanists probably have more in common with a chimpanzee troop than transhumans, let alone posthumans.

DP chats to Rajat Sirkanungo about transhumanism, spirituality and paradise engineering:
mp4
Shifting the Overton window is hard. Rather than a WHO-led Hundred Year Plan - the ideal! - we can anticipate slow, fitful, incremental change over hundreds of years, starting s with obvious genetic disorders. For sure, "Only a few hundred years to go!" doesn't have an inspiring ring. Perhaps there will be revolutionary socio-political shifts. The idea of life based on gradients of bliss is ultimately too compelling to go away - it just needs more powerful and persuasive advocates.

A critique of Nietzschean transhumanism:
"Infinite Monkeys: Nietzsche and the Cruel Optimism of Personal Immortality"

[on "La Revolucion de la Biofelicidad"]
Nuestros descendientes estarán motivados por gradientes de bienestar genéticamente reprogramado, órdenes de magnitud más ricos que las más sublimes experiencias actuales...
La Revolucion de la Biofelicidad
La Revolución de la Biofelicidad (mp4)

And the English version:
The Biohappiness Revolution
Perhaps using eugenics.org is a bit edgy.
But Darwinian life needs some serious recoding.

waterfall of obscure relevance

1 : 2 : 3 : 4 : 5 : 6 : 7 : 8 : 9 : 10 : 11 : 12 : 13 : 14

David Pearce (2022)
dave@hedweb.com


hedweb.com
HOME
2021 (FB)
2020 (FB)
2019 (FB)
2018 (FB)
2017 (FB)
2016 (FB)
2015 (FB)
2014 (FB)
Pre-2014 (FB)
Video Interview
Some Interviews
BLTC Websites 2022
The Philosophy Forum
The Abolitionist Project
Quora Answers 2015-22
Social Network Postings (2022)