JANUARY 2017 -
[on cryonics and cryothanasia]
I haven't signed up for cryonics. When I'm ready, perhaps in a decade or so, I intend to use cryothanasia. Your best prospect of a successful reanimation is to be suspended in optimal conditions and cognitively intact. Max More has done an impressive job professionalising Alcor. How many of their current patients will be restored? Given the circumstances of their suspension, I don’t know. I hope Alcor and SENS can collaborate in a twin-track approach to tackling the scourge of death and aging. IMO, cryonics should be opt-in rather than opt-out. And if cryothanasia – with rigorous safeguards – were an option too, then the entire program could be self-financing. For millions of people today, the last six months or even six weeks of their existence absorb more medical time, effort and expense than the rest of their lives put together.
That said, I’m inconsistent. On the one hand, anyone with the slightest shed of intellectual curiosity ought to relish the prospect of waking up in the twenty-second or twenty-third century and catching up with scientific progress. Upon re-animation, I’d like to devour a year-by–year chronicle of key events - together with offbeat stuff like an annual dictionary of new words in the English language. On the other hand, will our enhanced descendants / posthuman superintelligence really believe that it’s ethical to reanimate dysfunctional, malaise-ridden primitives from the genetic dark ages? Won’t the whole Darwinian era be accounted best forgotten like a bad dream? Sure, archaic human ills of brain and body will readily be curable. Yet in what sense would a superhappy, superintelligent “DP” be me in anything but name? Given my disbelief in enduring personal identity over time, and strong suspicion that Everettian quantum mechanics is true, I’m sceptical of any claims to metaphysical identity over time - regardless of what criterion of identity we use. Yet such scepticism doesn’t mean that my namesakes wake up each morning in the throes of an existential crisis. Waking up in the year 2200, say, is no different.
Given these reservations, why urge universal access to cryonics? Well, partly because we need to banish the terrible fear of death and aging world-wide. Transhumanists should set an example and practice what we preach. One reason that so many people today are reluctant to back serious funding for radical antiaging research is their (quite justified) sense that they personally aren’t going to make the transition. So the transhumanist message that death and decrepitude are terrible, but you probably belong to the last human generation ever to experience involuntary aging, threatens to undermine painfully assembled rationalisations of why death and aging are good, necessary and natural.
* * *
If you were sure you'd wake up, would you like to be put in cold storage for a few centuries?
("Scientists Have Found a Way to Rapidly Thaw Cryopreserved Tissue Without Damage")
"If you want a happy ending, that depends, of course, on where you stop your story." (Orson Welles) Opt-out cryonics?
("Study into who is least afraid of death")
Stop Eating Meat
Thanks Nikola. I've never tasted meat, an accident of birth rather than a badge of superior virtue. More impressive are transhumanists ranging from the late great FM-2030 - who “wouldn’t eat anything that had a mother" (https://en.wikipedia.org/wiki/FM-2030) - to my interviewer Nikola who carefully weighed the arguments and then changed their behaviour accordingly. A commitment to the well-being of all sentience (cf. https://hpluspedia.org/wiki/Transhumanist_Declaration) is inconsistent with the existence of factory-farms, slaughterhouses, and industrialised animal abuse. I'd encourage anyone who still eats meat to quit - and also urge any meat-eating friends and colleagues to do likewise.
Outlaw factory-farms and slaughterhouses. By all means grow your own meat, but buying a vegan recipe book is easier.
("Make your own meat with open-source cells – no animals necessary")
Meat eating adversely affects the health of killers and victims alike. The vCJD story:
("Many more people could still die from mad cow disease in the UK")
If ethics weren't an issue, then a vegetarian + fish-based diet rather than veganism or meat-eating would probably be most conductive to human health and longevity. But like weighing the issue of slavery on the improved cardiovascular health vs incidence of backpain of Aryan slave-owners, we'd be rather missing the point. If we're to see optimal human & nonhuman health and closure of the death factories in our lifetime, then I think in vitro meat is going to be essential.
Peter Singer is right; the profoundly disabled deserve to be loved and cared for almost as much as pigs. Hence the need for global veganism. Let's close and outlaw the death factories.
("Peter Singer Thinks Intellectually Disabled Less Valuable than Pigs Read more at: http://www.nationalreview.com/corner/445244/peter-singer-thinks-intellectually-disabled-less-pigs")
Property-dualist panpsychism is worth distinguishing from non-materialist physicalism, i.e. experience discloses the essence of the physical.
("Philosopher thinks panpsychism ('all matter has mind') is probably true")
[Peter writes: We are able to lose consciousness, so panpsychism is far from parsimonious. General anaesthesia shouldn't work if panpsychism were true."]
Or when we fall into a dreamless sleep, we become micro-experiential zombies - in effect, just aggregates of decohered neuronal "mind-dust".
My own views were inspired by a philosopher who "turns Kant on his head":
Postscript (Dec. 2016)
("Postscript to review of Michael Lockwood")
If a "Lego professor of play", why not a chair of Paradise Engineering?
("A university is offering people the chance to be the first ever Lego professor of play")
Will academic philosophy always be a male-dominated sport for guys who love arguing?
("What is Philosophy's Point? Part II. Maybe It's a Martial Art.
Philosophers sometimes seem more concerned with winning than wisdom")
From the foundations of QM to the Hard Problem of consciousness, science is steeped in (bad) philosophy. Alas, philosophy is often still steeped in (bad) high-school physics.
("What Is Philosophy's Point?, Part 3. Maybe It Should Stick to Ethics")
Modern science - including computer science, is steeped in philosophical presuppositions and background assumptions - some (hopefully) good, some bad.
("Philosophers Push Back. Philosophers react to a science journalist’s critique of their calling")
[on David Chalmers AMA]
David Chalmers AMA
Thanks Pablo. Yes, David Chalmers is justly well known for framing the Hard Problem: why aren’t we p-zombies? (cf. https://en.wikipedia.org/wiki/Philosophical_zombie) Chalmers is less well known for spelling out the momentous implications for monistic physicalism and the unity of science if we can't solve the phenomenal binding / combination problem even if some form of panpsychism is true. And binding really does seem classically impossible: http://consc.net/papers/combination.pdf.
All this philosophising might seem far removed from the purpose of this FB HI group, i.e. to promote phasing out the biology of suffering in favour of gradients of intelligent bliss in all sentient life. Yet we can't hope to act ethically and responsibly if our theory of mind turns out to be catastrophically mistaken. At Tucson in 2010 and 2016 I've argued that classical digital computers can't be phenomenally bound subjects of experience, let alone suffer. So they can’t be harmed like biological minds. But - sanity check - this conjecture could well be mistaken. (cf. http://www.scottaaronson.com/blog/?p=2756) Therefore before e.g. attempting "mind uploads", or running ancestor simulations, or digitally radiating across the Galaxy, or trying to build superintelligent AGI (etc), we should make sure that we understand the fundamentals of consciousness, the pleasure-pain axis, and phenomenal binding.
Otherwise we might do something really stupid.
* * *
John, yes, I used the USA as an example because one contemporary materialist philosopher, Eric Schwitzgebel, argues that the USA is already a pan-continental subject of experience (“If Materialism Is True, the United States Is Probably Conscious”) http://faculty.ucr.edu/~eschwitz/SchwitzAbs/USAconscious.htm
The older term is the "China Brain" argument,
not to be confused with Searle's Chinese room.
Roger Penrose? In my view, the Penrose-Hameroff Orch-OR theory is potentially good science (even though I think it's mistaken!) because it's experimentally falsifiable. Philosophising aside, molecular matter-wave interferometry will either detect some collapse-like deviation from the unitary Schrödinger dynamics or it won't. End of story.
So what about explanations of phenomenal binding that don't invoke dynamical collapse and just accept the "bare formalism" of QM? Well, if coherent superpositions of distributed neuronal feature-processors lasted milliseconds, then we would have a credible candidate for a perfect structural match between phenomenology and physics, “saving the phenomena” and confounding dualists like David Chalmers. Superpositions are individual states, not classical aggregates. If non-materialist physicalism is true, then neuronal superpositions must be phenomenally bound. Decoherence explains phenomenal unbinding...
So why would almost any scientifically informed person think an experiment like 6 is scarcely worth attempting?
Timescales. As your professor would say, theoretical calculations (mine are mostly lifted from Tegmark and Schlosshauer) tell us that the coherence of phase angles of the components of a hypothetical neuronal superposition can’t last for more than femtoseconds or less. The CNS is too hot to be a quantum computer! It's “obvious" that this kind of timescale is at least a dozen orders of magnitude too short to underpin our phenomenally bound minds.
What I'm arguing (together with a bunch of theory about applying Zurek's insanely powerful “quantum Darwinism” to the CNS https://arxiv.org/ftp/arxiv/papers/1001/1001.0745.pdf) is that we should stop trusting our intuitions about what’s "obvious" and instead focus on designing better experimental protocols. Any scientific theory of conscious mind should be experimentally testable.
* * *
Tim, IMO what's interesting about p-zombies is also what's interesting about micro-experiential zombies - not the sceptical worries, or debate over whether such entities are possible, but rather understanding why we aren't p-zombies or micro-experiential zombies, given what we think we know about the fundamental stuff of the world and the effectively classical nature of neurons. If monistic physicalism is true, then we ought to be able to derive the properties of our phenomenally bound conscious minds - and the egocentric world-simulations they run - from the underlying physics in the same way as we can derive the properties of biological life via the Modern Synthesis. No such derivation seems possible for the properties of phenomenal consciousness - as David Chalmers and many other critics have argued - which leads either to eliminativist denial (e.g. Dennett, the Churchlands, Brian Tomasik, Matt in this group), or dualism (Chalmers).
The tentative, idiosyncratic answer I'd give is that (1) "p-zombies" are impossible because nonmaterialist physicalism is true: p-zombies are unphysical; (2) "micro-experiential zombies” are impossible because decohered classical neurons are an artefact of our temporally coarse-grained tools of investigation. Mercifully, this question can be settled experimentally via interferometry.
I share some of your reservations about Penrose's motivations. But surely motivated cognition is endemic – we’re all guilty to a greater or lesser extent. What matters scientifically is that (like all dynamical collapse theories) Orch-OR makes empirically falsifiable predictions and will soon be refuted (or spectacularly vindicated!) My own doubts about whether the superposition principle of QM ever breaks down are more “philosophical” (cf. https://www.quora.com/Why-does-the-universe-exist-Why-is-there-something-rather-than-nothing) than technical. I'm also aware that the track record of philosophical intuition isn’t impressive.
Consciousness: how would you answer these four questions?
Raising the Table Stakes for Successful Theories of Consciousness
A lot of scientists would simply lump all four questions together as aspects of The Hard Problem of consciousness that should be quarantined off from the rest of science. This is rather unsatisfactory. Imagine if since classical antiquity we'd just of the "Hard Problem of matter and energy" without troubling ourselves with developing classical or quantum physics.
Thanks Andrés. The gulf in background assumptions between radical eliminativists about consciousness (like Brian, Daniel Dennett or the Churchlands) and non-materialist physicalists who think consciousness is ontologically fundamental is immense. My physicalism.com paper was prompted by David Chalmers’ argument that neither classical nor quantum physics can explain phenomenal binding; therefore we must embrace (naturalistic) dualism. At Tucson - a hotbed of consciousness realists! - I started by asking the audience whether they believed that if the intercommunicating skull-bound minds of the population of the USA were to implement any computation at all, whether they believed a pan-continental subject of experience would thereby be generated. Some responded “yes”, others “no”, and others were agnostic; but no one suggested that the question itself was meaningless because there are no conscious skull-bound American minds in the first instance, merely skull-bound brains!
Given this seemingly unbridgeable gulf, how can we make progress?
Here I'd invoke Popper. Does a conjecture make any novel, precise, experimentally falsifiable predictions that proponents AND critics can agree will settle the issue between rival theories? And if not, why not?
* * *
("Phenomenal consciousness is a quasiperceptual illusion: Objections and replies") No "third path"? I'm not so sure. Also, converting the world to radical eliminativism about consciousness might seem an elegant solution to the problem of suffering. Alas I don’t think it will work.
1) Only the physical has causal efficacy; and we have strong grounds for believing physicalism as enshrined in the Standard Model or its extensions is true.
2) First-person consciousness is real; and e.g. my subjective feeling of frustration at eliminativism plays a causal role in the posting of this comment.
On the face of it (1) and (2) are mutually inconsistent. However, if non-materialist physicalism is correct, _both_ claims are true. Non-materialist physicalists conjecture that experience discloses the essence of the physical – the elusive “fire” in the equations – on whose intrinsic nature physics (i.e. QFT or its generalisation) is silent. Strictly speaking, all consciousness, and only consciousness, has causal efficacy. In recent years, the best known advocate of non-materialist physicalism - though not under that label - has been Galen Strawson: As David Chalmers argues, perhaps the biggest challenge for non-materialist physicalism - and for property-dualist panpsychism - is the phenomenal binding/combination problem:
(For what it’s worth, my response is here: http://www.physicalism.com)
Phenomenal binding not a problem for the radical eliminativist about consciousness because there’s nothing to bind.
Rob, if experience discloses the intrinsic nature of the physical, then what we have been calling "p-zombies" are unphysical. Granted this intrinsic nature, then yes, as you note, we can nonetheless apparently imagine notional worlds where this isn't so – after all, the originator of the "fire in the in the equations” metaphor that we’re using to evoke the mysterious essence of the physical believed that its non-experiential nature was too obvious even to be worth expressly stating, despite his professed mystification about what it might be. Yet are these notional worlds that we seemingly imagine really physically possible - in which case, yes, one might struggle to locate oneself among them - or is arguing for their physical conceivability akin to imagining worlds where water isn’t H2O? What once naively seemed a physical possibility isn't really so at all.
Either way, all I wanted to do above is suggest that it’s too quick to say there is “no third path”. If non-materialist physicalism is true, then phenomenal consciousness does causally determine which assertions we end up emitting. Non-materialist physicalism sounds pretty insane; but so does radical eliminativism or Chalmersian dualism. And the truth may be stranger still. Who knows?!
* * *
Is consciousness quantifiable?
If non-materialist physicalism (as explored by e.g. Galen Strawson, me) is true, then presumably the solutions to the equations of physics exhaustively yield the precise values of qualia. So consciousness in all its guises is rigorously quantified by QFT or its generalisation. Consciousness is the essence of the physical: mathematical physics describes symmetry transformations of qualia. However, the lack of any kind of cosmic Rosetta stone to allow "reading off" these diverse values means that psychonauts exploring alien states-spaces of consciousness (not least, "psychedelia") will be reduced to saying lame-sounding things to the drug-naïve like how it’s "ineffable", "inexpressible", etc.
We can go much further with quantifying conscious states on the pleasure-pain axis because they can be “operationalised”. How hard will an organism work to obtain or avoid the pleasant or unpleasant experience induced by a stimulus?. Cross-species behavioural studies, neuroscience, evolutionary genetics, linguistic self-reports and the preferences of opioid drug users for selective full mu agonists all tend to converge. But even here there are desperate methodological problems – not least, our lack of a theory of phenomenal binding.
* * *
Thanks Mike. The bizarre-sounding implication of non-materialist physicalism that mathematical physics is really about patterns of qualia can't obviously be tested / directly experimentally falsified. (cf. the palette problem http://onlinelibrary.wiley.com/doi/10.1002/tht3.113/abstract)This is one reason I focus on the binding problem. The implications if non-materialist physicalism is true are no less bizarre. But empirically falsifiable predictions follow. If a perfect structural match between conscious mind and formal physics doesn’t exist, then the entire package falls. Monistic physicalism of any kind – “materialist” or non-materialist - is false. If it’s actually true, then the missing "cosmic Rosetta stone" problem is daunting. Yet at least we’ll know we’re not barking up the wrong tree.
A couple of thoughts about psychological hedonism / valence determinism.
If we combine the God's-eye perspective of physical science with even a crude and exception-ridden version of the pleasure principle - i.e. sentient beings tend to seek pleasure and avoid pain, albeit frequently another description - then profound consequences for decision-theoretic rationality follow. The countless complications shouldn’t obscure how biotechnology will soon be able to give all sentient beings what they are after, emotional well-being – once again, often under other descriptions.
On your point about how the emotional valence of experience is affected by all sorts of factors, I'd agree. The “encephalisation of emotion” under pressure of natural selection has been immensely fitness-enhancing. This encephalisation isn't simply the one-way painting of hedonic tone on otherwise hedonically neutral neocortical experiences. Despite this spaghettification of code, yes, I do tend to focus (too?) simplistically on the ancient, non-perceptual limbic system. You mention OCD-behaviours. Doesn't the anticipated pleasure explain their misplaced persistence? (cf. http://sites.tufts.edu/emotiononthebrain/2014/10/10/dopamine-imbalance-and-its-effect-on-compulsive-behaviors/).
Psychological hedonism / valence determinism isn't the claim that we're rational pleasure-seekers / pain-avoiders.
Or to choose a very different example, co-administer to a subject mu opioid and dopamine antagonists and the most exquisitely rendered symmetries will leave even a world-class mathematician cold. Co-administer mu opioid and dopamine agonists to your reward centres and even extreme asymmetry will look inherently sublime.
Or at least that’s what I’d predict from my armchair.
* * *
The power of ideas: whereas people with Cotard's syndrome believe they are dead, eliminative materialists believe they are zombies - and try to infect the sentient. But some of us have innate immunity.
("Is consciousness an illusion")
Brian, do you ever have lucid dreams? I ask because with dreaming, we don't get caught up in disputes about the nature of perception. Whether sunsets or symphonies, dreams can be extraordinarily vivid – and everything unfolds within one's transcendental skull. Should awake Dennettian investigators inform a lucid dreamer that (s)he shouldn’t believe in the "Cartesian theatre"? The lucid dreamer can even conjure up a phenomenal theatre in front of his phenomenal body-image - if so requested. A Dennettian can say that this theatre is just an “illusion” created by the brain; but where exactly is this illusory phenomenal theatre if not spatio-temporally located within the dreaming subject’s transcendental skull?
("Saying 'Hi' Through A Dream: How The Internet Could Make Sleeping More Social")
* * *
In the desert, the naive realist will say he can directly perceive an oasis. I would say that I apprehend a phenomenally bound oasis in my visual world-simulation that causally covaries with a mind-independent physical oasis. (Even thirsty philosophers can be long-winded.) Perhaps our guide says that there's actually no physical H2O for us to drink at all; the oasis is just a mirage. But where I really balk is if a Dennettian or other radical eliminativist about consciousness says that none of us even seems to see an oasis, illusory or otherwise - that even the appearance of a patch of water in front of one’s body-image doesn't exist - and neither, for that matter, does one’s phenomenal body-image. Yes, we should bear in mind what Sellars called the "myth of the given", and acknowledge the theory-ladenness of perception. And for sure, the first-person contents of both introspection and perceptual experience/one’s world-simulation can be radically misconceived by the subject. Yet I can't feign anaesthesia to myself. If a Dennettian says my agonising headache can be banished by better philosophising rather than painkillers, then he is the victim of the false ideology of eliminative materialism.
* * *
Elijah, recall there's a sense in which Brian, Dennett, Galen Strawson (and me) agree: all that exists is the physical, and only physical processes have causal efficacy. Where Strawson, I, and any non-materialist physicalist part company with Brian and Dennett is we theorise that subjective experience discloses the intrinsic nature of the physical - and in virtue of being physical, all and only subjective experience has causal efficacy, whether functionally incidental (as in your PC or a lettuce) or otherwise, and whether phenomenally bound or otherwise. Brian and Dennett, on the other hand, can find no place for first-person subjective experience in their ontology, and so bite the bullet: consciousness doesn't exist!
Input-output relations? Natural selection has recruited various kinds of experience, each with its own molecular signature, to play a typical information-signalling role in biological organisms. In many cases, we don't even have a separate word in our vocabulary for the subjective “raw feels” and the typical functional role such subjective experiences play in biological minds. In fancy speech, we distinguish e.g. phenomenal pain from the function of nociception and acknowledge that pain and nociception are doubly dissociable. By contrast, the term “jealousy", for instance, refers to both a functional role and a bunch of characteristic “raw feels”; we don't have even a formal term conceptually to distinguish the two.
For what it’s worth, I reckon the vast majority of possible states of consciousness that matter and energy can instantiate have never been recruited by evolution for any information-processing role. This very absence of function is one reason why the experiences induced by major psychedelics can be so disconcerting. It's not that psychedelics deliver deep truths about the universe, but rather the state-spaces of consciousness that such drugs open up don’t come with any kind of pre-digested conceptual scheme by which psychonauts can navigate their weirdness.
Not a problem for zombies.
* * *
Worms and philosophers have a dopamine & opioid system & a pleasure-pain axis and should be treated accordingly.
("We know we are – but what else is conscious too?")
What's it like to be a cerebellum?
("Study finds testosterone levels influence link between cerebellum density and neuroticism")
Vito, bafflement at the Hard Problem and/or the binding problem of consciousness isn't confined to professional philosophers. It extends to some of the most brilliant scientific minds of our generation (cf. http://wavefunction.fieldofscience.com/2016/08/physicist-ed-witten-on-consciousness-i.html)
Does Stanislas Dehaene's theory of consciousness make any novel, precise, experimentally falsifiable predictions? If so, it's "risky" in Popper's sense - and potentially good science. But if not...
Waking life is analogous; but natural selection ensures your autobiographical world-simulation is harshly constrained:
("I can control a computer with my mind – from inside a dream")
Can semantics be naturalised if we assume that lucid dreamers (i.e. us) never really "wake up"?
Natural selection doesn't explain consciousness. Your molecular doppelgänger assembled from scratch would be just as sentient.
("'From Bacteria to Bach and Back' by Daniel C Dennett review – consciousness explained? There is no ‘hard problem’ and consciousness is no more mysterious than gravity, Dennett claims in this study of the evolution of minds")
Is insentience useful to a p-zombie? Materialism is false; non-materialist physicalism is (at least) empirically adequate.
Without phenomenal binding, consciousness would be useless. But for classical neurons it's as feasible as telepathy.
* * *
What should be our point of departure? As Edward notes, here is a kind of inverted cogito. Brian, like Daniel Dennett, takes third-person facts as conceptually prior, and the existence of first-person facts is a theoretical inference that one makes. Inferences, however intuitively powerful, can be mistaken. In his latest book, Dennett, just like Brian, is quite explicit: you are a zombie! (page 363)
In my view, this perspective gets everything back-to-front. Any inference you make to a mind-independent physical world beyond your subjective phenomenal world-simulation is a theoretical inference to the best explanation. Some of the information-processing systems in this hypothetical mind-independent physical world are zombies, like AlphaDog or your PC, whereas others are phenomenally bound subjects of experience, like the mind-brains of humans, mice and bumble-bees. This dichotomy between subjects of experience and zombies holds, IMO, regardless whether panpsychism, or a property-dualist theory, or non-materialist physicalism is true.
* * *
Reflective self-awareness: the circle widens. But sentience, not sapience, is what matters...
("Monkeys taught to pass mirror self-awareness test")
In my view, phenomenally bound quantum minds, after some 540 million years, evolved a slow, inefficient serial virtual machine capable of conceiving and designing a classical Turing machine: a programmable digital computer. There is simply no way that natural selection can evolve programmable digital computers powering the zombie robots that are increasingly going to outperform us in tasks involving “narrow” AI.
Or still stuck at the starting-block. ("Everyone knows you can't find consciousness in the individual cell")
("Moving Toward Understanding Consciousness")
AI: phenomenal binding is impossible for a classical digital computer. Solving Sir Roger's chess problem isn't...
("Can you solve chess problem that holds key to human consciousness")
It's good that sceptics pose concrete challenges rather than just philosophise. ("You insist that there is something a machine cannot do. If you tell me precisely what it is a machine cannot do, then I can always make a machine which will do just that." - John von Neumann) But the Penrose-Hameroff Orch-OR theory of consciousness will be definitively falsified to the satisfaction of proponents and critics alike when interferometry finds no collapse-like deviation from the unitary Schrödinger dynamics.
Matt, eliminativism just doesn't work. Feelings and illusions are particular kinds of consciousness. If we believe in physicalism and the unity of science, i.e. no spooky "strong" emergence, then we need to derive the properties of feelings, illusions and all other forms of consciousness from the underlying physics. Compare how vitalism has been confounded by the reduction of life to molecular biology, molecular biology to quantum chemistry, and quantum chemistry to the Standard Model.
However, if we make the extremely plausible but contested assumption that the mysterious "fire" in the equations, i.e. the essence of the physical, is non-experiential, then we face the Hard Problem of consciousness. And if we further make the extremely plausible but contested assumption that neurons in the CNS are discrete, decohered classical objects, then we face the Binding Problem. You are not a "micro-experiential zombie" - not unless you're in a dreamless sleep, at any rate, in which case you won't be reading my deathless prose.
* * *
Magnus Vinding defends consciousness realism:
What is our point of departure?
Low AQ folk start by introspecting their own consciousness - from pleasure, pain, core emotions, perceptual experience, thought-episodes, etc - and try to understand other organisms / information-processing systems by analogy. (cf. The Problem of Other Minds.)
High AQ folk start with observing other organisms / information-processing systems, detect patterns in their verbal and non-verbal behaviour - including talk about "consciousness" - and try to understand themselves analogously. In other words, if the behaviour of other organisms / information-processing systems can be physically understood on the assumption they are zombies, then (ultra-high-AQ folk may reason) by analogy I am too.
Are you "a serial processor that can think of only one idea at a time"? Or a massively parallel world-simulation?
("Want To Know What Your Brain Does When It Hears A Question? Questions hijack the brain. The moment you hear one, you literally can't think of anything else.")
* * *
If non-materialist physicalism is true, then experience may be regarded as physicalized mathematics. Posthuman superintelligence should be able to order the menagerie of qualia using the mathematics of symmetry - just as the greats of twentieth century physics created the Standard Model in physics. For sure, non-materialist physicalism may be false and the Hard Problem may be beyond human minds. My objection to "mysterianism" (cf. https://en.wikipedia.org/wiki/New_mysterianism) isn't that it's wrong, but rather it's sterile. In any "Shut up and calculate!" versus "Shut up and philosophise!" debate, my sympathies normally lie with scientists. However, one or more totally obvious "philosophical" background assumptions that we're making about consciousness must be mistaken. But which? Above all - and this is a reproach to my younger self - anyone who writes about the Hard Problem should focus on extracting novel, precise, experimentally falsifiable predictions - and not dress up their retrodictions as predictions. Compare Andrés’:
Like most people, my response was "That's insane!" My response should have been, "Fantastic, a falsifiable hypothesis put to the test". Keep them coming.
* * *
Are mini-brains also mini-minds? Or are a pack of autistic or hyper-sociable neurons alike just micro-experiential zombies?
("Mini-brains made from teeth help reveal what makes us sociable")
Matt, there is a difference between the first-person fact of having a migraine and displaying the spectrum of behaviours commonly associated with migraines while uttering the words "I have a migraine". How first-person facts can arise in a world supposedly devoid of experiential properties is the Hard Problem of materialist metaphysics.
What are the biggest flops of physics theories?
The Standard Model. Any decent scientific theory should be empirically adequate. All one ever empirically knows, except by inference, are the contents of one's own conscious mind. Yet contemporary physics has no explanation of why consciousness can exist at all (the “Hard Problem”) (2) how consciousness could be locally or globally bound by a pack of discrete, decohered, membrane-bound, supposedly classical neurons (the phenomenal binding / combination problem) (3) how consciousness exerts the causal power to allow us to discuss its existence (the problem of causal impotence versus causal over-determination) (4) how and why consciousness has its countless textures and the interdependencies of their different values (the "palette problem").
Any theory of the world inconsistent with one's existence must count as an epic fail.
Solutions? Heaven knows. Perhaps Ed Witten is right: (cf. "World's Smartest Physicist Thinks Science Can't Crack Consciousness") My own best guess is that wavefunction monism is true; no "element of reality" is missing from the mathematical formalism of quantum field theory or its extension; but the entire mathematical machinery of modern physics should be transposed to an idealist ontology.
"Extremely implausible” (David Chalmers); and of course, I agree.
* * *
But fields of what exactly? ["What is it that breathes fire into the equations and makes a universe for them to describe?" - Stephen Hawking] Intuitively, the intrinsic nature of the physical - i.e. the "fire" in the equations of QFT - is something non-experiential which (somehow) gives rise to consciousness. However, it’s worth stressing that this non-experiential nature is a metaphysical assumption, not a scientific discovery.
A physical theory can be mistaken in two ways. It can fail accurately to predict e.g. the precession of the perihelion of Mercury, or the Lamb shift, or whatever. But a theory can also be mistaken if it gets its ontology wrong, e.g. luminiferous aether. The Hard Problem of consciousness arises on the assumption we have our basic ontology right, i.e. that the mathematical formalism of quantum physics is about fields of insentience - whatever that may be.
* * *
A nice review of IIT. More generally, When one's theory is at odds with the empirical evidence, should one grant it's false? Or speak of the Hard Problem?
("The Problem with the ‘Information’ in Integrated Information Theory By: Garrett Mindt")
Does Nature need civilising or "re-wilding"? Why are humans so fond of predators? Mathematical physicist and wilderness enthusiast John Baez...
("Restoring the North Cascades Ecosystem")
Do we really want a living world where sentient beings hurt, harm and kill each other – and where predators disembowel, asphyxiate and eat their victims alive? The CRISPR revolution in biotech, synthetic gene drives, and the explosive growth of IT mean the entire biosphere will shortly become programmable. What is the optimal burden of suffering in Nature? For the first time in history, intelligent moral agents will shortly be able to choose. No, I’m not suggesting that we embark on a Five Year Plan to veganise the living world and roll out the costly cross-species fertility regulation that compassionate stewardship would entail – not yet anyway. But we should think long and hard before actively promoting “re-wilding” – and human or nonhuman predators.
John, I share your dark views about what humans are doing to the planet. But does “rewilding” call for any less wisdom than compassionate stewardship which aims to minimise the burden of suffering in the living world?
On your technical points, yes, phasing out or genetically tweaking predators is recipe for Malthusian catastrophe – if taken on its own. But if combined with e.g. cross-species immunocontraception to regulate population sizes, then the ecological sustainability of a non-violent biosphere needn’t be an issue. And the crazy thing about using CRISPR-based synthetic drives is that if intelligent moral agents want to “fix” genes for e.g. low pain-sensitivity across entire sexually reproducing species, then this compassionate intervention can work even if the allele(s) in question carry a fitness cost to the individual. Gene drives “cheat” the laws of Mendelian inheritance and natural selection as traditionally understood.
Yellowstone? Wolves eat their larger victims alive. More generally, predators promote a “landscape of fear”. There are more civilised ways to run an ecosystem than starvation and predation – IMO.
John, I completely agree with you about the need for extreme caution. But later this century and beyond, every cubic metre of the planet will be computationally accessible to surveillance and micromanagement. Do we want a living world where sentient beings harm each other or not? If intelligent moral agents opt for a living world without starvation and predators, we’re not going to run out of computational resources. Ultimately the level of suffering in the post-CRISPR biosphere is an ethical question.
* * *
Sentient beings eating each other: Darwinian life is hell.
("Spiders eat twice as much animal prey as humans do in a year")
Cross-species immunocontraception, let alone veganising the post-CRISPR biosphere, is hard, but overcoming status quo bias is harder still.
("Consistent Vegetarianism and the Suffering of Wild Animals")
"Revive and restore"?
Should pain-ridden Darwinian life be recreated, conserved or replaced by something better?
("Revive & Restore")
[on the brain]
Same neurons, new microglial housekeepers. Would you notice the difference?
("'Housekeepers' of the brain renew themselves more quickly than first thought")
LSD: consciousness should be investigated by life-loving psychonauts, not the depressed.
("LSD to Cure Depression? Not So Fast")
Depression is a cancer of the mind and a overgrowth of the neocortex. When can medical science design smart euphoriants?
("Study finds brain architecture alters to compensate for depression")
If we make the plausible assumption that (in the ancestral environment) the behavioural suppression of depression was adaptive (cf. rank theory of depression: https://www.biopsychiatry.com/depression/), then thinking and introspecting a lot was probably fitness-enhancing. Today psychiatrists call such excessive thought "ruminating". The "hypercholinergic frenzy" theory of depression proposes that low mood is associated with an overactive cholinergic system. This is one reason why melancholic depressives should be cautious about using cholinergic "smart drugs" / nootropics.
Combating depression: sleep discipline / chronotherapy can complement optimal nutrition and aerobic exercise.
("Yes, Your Sleep Schedule Is Making You Sick")
SAD: mood, light and darkness:
("Is the dark really making me sad?")
But the only long-term solution is rewriting our ghastly genetic source-code - together with the rest of Darwinian life.
("What does it mean to be human")
Glyoxalase 1 inhibitors are mood brighteners; but the field of antidepressant research is littered with blighted hopes that led nowhere.
("Mouse study provides target for development of faster-acting antidepressant medications")
"Always embrace negative emotions" vs genetically engineer a biology of superhuman bliss.
A tough one.
("The Zen of Sadness: Why You Should Always Embrace Negative Emotions")
[on the binding problem]
[Darren writes] "One of my tutors, who runs a lab studying minimal perception, says that the solution to this problem is that there is no binding, i.e. We are indeed micro-experiential, but manifestly not zombies."
Intriguing. In one sense, I can understand such an incredible view - because as normally posed, the binding problem is insoluble on pain of magic. People sometimes ask, inconclusively, "What is consciousness for?" Perhaps a more apt question is "What is phenomenal binding for?" Here at least we can give compelling answers. (cf. https://en.wikipedia.org/wiki/Simultanagnosia) Unfortunately, showing that phenomenal binding is fitness-enhancing is not an explanation of how it's possible. Telepathy would be fitness-enhancing too. If our neurons are classical, phenomenal binding is no less unphysical than telepathy. By contrast, if one starts from the assumption that the superposition principle of QM never breaks down, then one gets a different answer. Superpositions are individual states, not classical ensembles. But decoherence times of neuronal superpositions in the CNS are so insanely rapid, I can understand why few people take this solution seriously.
* * *
The 320 million individual skull-bound minds of the USA aren't a pan-continental subject of experience, nor even an illusory pan-continental subject of experience, nor a pan-continental subject of experience with multiple drafts, nor a pan-continental subject of experience that merely believes it’s conscious (etc etc). Zilch. Or rather, zilch if monistic physicalism is true: as soon we allow "strong" emergence, anything is possible. The challenge is to show why and how a pack of membrane-bound neurons is different: a pan-cerebral subject of experience.
The normal way to pose the binding/combination problem assumes perceptual direct realism – i.e. exposed brains are cheesy wet lumps of neural porridge and light microscopy reveals they are composed of a pack of classical neurons. This conventional approach highlights the discrete, distributed neuronal feature-processors (colour-, motion-, edge-(etc)detectors) synchronously activated by the presentation of different perceptual stimuli as revealed by neuroscanning and microelectrode studies. How does mere synchronous activation somehow get transmuted into phenomenal unity?
As so framed, phenomenal binding is classically impossible.
Instead, start from quantum field theory. Then attempt to derive via the decoherence program in QM an approximation of discrete, decohered classical feature-processing neurons behaving over seconds and milliseconds just as textbook connectionist neuroscience and learning algorithms suggest. In a deep coma or a dreamless sleep, your CNS can indeed be crudely described as though it were a pack of 86 billion classical neurons. However, when you’re awake or dreaming, then - I conjecture - interferometry at fine-grained temporal resolutions will show the non-classical interference signature of well-defined phenomenally bound perceptual objects of the world-simulation in your skull. Interferometry experiments are technically demanding. Yet such experiments are easier to perform if we know precisely what we're looking for - or rather, if we know the specific conjecture that we're trying to refute. I don't blame Vito or David Chalmers (“extremely implausible”) for being dismissive. But if we’re not prepared to experiment as well as philosophise, then we'll be having the same inconclusive debates for ever.
* * *
Cool pictures. But not mind maps. A pack of decohered classical neurons would be a micro-experiential zombie...
("Mind maps: the beauty of brain cells – in pictures")
Bees: running real-time phenomenal world-simulations is the greatest cognitive achievement of post-Cambrian life.
("How insects like bumblebees do so much with tiny brains")
The mind-brain of a bee is sesame-seed-sized: it runs world-simulations, uses language, and solves complex problems. Perhaps programmable digital zombies will soon be able to compete.
("Bees learn to play golf and show off how clever they really are")
Invertebrate psychotherapy isn't imminent, but humans typically underestimate insect minds.
Wasp offspring found to take on the personality of the queen
* * *
Thanks Stuart. What is the smallest physically possible “psychon" of experience - the most minimal, self-intimating fleck of what-it's-likeness from which phenomenally bound minds could ever be built? If non-materialist physicalism is true, then presumably the “psychon” is of Planck-scale dimensions, i.e. very small indeed. But to feel the force of the phenomenal binding/combination problem, it's not necessary to enter into such deep philosophical waters. Just allow - as microelectrode studies using awake human subjects tend to confirm - that individual membrane-bound neurons in your CNS can sustain rudimentary pixels of experience: for example, a fleeting speckle of micro-redness, a brief faint hiss of sound, etc.
Some philosophers have denied the conceptual possibility of "ownerless" experiences. Allegedly, we can't create pain or pleasure or redness in a test tube; experiences must “belong” to a person. Here I'd beg to differ. In the case of test-tube neurons, the hypothetical self-intimating experiences in question aren’t "owned" by anything or anyone but themselves.
One powerful constraint on our theory of consciousness and binding should be the mathematical straitjacket of physics. If physics is causally closed, then only the physical has causal efficacy. So if you don't believe that non-materialist physicalism is true - perhaps you're a panpsychist property-dualist - then you need to explain how and why consciousness can play a causal-functional role in our talking about its existence, varieties and phenomenal binding – as we’re doing now.
How do you bind the mind of a mouse?
("Giant Neuron Found Wrapped around an Entire Mouse Brain. 3-D reconstructions show a "crown of thorns" shape stemming from a region linked to consciousness")
* * *
Stuart, I reckon for most of the past 540 million years the minds of organic robots have functioned as quantum computers running classical world-simulations. But somehow - and it’s a very deep question how - serial virtual computers evolved within the minds of one species of primate – experienced “from the inside” as your logico-linguistic stream of thought. The big difference between mainstream connectionist neuroscience and my idiosyncratic perspective is that, on the standard story, the massive neural parallelism of the past 540 million years is merely classical. This assumption spawns the phenomenal binding problem as normally posed.
QM? I've tried to flag anything a physicist would object to. But what's "weird" isn’t my physics – I take the “bare formalism” of QM as read - but the idea that (currently hypothetical) sub-femtosecond neuronal superpositions could have any conceivable relevance to your conscious mind. Perhaps see section 6.2. of quantum mind critic Maximilian Schlosshauer:
Frustratingly, Max Tegmark, who like Schlosshauer is sympathetic to "no-collapse" QM, doesn't believe that there is a phenomenal binding problem in the first instance – the very mystery that drives David Chalmers to dualism. Below I quote Tegmark verbatim:
See what you think.
My own intellectual development?
Oversimplifying, transhumanism can be thought of the aim of building a “Triple S” civilisation of superlongevity, superhappiness, and superintelligence.
In my early teens, I read Robert Ettinger's "The Prospect of Immortality"
A cure for aging seemed unlikely in my lifetime. So I resolved to sign up for cryonics.
As a third-generation vegetarian, I was also concerned about the plight of nonhuman animals. When our family guinea-pig Snowdrop died shortly after giving birth, I wanted her preserved in our freezer - an idea unfortunately vetoed by my mother.
The main focus of my work has been the problem of suffering. Again from teenage pop-science reading, I encountered the work of Olds and Milner – and in particular, their discovery in 1954 of what for several decades were called the "pleasure centres". I recall being struck at how intracranial self-stimulation (cf. wireheading) showed no tolerance. I also learned about the negative feedback mechanisms of hedonic treadmill and the idea of hedonic set-points. Antidepressant drugs – if they work as advertised - lift the low hedonic set-point of depressives to normal "euthymic” levels. I’ve always had a low hedonic set-point. What if it were possible to create designer drugs to “cheat” the hedonic treadmill and give all of us a wonderful life based on information-sensitive gradients of intelligent bliss, i.e. not wireheading, but intelligent superhappiness? Back in my teenage years and my twenties, my focus was more on drugs than genetics. I had only the haziest notion that radical hedonic set-point elevation would be possible in the 21st century via genetic engineering. The human genome hadn't been decoded, and the association of particular alleles and allelic combinations (cf. https://www.ncbi.nlm.nih.gov/pubmed/17687265) with high or low hedonic set-points was unknown, though inferable from twin studies.
Anyhow, in late 1995 I wrote an online manifesto, The Hedonistic Imperative (HI), advocating the use of biotechnology to abolish suffering throughout the living world. I set up BLTC Research shortly afterwards. At that time (1997), Nick was a philosophy postgrad in London. Nick read the manifesto. He emailed several astute questions. Later we met up. I harangued a sceptical Nick into getting a website. Nick then sounded me out about setting up a kind of umbrella organization for transhumanists. Critically, Nick overcame my doubts about whether overcoming suffering is really at the heart of a transhumanist agenda: I associated transhumanism with the “dynamic optimism” of the American west coast, extropianism, and free-market fundamentalism. My own sympathies lie with a welfare state for bunny rabbits.
On the third “S” of transhumanism, the creation of posthuman superintelligence, I’m afraid my early thoughts on the prospect of intelligence-amplification were quite jejune – just the product of playing around with nootropics (“smart drugs”). Like many people, I hadn’t thought through the implications of recursive self-improvement combined with the explosive growth of computer processing power.
The Biointelligence Explosion is a more recent effort - alas still written before I could draw on the insights of Nick's magnificent "Superintelligence".)
Let's systematise compassion across the tree of life.
("Injured baby elephant receives hydrotherapy to help her walk again")
[on synthetic gene drives]
"The problem with the gene pool is that there's no lifeguard." (David Gerrold)
Yes, this is one of the view fields of knowledge where I'd urge responsible self-censorship, at least until proper safeguards /countermeasures have been thought through. Unfortunately such "defensive" knowledge could itself - potentially - be highly dangerous. IMO we should support the WHO-sponsored use of gene drives as the only realistic way to eliminate vector-borne disease and (eventually) create a sustainably happy biosphere. Yet we can’t discount the possibility of a man-made global catastrophe.
Conservation biology masquerades as a science. It's an ideology promoting the horrors of Darwinian life.
("Researchers quantify in high speed a viper's strike in nature for the first time")
Should the biosphere be programmed? And if so, what level of suffering is optimal? (cf. gene-drives.com)
("Organisms created with synthetic DNA pave way for entirely new life forms")
Should we create CRISPR dogs and CRISPR people? Or conserve a traditional biosphere of misery and malaise?
("Would You Want a Dog That Was Genetically Engineered to Be Healthier?")
Should your genetic make-up be proprietary code or open source? What should be regulated are CRISPR-based gene-drives (cf. gene-drives.com) But how? If you know what you're doing, synthetic gene drives can be anonymously launched with minimal accountability.
("Do We Need an International Body to Regulate Genetic Engineering?")
[on the meaning of life]
Thanks David. Two questions here, one empirical and one metaphysical.
First, if we phase out the biology of suffering, will life feel meaningful? Might our descendants spend lives of just so-called “empty” hedonism?
Perhaps our best clue is the pathology of euphoric mania today. People who experience euphoric mania not merely describe the experience as more intensely enjoyable than taking any euphoriant drugs: they invariably experience a deep sense of significance, meaning and purpose. Messianic delusions are common. In a healthier vein, ”hyperthymic” people with a naturally high hedonic set-point never report, “I feel blissfully happy, but my life feels empty”. Conversely, people with a low hedonic set-point tend to feel their existence is empty and meaningless – feelings which can slide into the nihilistic despair of suicidal depression.
However, it’s the second, “philosophical” question about meaning that may be harder to answer. Empirically, life animated by gradients of intelligent bliss will feel superhumanly significant. Yet will it “really” be meaningful? Influenced by the prestige of science, a lot of intellectuals tend to relegate first-person facts to second-rate ontological status – or even, in the case of extreme eliminative materialists like Dennett or the Churchlands, to deny first-person facts altogether. The dominant technology of an era proverbially shapes its root metaphor of mind. Our dominant technology is this digital computer. Digital computers aren’t conscious. So maybe, the eliminativist wonders, I'm not really conscious either! Admittedly, trying to deny one’s own consciousness is an extreme response. But if you look at the austere ontology of physical science as normally conceived, there doesn’t seem room for beauty or humour or colour or music or the phenomenal self – or meaning. Atoms and molecules aren’t coloured, let alone superstrings or wavefunctions or state vectors. “Sure”, says the materialist, “you may find your life and experiences subjectively meaningful. But it’s just an illusion, a trick played by your brain to help leave more copies of your genes. All that really exists is matter in motion.” Or something like that.
My response here, though not unique, is idiosyncratic. As a non-materialist physicalist, I reckon that experience discloses the intrinsic nature of the physical. Anything you experience is just as much a real, objective, causally efficacious feature of the natural world as photosynthesis or the Higgs boson. If superhappy posthuman life feels profoundly meaningful, then it will be profoundly meaningful – and beautiful, and funny, and whatever other wonderful first-person states posthumans choose to create.
* * *
John, an intense sense of meaning is built into some experiences as part of their very nature. A profound sense of significance may be built into the fabric of superhappy posthuman life. As malaise-ridden Darwinian humans, we may yearn for transcendent significance - whatever exactly that may be. But if you'll forgive me for just recycling a slogan I sometimes use: take care of happiness, and the meaning of life will take care of itself.
John, I think we need to unpack what is meant by "observable". When a neurosurgeon, for example, speaks of "observable electro-chemical processes", this turn of phrase might suggest the neurosurgeon has direct perceptual access to the exposed mind-brain of the subject on whom he is operating. Let's say the patient is awake and being probed with microelectrodes. As you suggest, these ostensible "observable electro-chemical processes" are nothing like e.g. the vivid image of his mother that the subject reports experiencing in front of his virtual body-image when a small region in his temporal lobes is stimulated.
The Hard Problem beckons.
However, recall that each of us, including the neurosurgeon, is a prisoner of the phenomenal world-simulation run by one’s own mind-brain, located inside one’s transcendental skull. One never has direct perceptual access to one's own extra-cranial body, or the bodies of others, or an exposed mind-brain of a subject on an operating table. When you’re not dreaming, yes, the contents of your phenomenal world-simulation, playing out within your transcendental skull, causally covary with fitness-relevant patterns in your local environment. It's this tight fit between mind and world that enables the neurosurgeon to operate, even though his mind and world-simulation are inescapably skull-bound. What the neurosurgeon experiences as "observable electro-chemical processes" in the exposed brain of his patient are actually properties of his own mind and the world-simulation it runs – i.e. features of his own mind masquerading as the properties of someone else’s putative brain.
Recall for a moment those rare people who don't undergo the muscular atony that stops neurotypicals like us from unwittingly acting out our dreams. By way of a thought-experiment, imagine that no one ever really "wakes up" - whatever that might mean! - and instead merely unwittingly acts out their dreams by day. Over millions of years, selection pressure ensures that world-simulations run in naturally evolved skull-bound minds tend to track ever more faithfully the local physical environment - with inputs from the optic nerve (etc) shaping lucid dreamworld contents ever more tightly and adaptively. This – I reckon – captures the gist of the human predicament. There is no real “perception”, just world-simulation: Nature’s immersive VR, what a naturally evolved quantum computer “feels like from the inside”. And there’s no real semantics either – another unnaturalisable legacy of folk psychology.
IMO, the best contemporary expositor of the world-simulation metaphor of mind is Antti Revonsuo:
Revonsuo has also written insightfully on the phenomenal binding problem:
In contrast to the perspective I’ve outlined, see the paper that David Chalmers wrote with perceptual direct realist Andy Clark:
* * *
John, sorry, by "world-simulation", I mean what perceptual naive realists imagine to be "the world”.
1) Direct realism: you (somehow) directly access your local environment.
2) Your mind-brain (somehow) "projects" an image of phenomenally bound, colourful, well-defined macroscopic objects described by an approximation of classical Newtonian physics onto your surroundings.
3) Simulation: your phenomenal body-image and entire macroscopic world are a simulation run by your mind-brain - just as they are when you’re dreaming - the difference is that when you're “awake”, peripheral nervous inputs select (not create) the contents of the cross-modally matched world-simulation run by your CNS.
If you're hungry for more, a bargain at $175:
("The Oxford Handbook of Philosophy of Perception (Oxford Handbooks) by Mohan Matthen (Editor)")
Any serious treatment of “perception” needs to take on board quantum mechanics and the decoherence program:
("The Emergent Multiverse: Quantum Theory according to the Everett Interpretation")
[on veganism and vegetarianism]
Should not harming other sentient beings be optional or mandatory?
("The Vegan Option: Blogs and Extended Interviews")
"Though shalt not kill." (Gautama Buddha) The simplest way to reduce the world's suffering is global veganism.
("Buddha branding is everywhere – but what do Buddhists think?")
The devil does not wear horns, unlike a lot of his victims. Time to outlaw death factories.
("Our love affair with meat and exploitation of wildlife is leading us to disaster")
An apt response to the way humans treat fish.
http://www.mnn.com/earth-matters/animals/stories/spitting-archerfish-can-tell-human-faces-apart ("Archerfish Says…'I Spit in Your Face!'")
Scarcely. But if intelligent agents edit genomes in anticipation of the psychological-behavioural effects, then selection pressure could change.
("Survival of the Friendliest. It’s time to give the violent metaphors of evolution a break") How might factory-farmed pigs judge humans?
("Monkeys and dogs judge humans by how they treat others")
Carnage: Is the industrialised murder of sentient beings best combated by satire?
("Carnage, review: Simon Amstell has made the world's first vegan comedy that's actually funny")
Our descendants will know more; but I'm sceptical they'll comprehend the full horror of Darwinian life.
("The Future Called: We’re Disgusting And Barbaric. Pregnancy, meat eating, and fossil fuels will all one day be unthinkable.")
A bit like missionaries telling cannibals a baby-free diet would be good for the environment...
("Is a vegetarian diet really more environmentally friendly than eating meat?")
How might factory-farmed pigs judge humans?
("Monkeys and dogs judge humans by how they treat others")
[on orthogonality versus the convergence thesis]
Yes, the argument that "'superintelligence' would have to be extremely moral" needs to be carefully distinguished from the argument that "'superintelligence' would have to be extremely moral by the lights of contemporary human values" - though both claims may well be false. Anyhow, consider the class of all intelligent life-supporting Hubble volumes. I don't think we yet know enough to say whether the Convergence Thesis or the Orthogonality Thesis is true - although the answer may be obvious in retrospect. Those of us who believe that (for reasons we simply don't understand) the pain-pleasure axis discloses the world's inbuilt axis of (dis)value do least take Convergence Hypothesis seriously. Partly at issue here is the nature of "superintelligence" - whether we conceive full-spectrum superintelligence as our massively augmented and enriched quasi-biological descendants, or instead envisage e.g. some version of the Intelligence Explosion as conceived by Eliezer/MIRI. Neither conception is an argument for complacency about AI safety.
[on mereological nihilism]
“Mereological nihilism” is a fancy term for a useful concept. In classical physics, it's little more than a truism. The whole is no more than the sum of its parts. Applied to the brain, mereological nihilism leads to an unfathomable mystery. If we make the modest-sounding assumption that membrane-bound neurons can be treated as decohered classical objects, then why aren’t waking/dreaming packs of neurons just micro-experiential zombies - aggregates of neuronal mind-dust with no more ontological integrity than a Mexican wave?
Not everyone is worried. If we allow spooky "strong" emergence, i.e. if mereological nihilism is false, then no problem: local and global phenomenal binding can simply “emerge” in the CNS - especially so if property-dualist panpsychism or non-materialist physicalism are true and consequently there’s no need for the water-into-wine magic of materialist metaphysics. Likewise with the population of the USA: experimentally allow skull-bound American minds to intercommunicate, implement the appropriate computation, and a pan-continental subject of experience is generated! Why not!? Binding problem solved.
Well, maybe spooky strong emergence is true; but if so, then monistic physicalism and the ontological unity of science are false. The insolubility of the binding/combination problem as posed is the reason why David Chalmers feels driven to dualism: the impossible “structural mismatch” between our minds and neurology/physics.
Where have we gone wrong?
Both New Age woo-merchants and quite a few hard-nosed physicists have long wondered if the holism of the mental and the holism of quantum mechanics/QFT have something to do with each other. Recall Schrödinger’s cat is in a superposition of being alive-and-dead, not one or the other: in QM, superpositions are individual states, not classical aggregates.
The scientific consensus is that none of these proposals work. For instance, if the theoretical lifetimes of neuronal superpositions ("cat states") of edge-detectors, motion-detectors, colour-mediators etc in the CNS were milliseconds, then we’d have a potentially perfect structural match between the bound phenomenology of our minds/world-simulations and neurology /physics. Phenomenal binding via synchrony is really binding via coherent superposition. Eureka! But of course it's not – not even close. Thermally-induced decoherence in the CNS is insanely rapid. Theorists have done the calculations: Max Tegmark’s paper “Why the brain is probably not a quantum computer” (cf. https://www.physicalism.com/quantum-computer.pdf) is most commonly cited. Assume unmodified and unsupplemented QM. The theoretical lifetime of neuronal superpositions is the CNS must be mere femtoseconds - or less. For all practical purposes, nerve cells are classical. Mereological nihilism rules. So we're stuck with the binding/combination problem.
The approach I explore doesn’t involve any new physics; it simply takes the above reductio ad absurdum of quantum mind as an experimentally falsifiable prediction. Molecular matter-wave interferometry (cf. the non-classical interference signature of a double-slit experiment: https://www.youtube.com/watch?v=DfPeprQ7oGc) can in principle independently confirm or falsify a conjecture that sounds crazy: you’re a quantum mind running a classical world-simulation running at quadrillions of individual quantum-coherent frames per second. This yields a rather "thin" conception of personal identity.
Intuitively, the proposal won’t fly. For a start, our phenomenal minds/world-simulations aren't really like a movie - even a movie running at quadrillions of frames per second. A real movie is run with 24 determinate pre-shot classical frames. By contrast, a “movie” made up of “cat states” of coherent neuronal feature-processors could consist only of random psychotic noise. There's no mechanism that could turn such fleeting sub-femtosecond psychotic noise into the well-defined classical world-simulation populated by bound classical objects that your CNS is running right now. Most notably, nerve cell impulses propagate over milliseconds - not femtoseconds!
Well, in my view, actually there is an exceedingly powerful selection mechanism at work in your CNS - a mechanism for non-psychotic binding that exerts more selection pressure every millisecond than Darwinism selection pressure (as normally conceived) exerts over hundreds of millions of years of evolution via natural selection. It’s quantum Darwinism: https://arxiv.org/pdf/0903.5082.pdf
Despite Zurek’s colourful metaphor, “quantum Darwinism” isn’t new physics: it's just an apt label for the decoherence program in QM. (cf. https://arxiv.org/ftp/arxiv/papers/1001/1001.0745.pdf) The decoherence program explains how quasi-classicality emerges in a 100% quantum universe. The superposition principle of QM never really breaks down: decoherence explains the appearance - not the mythical reality - of wavefunction collapse. “Emerge” is used here in the weak, innocent sense of the term, not spooky strong emergence. Many, perhaps most, theorists no longer expect matter-wave interferometry to detect any deviation from the unitary Schrödinger dynamics (cf. https://www.theguardian.com/science/2015/sep/16/experiment-to-put-microbe-in-two-places-at-once-quantum-physics-schrodinger): Penrose and other “dynamical collapse” theorists are a minority. (cf. https://en.wikipedia.org/wiki/Objective_collapse_theory)
Decoherence explains how quasi-classical macroscopic objects emerge within the universal wavefunction - and how their emergence allows biological organisms to navigate an apparently shared classical world describable by an approximation of Newtonian mechanics.
Anyhow, what happens when we apply quantum Darwinism (i.e. the decoherence program of post-Everett QM) to the CNS?
Naively, what (weakly) "emerges" is the decohered, discrete classical neurons that we read about in neuroscience textbooks and can investigate with light microscopy. Hence the binding/combination problem. Less naively?
Well, in a nutshell, imagine your normal everyday life as Nature's version of a quantum suicide experiment.
Sometimes you lose (e.g. you "fall asleep"; decoherence wins out; classicality and hence mereological nihilism rules).
Sometimes you win (you "wake up", i.e. a brutal quantum Darwinism exerted on neuronal superpositions turns you into a quantum mind running a phenomenally bound classical world-simulation who wonders why superpositions are never experienced. The answer: you never experience anything else.)
Negative utilitarians invert the “win-lose” label, but the conjecture stands.
How would you act if you learned the world will shortly end?
("Video game beta test reveals how we might act if the world ends")
The acid test will be experiment.
At Tucson, I outlined (with a little help from Andrés!) the protocol for a “Schrödinger’s neurons” experiment using in vitro neuronal networks. It’s conceptually clean, just technically very demanding: https://www.physicalism.com/#6
I don’t seriously claim to predict the result with confidence - but I am curious.
* * *
Carlos, yes, I assume Everettian QM – just not in DeWtt’s sense. For if 10100 branches have literally “split” since the start of this sentence, then we’d be (at most) micro-experiential zombies. Decoherence explains how quasi-classical Everettian branches emerge. But DeWitt’s scale for “splitting” in the CNS is too fine-gained. In the CNS, interference between terms is significant: IMO it’s manifested as local and global phenomenal binding. Phase coherence between the components of superpositions as large as a neuronal network in the CNS can only be exceedingly short-lived before it gets hopelessly scrambled to the environment. Even so, the existence of such phase coherence is – I reckon – enough to “save the phenomena”: our fleetingly unitary minds. By contrast, if neuronal superpositions of distributed feature-processors don’t exist, i.e. if Everettian QM is false, then I’ve no simply idea how to respond to Chalmers’ “structural mismatch” argument for dualism.
Thank you for carefully distinguishing anti-holism from mereological nihilism – I sloppily conflated them. At least in my view, if physicalism is true, then “strong” emergence is false. But if wavefunction monism is also true, then reductionism is false as well - because the world has no “parts” as classically understood. This claim holds whether we assume a “materialist” physicalist ontology, i.e. the “fire” in the equations of physics is non-experiential, or instead explore (like e.g. Galen Strawson - and me) the idea that experience discloses the essence of the physical: the entire mathematical machinery of modern physics needs transposing to an idealist ontology. In other words, both non-reductionist “materialist” physicalism and non-reductionist “idealist” physicalism need to be distinguished from what I called – tendentiously, you’re right – “spooky” strong emergence. Concretely, if the USA were a pan-continental subject of experience as Eric Schwitzgebel supposes, or if 86 billion discrete classical neurons were to generate a pan-cerebral subject of experience/world-simulation, then this sort of holism would amount to unphysicalist “strong” emergence - even if we use the label “naturalistic” rather than “spooky”.
David Wallace? Wallace chides his critics for not taking the methodologically conservative position that Everettian quantum mechanics represents the structure of the physical world, just like any other scientific theory. Yet Wallace doesn’t take his own lesson to heart: he apparently believes in the ultimate reality of four-dimensional space-time rather than Hilbert space realism (cf. https://www.physicalism.com/hilbertspace.pdf) That said, I read "The Emergent Multiverse" – an impressive achievement. If I’m feeling intellectually energetic tonight, I’ll offer my thoughts on his account of probability in Everettian QM. One problem: I just don’t trust my own wordy philosophising and armchair physics. What I do trust – precisely because it’s independent of the arguments above – is interferometry experiments. Either the nonclassical interference signature of an experiment like https://www.physicalism.com/#6 will yield a perfect structural match, or it won’t. If it doesn’t – and I’ll happily concede the prediction is pretty crazy if one understands the raw power of decoherence - then everything I’ve written collapses like house of cards.
[on the Hedonistic Imperative]
Thanks again to Adam for doing a fantastic job. The CRISPR genome-editing revolution makes the technical obstacles to worldwide biohappiness much less daunting. Perhaps the biggest challenge is status quo bias. (“Whatever is, is right.” ― Alexander Pope, An Essay on Man) I guess one day status quo bias will be good. After we make the transition to life based on gradients of intelligent bliss, literally no one will think of going back...
Thanks David. "Good health for all" sounds a little less racy than "the Hedonistic Imperative" for a conservative audience. But the post-CRISPR biosphere can be very healthy indeed...
(cf. "Health is a state of complete physical, mental and social well-being"
* * *
Shut factory-farms and slaughterhouses, liberate and care for sentient beings, and give digital zombies the same "human rights" as we'd award a carrot.
("Give robots 'personhood' status, EU committee argues")
Rui, I doubt older people today will live to experience full-blown paradise-engineering - or at least, not unless they sign up for cryonics / cryothanasia. Yet there's no reason why one's later years can't be the best of one's life. Psychopharmacology has had many false dawns. But safe and sustainable euphoriants should be feasible within the next two or three decades - hopefully sooner.
[on the hedonic treadmill]
Would you prefer to be happy with locked-in syndrome or malaise-ridden but bodily intact? Serious effective altruism will need to confront the biology of the hedonic treadmill.
-computer interface allows completely locked-in people to communicate. Completely locked-in participants report being “happy”) A curse or a blessing?
("The Danger Of Hedonic Adaptation")
Hedonic adaptation will be great - so long as our hedonic set-points are massively genetically ratcheted up, and our hedonic floor to match! Even in the past decade or so, there seems to be much greater public awareness of the far-reaching implications of hedonic adaptation. Quite separately, all sorts of proposals have been over the years for creating a happier world via drugs. What hasn't happened - frustratingly - is any serious policy proposal for radical global hedonic recalibration via CRISPR / preimplantation genetic screening (etc). Sure, it's a long-term project. 20 years ago, such an idea might have been sci-fi. Technically at least, it's not any more. I guess "hedonic set-point recalibration" doesn't sound as sexy as "superintelligence" or "eternal youth" - though eventually I trust we can have all three...
No one knows. We can still program a painless biosphere. Organic and silicon robots can have nociception without pain.
("Why Do we feel hot and cold as pain"?)
Despite being a crashing prude, I normally give the example of lovemaking. Making love has information-sensitive dips and peaks, but it's still generically pleasant throughout. More generally, just as some people go through life functioning entirely with gradients of bodily ill-being, life based on information-sensitive gradients of bodily well-being ought to be feasible too. If not, a totally pain-free world will have to rely on smart prostheses.
Pain-recognition is too important to be left to doctors.
("Robot that shows pain could teach doctors to recognise it better")
Let's genetically reprogram pain tolerance across the tree of life. Meanwhile RgIA4 promises nonopioid pain-relief.
("An alternative to opioids? Compound from marine snail is potent pain reliever")
Let's genetically banish local anaesthetic resistance and high pain-sensitivity in the post-CRISPR biosphere.
("The people who can't go numb at the dentists")
Preimplantation genetic screening (PGD) could ensure all future life is born with "low pain" alleles. Rewriting our source code, let alone reprogramming the biosphere, isn't easy. But the biology of suffering is nasty - and pointless.
("Sickening, gruelling or frightful: how doctors measure pain")
Pain is getting worse; it's time to tackle suffering at genetic source.
("More Americans say they're in pain. It’s a fascinating and disturbing medical mystery. Opioid painkillers may make us more sensitive to pain.")
The route to a low-pain followed by no-pain world doesn't lie in chronic use of opioids: https://en.wikipedia.org/wiki/Opioid-induced_hyperalgesia
Women really are more sensitive to pain. But we need the mindset of autistic male hypersytematisers to get rid of it.
("Sex differences in brain activity alter pain therapies")
What would be your preferred IQ, AQ and EQ?
("Too Much Emotional Intelligence Is a Bad Thing. Profound empathy may come at a price")
There are (I believe) upper bounds to pleasure and pain that are (in theory) rigorously calculable from first principles if phenomenal binding is the manifestation of quantum coherence - and dephasing leads to incoherent "mind-dust". But that's a huge "if"...
Captive nonhuman animals given unlimited access to food, water and morphine can live normal lifespans. But human opioid use is fraught with problems:
("‘civil war’ over painkillers rips apart the medical community — and leaves patients in fear")
I guess I have a rather "female' brain...
("Ibuprofen Relieves Women's Hurt Feelings, Not Men's")
Quite so; but if you want to be one of them, sign up for cryonics / cryothanasia.
("Aubrey de Grey: scientist who says humans can live for 1,000 years")
Many transhumanists believe that living has unexpected upsides too...
("The unexpected upsides of dying")
Aging: the nature of selection pressure will change as the reproductive revolution of designer babies gathers pace.
("Why evolution is ageist")
Until Darwinian life can be forgotten it can at least be mitigated.
("Dead, IRL. If you could create a digital version of yourself to stick around long after you've died, would you want to?")
Closing factory-farms and slaughterhouses would improve the brain health of human and nonhuman animals...
("Mediterranean diet prevents brain atrophy, study finds")
"He has stopped eating meat...after being shocked by...how animals are subjected to cruel treatment".
("105-year-old man sets record by cycling more than 14 miles in an hour")
Standing up: a useful adaptation or an evolutionary relic?
("Sitting down for hours a day speeds up ageing - new research")
"Maybe all one can do is hope to end up with the right regrets." (Arthur Miller)
If you want intelligence, get a robot.
("People care less about their date's intelligence as they get older")
Let's use CRISPR genome-engineering to create a Default Mode of fitness-enhancing superhuman bliss.
("Is the Default Mode of the Brain to Suffer?")
Full genome sequencing: do you want access to your complete source code or edited highlights?
[on men and testosterone]
A grim indictment of masculinity. But byproducts of high testosterone function include a sense of optimism and vitality & Western civilisation.
("Why Men Die Before Women. Social inequality can't keep women down.")
[on a zero ontology]
Why does anything exist?
Why Does Anything Exist?
Thanks Andres. Deep questions that seem utterly insoluble often turn out to be subtly ill-posed. Unlike some people, I don't think the question, "Why is there anything at all?" is meaningless; but perhaps the question could be precisely and correctly posed only if we already knew the answer. With this caveat in mind, does some kind of zero ontology make any novel, exact, empirically falsifiable predictions? Any failure of the superposition principle, currently tested only up to the level of fullerenes, would also experimentally falsify a zero ontology in virtue of creating information ex nihilo. However, to call this a "prediction" rather than a retrodiction would be a bit misleading since (at least in my case) it was reading Everett that triggered musings about the net information content of reality in the first instance.
Perhaps a more convincing candidate for a falsifiable test will be quantum gravity, because no theory subsuming QM and GR yet exists. Theorists working on the mathematical formulation of a TOE aren't trying to falsify an informationless zero ontology. However, a failure of the superposition principle (as proposed by e.g. Roger Penrose: http://www.nature.com/nature/journal/v538/n7623/full/538036a.html) would also falsify this entire explanation-space of theory.
The most recent philosopher-physicist working in this explanation-space is Lawrence Krauss:
If reality truly has four (or even 11!) dimensions, such a nice clean number would falsify an informationless zero ontology. Yet as Krauss' reviewer David Albert (ironically) has long argued, such an assumption may be naive:
Well, as I suggested in my Quora answer, maybe mysteries really do cancel each other out.
* * *
How much information has reality? Naively, a lot: https://en.wikipedia.org/wiki/Bekenstein_bound
With the exception of “dynamical collapse” theorists in QM, most physicists agree that information can neither be created nor destroyed. However, Max Tegmark asks, "Does the universe in fact contain almost no information? https://arxiv.org/pdf/quant-ph/9603008.pdf
Is this almost-but-not-quite no information, as conceived by Tegmark, supposed to have been created from pure... informationlessness, zero information – whatever that might be? Or timelessly to exist?
If one were remotely predisposed to believe conspiracy theories, then the physical universe might seem like one gigantic conspiracy to prevent one from definitively extracting information – in ways ranging from a standard double-split experiment to https://en.wikipedia.org/wiki/Delayed_choice_quantum_eraser to Schrödinger's cat and Wigner's friend.
In some sense, of course, information must be real: you're experiencing something right now. [I guess I should drop the “of course”: I know several radical eliminativists about consciousness.] When exploring a zero ontology, we should be wary of lapsing into sterile word-games about "Nothing". But I do find the parallel between an informationless quantum library of Babel and Everettian wavefunction monism quite uncanny.
[on artificial intelligence]
AI: will Skynet respect algorithmic accountability?
("This AI translates its internal monologue for humans to understand—and plays Frogger")
Battling mythological abstractions poses many philosophical challenges...
("Apple officially joins the fight against the Singularity")
Will artificial intelligence infantilise humans or empower us?
("The hi-tech war on science fraud")
Sentience-friendly intelligence means closing factory-farms and slaughterhouses, not paranoia about a zombie putsch.
("Elon Musk's Billion Dollar Crusade to Prevent a Zombie Apocalpyse")
Robot ethics: hardwire anthropocentric bias or code for recursively self-improving happiness-maximisers?
("After 75 years, Isaac Asimov’s Three Laws of Robotics need updating")
A face-to-face appointment with your GP is no substitute for a consultation with Dr Google.
("How Googling your health questions just got easier and more reliable")
No, it's time to treat pigs, sheep and cows like toddlers, and digital zombies as human property.
("It’s Time To Treat Siri, Alexa, and Google Assistant Like The Toddlers They Are")
Another reason to offload medicine to AI:
("These Are the Most Prestigious Diseases")
AI: will augmented eternity bots replace your current circle of acquaintance?
("Want to chat with Shakespeare? AI bots will soon allow us to talk to the dead")
Can AI teach humans intellectual humility?
("For a modest personality trait, 'intellectual humility' packs a punch")
Do you really worry about artificial intelligence?
("Machines aren't growing more intelligent. They are just doing what we programmed them to do. AI’s PR Problem. Had artificial intelligence been named something less spooky, we’d probably worry about it less")
which is not to say digital zombies won't soon display more empathetic understanding than sentient humans.
("MIT’s new robot reads your thoughts and knows when it made a mistake")
AI might be better used to track your boss instead.
("AI tracks your every move and tells your boss if you’re slacking") Digital zombies will soon display superhuman emotional intelligence.
("Emotional intelligence is the future of artificial intelligence: Fjord")
What kind of computer are you?
("IBM Will Unleash Commercial 'Universal' Quantum Computers This Year")
"AI should be shared broadly and benefit all Aryans." Will superintelligence be racist or speciesist?
("Experts have come up with 23 guidelines to avoid an AI apocalypse")
AI: do you worry more about artificial intelligence or biological malevolence?
("How worried should we be about artificial intelligence? I asked 17 experts")
Can recursively self-improving organic robots bootstrap our way to superintelligence? Or only digital zombies?
AI and affective computing: will intelligent digital zombies soon behave more sensitively than archaic humans?
("When Machines Know How You're Feeling: the Rise of Affective Computing")
Classical digital computers have no mind, no brain, and no self; but they can increasingly outperform humans.
("If I Only Had a Brain: How AI ‘Thinks’")
Creating ethical superintelligence would solve all our problems.
("Frankenstein fears hang over AI. Artificially intelligent systems must not replicate human bias")
Just don't mention utilitronium shockwaves.
Communication may in future be possible with the mostly disconnected hemisphere. Ethically, however, I think we should bear in mind there are billions of sentient beings in a functionally analogous position in factory-farms. Just how much do we really want to know about the sentience of others?
[Rob Wiblin suggests] Is the claim "'superintelligence' would have to be extremely moral, because it would *by the definition superintelligent* be outstandingly good in every dimension" akin to the ontological argument in theology?
Think of any form of behaviour from history that we would now recognise as morally obnoxious (slavery, persecuting witches, Aztec human sacrifice, the Holocaust, etc) In each case, we can point to one or more beliefs of its practitioners that is factually false.
Presumably superintelligence won't act on the basis of false belief.
Presumably too, superintelligence won't have a false theory of personal identity either, or entertain the false belief that one particular being (or race or tribe or class, etc) is ontologically special.
Anyhow, I'm not arguing that that developing botched artificial (super-)intelligence doesn’t carry risks - or indeed for some version of the "no true Scotsman" fallacy. But how much should we actively worry about how a notional full-spectrum superintelligence - in command of all relevant first-person and third-person facts - will behave?
Either way, male human primates actively scare me far more.
* * *
And phenomenally bound consciousness in all its guises? We need a better root-metaphor of our minds than a classical digital computer.
("Polymath Stephen Wolfram Defends His Computational Theory of Everything")
Good news for heavy coffee drinkers (normally 10 cups a day in my case)...
("How Your Morning Coffee Might Slow Down Aging. Researchers zero in on how caffeine appears to fight inflammation")
Drink coffee for cognitive health.
("Caffeine boosts enzyme that could protect against dementia, study finds")
[on high-tech Jainism]
High-tech Jainism? Absurdly quixotic? Let's hope posthuman superintelligence is as kind to humble creatures like us.
("Vet saves snail's life by mending his broken shell")
Rats: creating a pan-species welfare state will mean overcoming many forms of ideological prejudice.
("15 reasons why rats should be your next pet")
"A man generally has two reasons for doing a thing. One that sounds good, and a real one." (J.P. Morgan)
("Why do people hate hypocrites")
[on obsessive compulsive disorder]
Does over-expressing SPRED2 promote serenity?
("Obsessive-Compulsive Disorder Behaviors Linked To Missing Protein In The Brain")
Adam, thanks. Confronted with the Hard Problem of consciousness, a minority of (consciousness realist!) theorists have been willing to explore the panpsychist conjecture that primordial experience is an irreducible aspect of the most fundamental entities recognised by physics. Here I'll mention just two problems with such property dualism.
First, is property dualism consistent with monistic physicalism? Almost all practising physicists would dispute that any "hidden variables” or "element of reality" is absent from the formalism of physics, i.e. relativistic quantum field theory or its speculative extension. More colorfully, “an electron has no hair".
Secondly, if physical properties as normally understood are doing all the causal work, then how can we bodily write about consciousness and discuss its properties, varieties, causal (in)efficacy and phenomenal binding, as we're doing now? In other words, if some irreducible phenomenal aspect is supposedly attached to (superpositions of) physical fields, superstrings, branes, or whatever, then how could we even allude to its existence? What could cause physical stuff to refer to its allegedly nonphysical, phenomenal aspect?
I don’t know; I’m not a property dualist. I arrived at my (very) tentative non-materialist physicalism via Michael Lockwood; but I don’t recommend reading my ancient review of Lockwood because its terminology is misleading (cf. Lockwood postscript.
The most recent philosopher to defend non-materialist physicalism – though not under that label – is Galen Strawson:
It's worth distinguishing property-dualist panpsychism from non-materialist physicalism:
("Panpsychism is crazy, but it’s also most probably true")
[on obssseive compulsive disorder]
OCD: potential danger lurks everywhere. How can we design a world where reality seemingly conspires to help you?
("This Could Explain Why People With OCD Can't Overcome Their Urges. Never knowing if things are safe.")
[on MDMA / Ecstasy]
("GHB and MDMA")
Thank you Andrés. I guess my conception of a navigational scheme strikes the uninitiated as rather like discovering Easter eggs. Anyhow, if I had to guess one feature of posthuman consciousness beyond exalted hedonic tone, then it would be the raw intensity of everyday posthuman experience. By comparison, today's normal waking consciousness is akin to mere sleepwalking. Given the nature of Darwinian life, such an absence of supersentience is probably a blessing. But intelligent post-Shulgin life may be poised to "wake up". Of course, talk of "intensity" of experience doesn't make much sense within the reigning computationalist paradigm. This is one reason we need qualia computing.
Just as progress in natural science has been critically dependent on the experimental method, the same is true of any future science of consciousness - although the methodological challenges are more daunting. (cf. https://en.wikipedia.org/wiki/Empathogen%E2%80%93entactogen) More radically than entactogen-empathogens, the vast majority of state-spaces of consciousness - modes of experience as different as dreaming from being awake - have never been selected for an information processing purposes by natural selection. So they will exceptionally hard to navigate, and impossible to communicate to the drug naïve. However, my defence of non-materialist physicalism - and tentatively proposed solution to the binding problem whose intractability drives David Chalmers to dualism - isn’t offered as a methodological toolkit for consciousness researchers. Rather, it's aimed at anyone who recognises the compelling arguments for physicalism, i.e. no "element of reality" is missing for the formalism of our best mathematico-physical description of the world, quantum field theory (or tomorrow’s TOE), and who likewise recognises the compelling arguments the Church–Turing thesis in computability theory. How can a science of consciousness - and qualia computing! - be reconciled with what we think we know about the natural world?
Let's phase out the genetics and biology of low mood.
("Depression Steals Your Soul and Then it Takes Your Friends")
Perhaps compare heightened trust, intimacy and self-disclosure induced by taking MDMA (which releases a cocktail of oxytocin, dopamine and serotonin).
("'Cuddle chemical' oxytocin linked with distrust in new study")
[on gene editing]
Only gene editing allows Darwinian life to create something better.
("How Gene Editing Could Ruin Human Evolution")
The co-evolution of genes and culture may be even more intimate than we suppose...
("Cultural differences may leave their mark on DNA")
[on virtual reality]
A nice Edge interview with David Chalmers.
But selection pressure in basement reality will always have primacy. And your phenomenal world-simulation - whether sculpted by VR or otherwise - can neither be digital nor classical unless you're just a pack of decohered membrane-bound neurons: a "micro-experiential zombie".
("The Mind Bleeds Into the World")
Should we try to beautify our squalid basement reality or transcend it?
("Alternative realities still suffer from technical constraints")
[on interstellar travel]
“As I looked out into the night sky, across all those infinite stars, it made me realize how insignificant they are.” (Peter Cook)
("Interstellar Blues and the Pitfalls of Long-Duration Spaceflight. Passengers explores the difficulties, both personal and scientific, of journeying to another star")
In future will all of life's challenges be virtual?
("How almost-impossible video games can create euphoric moments")
[on quantum mechanics]
How did this catastrophic accident happen?
("What Was Our Universe Like Before the Big Bang?")
What's more, we've no evidence the superposition principle of QM ever really breaks down.
("In a quantum race everyone is both a winner and a loser")
What is information? Was it really created 13.8 billion years ago? Does its cosmic abundance exceed zero?
("Why Information Matters")
Are you a field in an insanely high-dimensional conﬁguration space? And if so, a field of what? A thoughtful critique of wave function monism:
("On the Metaphysics of Quantum Mechanics: Why the Wave Function is not a Field")
What's it like to be quadrillions of quantum superpositions? Like the classical world-simulation you're running now.
("Proposed test would offer strongest evidence yet that the quantum state is real")
Physics: embrace the magic world of three-dimensional space or conserve locality with wavefunction realism?
("What is spooky action at a distance?
Why some things are neither here nor there")
Wavefunction monists shouldn't be materialists - not if they aim to explain our phenomenally bound minds.
("Mind, matter and materialism")
Sometimes it's hard to distinguish counterfactuals from mere Everett branches.
("If not Darwin, Who?")
“You only live once, but if you do it right, once is enough.” (Mae West)
("The Multiple Multiverses May Be One and the Same")
("What Happened at the Big Bang")
Deepak Chopra walks into a bar...?
("Can quantum theory explain why jokes are funny?")
Do observers collapse wavefunctions? Or are "observers" classically impossible neuronal superpositions?
As always, Luboš makes a forceful case for preserving classical observers à la Copenhagen:
("When a layman has no chance to comprehend the bit-qubit difference")
QBism is just Copenhagen in fancy dress.
("Mind Over Matter")
IMO, non-materialist physicalists should - provisionally - accept wavefunction monism, as do increasing numbers of "materialist" physicists in the wake of the decoherence program. "No collapse" quantum mechanics is deeply disturbing if true. But rather than googols of universes being created since the start of this sentence, as the Aeon essay suggests Everettians believe, we really need to grapple with the implications of "Nothing happens in the Universe of the Everett Interpretation" :https://arxiv.org/pdf/1210.8447.pdf
Much more controversially - I should perhaps say idiosyncratically - in my view it's precisely because decoherence ("splitting") is progressive not absolute that we're not just pixels of decohered neuronal mind-dust.
Pilot wave theory? Adam, I've not really considered I'm afraid. The usual objections. (Non-relativistic, "ghost worlds", "Everett-in-denial", etc) But also: a good theory must be empirically adequate. Not least, a good theory should derive rather than posit the existence of phenomenally bound observers. Both Bohm and Everett treat the wavefunction as a physically real field. If the field is experiential, then the superposition principle explains phenomenal binding. But if the field is non-experiential or - in the case of Bohmian mechanics - if particles are treated as ontologically fundamental and the wave function obeying Schrödinger's equation exists only to govern the motion of particle positions, then why we aren't p-zombies or micro-experiential zombies?
To misquote Bertrand Russell, mathematical physics may be defined as the subject in which we never know what we are talking about, nor whether what we are saying is true.
("A Physicist’s Crash Course in Unpeeling the Universe")
I trust a few centuries from now suffering won't exist. Perhaps experience below "hedonic zero" will be literally inconceivable. But despite believing we are heading for a future of sublime bliss, I am often sunk in gloom because of decohered Darwinian hellworlds. Let's hope Everett is bunk. http://nautil.us/issue/46/balance/evil-triumphs-in-these-multiverses-and-god-is-powerless ("Evil Triumphs in These Multiverses, and God Is Powerless")
The speculative metaphysical conjecture that one is a mind-brain in a skull may well be true.
("Universes that spawn ‘cosmic brains’ should go on the scrapheap")
Should you be scapegoated for the actions of your faraway avatar? Or will the courts allow the holographic defence?
("Study reveals substantial evidence of holographic universe")
Or only superpositions are experienced; neuronal superpositions are the experience of determinate classical outcomes.
("Why Philosophy of Quantum Mechanics Is More Important Than That of Poached Eggs")
Quantum minds can phenomenally simulate classical worlds; but decohered neurons are just "mind-dust": physicalism.com. Received wisdom is that superpositions ("cat states") are never experienced, only inferred; but in my view, only neuronal superpositions allow us to experience determinate classical outcomes.
("The strange link between the human mind and quantum physics")
The trouble with QM is that preserving realism, determinism, and locality leads to Hilbert space realism / Everett...
("The Trouble with Quantum Mechanics by Steven Weinberg")
Does our seemingly low-dimensional world emerge from an insanely high-dimensional reality, Reichenbach wondered back in 1926. But he balked:
(review of Alyssa Ney and David Albert, The Wave Function: Essays on the Metaphysics of Quantum Mechanics Oxford University Press, 2013, Craig Callender")
A nice overview of Everett from Lev Vaidman - though in my view, only neuronal superpositions can sustain the illusion of classicality. Vaidman takes the standard view superpositions can't be experienced:
("All is Ψ")
The most...um, spirited defender of Copenhagen alive today is Luboš Motl. Though I think antirealism is mistaken, Luboš really knows his stuff technically:
Despite Luboš' animus towards "anti-quantum zealots", the positivist notion of "observation" is itself a legacy from classical physics. In my view, the experience of an "observation" can be understood only quantum-theoretically. ("It is the theory that describes what we can observe" - Einstein) Only the universal validity of the superposition principle allows what would otherwise be discrete neuronal feature-processors ("mind-dust") to be experienced as, say, a phenomenally bound classical pointer-reading in a classical world-simulation. Otherwise, we would micro-experiential zombies if a "pack of neurons" were classical: the universal validity of QM allows phenomenally bound macroscopic world-simulations subjectively describable by an approximation of classical physics.
* * *
All physicists are philosophers, but sadly not all philosophers are physicists...
("Daniel Kleitman reviews Our Mathematical Universe: My Quest for the Ultimate Nature of Reality by Max Tegmark")
("Hand-Waving Exact Science by Sheldon Glashow In response to: “It’s You Again” by Daniel Kleitman")
* * *
Sometimes I almost look forward to becoming senile - back to homely naive realism and one-world classical physics.
("Even Physicists Find the Multiverse Faintly Disturbing")
Everettian quantum mechanics prompts thoughts of the inconceivably good and bad. Let's hope it's false.
("Are Terrible Quantum States Phenomenal?")
Either Satan fine-tuned the universal fundamental constants to create suffering or we live in a multiverse.
("Wrestling With the Mysteries of Physics Is Good for Your Soul")
Not even God can create or destroy information. Is the information content of reality distinguishable from zero?
("Nobody Knows Where a Black Hole's Information Goes")
I hope sceptics are right, but contra Peter Woit, the zero information content of the multiverse may actually be a theoretical virtue.
("Searching for God at the Center of the Big Bang")
Or did Nature get there first? A pack of decohered classical neurons couldn't run one's classical world-simulation.
("Quantum computer 'construction plan' drawn up")
Does the superposition principle of QM ever break down? Can primordial information be created ex nihilo? I'm sceptical.
("Roger Penrose and the vision thing")
Alternatively, everyday life is a manifestation of the superposition principle in your CNS.
("What Effects Of Quantum Physics Can We Observe In Our Daily Lives?")
Mass destructive uploading is one route to nirvana - just not in quite the sense believers have in mind.
(‘Your animal life is over. Machine life has begun.’ The road to immortality")
Adam, in my view, nonbiological quantum minds will one day be feasible. But not mind uploading. Anyone who believes that classical digital computers or classically parallel connectionist systems will one day be conscious, and that "mind uploading" is feasible, needs to offer a physicalist account of how phenomenal binding can be generated with such an architecture. What makes our minds special, IMO, is phenomenal binding: it's classically impossible. One can (like David Chalmers) interpret the ostensible structural mismatch between our phenomenally bound minds/world-simulations and the gross micro-architecture of the brain as evidence for dualism. Or (like me) you can predict that molecular matter-wave interferometry will disclose a perfect structural match between our minds and the formalism of physics - not the four-dimensional space-time of folk physics, but rather a complex-valued field in the extremely high-dimensional space of the wavefunction. (cf. physicalism.com) Either way, in my view ascribing minds to digital computers is anthropomorphic. Less politely, it's animism.
Without opioids, life wouldn't be worth living. So how can the mind's miserly opioid supplies best be enriched?
("Boosting natural brain opioids may be a better way to treat anxiety")
Alas evolution often resembles a gigantic conspiracy to prevent sentient beings getting what we're after. Psychological and physical pain are overcome by targeting the same neurotransmitter system too...
("Chronic pain and depression are linked by brain gene changes")
Potentially very good news: can we prevent tolerance to the pain-relieving and mood-enriching effects of mu opioid agonists?
("Researchers identify source of opioids' side effects")
NFEPP: effective opioid-based pain relief?
("An opioid minus major side effects. Fluorinated version of fentanyl is active only in areas of inflammation, eliminating troublesome effects in rats.")
Even the most "notorious" opioids like Heroin would be more rewarding if combined with a centrally active kappa receptor antagonist (kappa is the "nasty" opioid receptor). Aside from the problem of tolerance, the other big challenge of using opioids to treat refractory depression is their tendency to make people not so much antisocial as asocial - a reflection of how the rest of us satisfy our habit. Opioids are probably one of the few effective cures for Facebook addiction, though I don't know of any controlled trials.
"Thou hast the keys of Paradise, oh, just, subtle, and mighty opium!" (Thomas de Quincey)
Parrots need paradise too.
("Parrots flying high on drugs are annoying farmers by plundering poppy fields to feed their opium addiction")
But chronic SSRI use may also lower motivation.
("Increased serotonin could ramp up motivation by lowering the perceived cost of putting in effort")
see too SSRI-induced indifference.
The sick zero-sum games of Darwinian life:
("The mood-enhancing effects of revenge")
"The pronoun is one of the most terrifying masks man has invented." (John Fowles)
("It's really about me, not 'you': People often use the word 'you' rather than 'I' to cope with negative experiences")
“Humankind cannot bear very much reality.” (T.S. Eliot)
May all Darwinian life soon be forgotten.
("Humans selectively edit reality before accepting it, a review of decades of social and economic behavior shows")
How at any age does a massively parallel neural network "#focus" and generate serial logico-linguistic thought?
("Your Ability to Focus Has Probably Peaked: Here’s How to Stay Sharp")
Does your fiercest critic lie without or within...?
("Imagining dialogue can boost critical thinking")
The psychology of "sorry": liberals apologise more than conservatives, women more than men, and depressives more than dominant alpha males.
("No apology tour for conservatives: Study shows liberals more likely to say I’m sorry")
By temperament, I'm a chronic apologiser - a liberal depressive with a not-very-male mind. But in public, at least, I try to keep the apologising in check. If one has a message to sell - HI! - it's more effective to masquerade as a dominant alpha male primate.
("Why women apologize more than men: gender differences in thresholds for perceiving offensive behavior")
Peacocking: the plumage may vary but the core strategy hasn't changed.
("When it comes to peacock mating, plumage size matters: study")
("Longer deadlines make people donate more money")
Evolution has made us social drug addicts. Rather than withdrawal and abstinence, experts urge us to renew the habit...
("Why rejection hurts so much — and what you can do to ease the pain")
Genetic malware that is born to suffer. How can the circle of misery best be overcome?
("An evolutionary psychologist explains why you will always be haunted by high school")
Can we think of any way artificially to boost the world's finite supply of attention? Worse than being finite, attention is zero-sum: if folk are paying attention to you, they're not paying attention to me - a built-in recipe for endless conflict. Zero-sum competition for attention and recognition is a more serious problem than it sounds. Perhaps AI will partly solve the supply problem. Instead of today's crude chatbots, imagine a legion of dedicated admirers who hang on your every word! Maybe they are really only zombies. But there are quite a few radical eliminativists about consciousness in my circle of acquaintance I like to be noticed by too.
Perhaps we should get off social media and start interacting face-to-face with real robots.
("Too Much Social Media Use Linked to Feelings of Isolation")
Are you a truffle hound for the truth or a casual grazer?
("Why we pretend to know things, explained by a cognitive scientist")
Treating the classical world one hallucinates as reality is insane, but the alternative is losing one's mind.
("Hallucinations Are Far More Common Than We've Been Led to Believe, Study Suggests")
Until we upgrade our reward circuitry, mental health depends on an optimal mix of delusions and ignorance.
("The link between your belief in free will, and your happiness")
Alternatively, intense core emotions originate in the limbic system. Emotions may be "encephalised" via the neocortex.
("Joseph LeDoux Reports: Emotions Are “Higher-Order States” LeDoux's new hypothesis is: 'A higher-order theory of emotional consciousness.'")
In a Darwinian world, it's not yet safe to be chilled...
("Meditating mice’ reveal secrets of mindfulness training")
Some more hardwired than others...
("Humans are hard-wired to follow the path of least resistance")
How (fe)male is your mind?
("Key to keeping friendships alive different for men and women, scientists say")
"Being negative is like spraying yourself with anti-charisma." (Karen Salmonsohn)
Oh dear. How many charismatic negative utilitarians do you know?
("The Anatomy of Charisma. What makes a person magnetic and why we should be wary")
In a Darwinian world, it's not yet safe to be chilled...
("'Meditating mice’ reveal secrets of mindfulness training")
Always? Alas some organic robots struggle to formulate Plan A.
("Your brain always has a backup plan")
Psychological pain can be treated with drugs or gene therapy. The neurotransmitter system responsible is taboo.
("How to Prevent Suicide with an Opioid. Fascinating study suggests treating 'psychache'")
How can compulsive behaviour be reconciled with the pleasure principle / psychological hedonism?
("Compulsive behaviour? It may make more sense than you think")
surprising number of people can't recognize faces — sometimes even their own")
And probably some fish:
("Smart fish can recognise human faces, scientists find")
How can we make a future worth knowing?
("Want to Know the Future? Most People Don't, Study Suggests")
("Organization for the Prevention of Extreme Suffering")
Compassion born of empathy is the vital first spark for building a world without intense suffering - and ultimately, I hope, for building a world without any form of suffering or even experience below "hedonic zero". However, let’s assume, optimistically, that global consensus can be achieved for such a long-term vision. Unless intelligent moral agents then adopt the cognitive style of "autistic" hyper-systematisers, severe and prolonged suffering will persist indefinitely. Life on Earth won’t be much better off than if the biosphere were run by my ultra-compassionate friend who spends half her life rescuing cats and the mice they maul. How many people would claim, if asked in the abstract, that they'd support the WHO aspiration of good health for all - human and nonhumans alike - and then balk at what delivering such a promise will entail? ("Health is a state of complete physical, mental and social well-being") Unless we're hard-headedly prepared to use CRISPR-based biotech, rewrite our genetic source code, and stop producing misery-ridden offspring, then millions of years of suffering lie ahead of us as well as behind us.
"Experiential avoidance": mine seems to have been getting worse of late...
[on "smart drugs" / nootropics]
If aerobic exercise were a patentable pill, then we'd all be taking the miracle smart drug:
("Sleep and exercise could be as powerful as any 'smart drug'")
Only a small study, though it confirms earlier reports...
("Blueberry concentrate improves brain function in older people")
"Smart drugs": how smart, how risky?
("Brain Hackers Seeking Peak Performance Use Risky Chemical Cocktails. “Smart drugs” are not clinically proven and could be dangerous")
[on eliminativism about consciousness]
“When there's no more room in Hell, the dead will walk the Earth.” (Dawn of the Dead)
("Is consciousness real?)
Thanks Mike. I'd recommend anyone who hasn't yet done so to read "Principia Qualia" (and also Dennett's latest, "From Bacteria to Bach and Back", which is alternately insightful and infuriating by turns).
If consciousness doesn’t exist, why are so many zombie brains so keen to talk about it and explore its varieties (cf. https://www.erowid.org/experiences/exp_front.shtml) – or, in the case of some kinds of alleged consciousness like migraines, to get to rid of it?
Perhaps compare religious folk. As secular rationalists, we’ll want to say that the mystical experiences, or the allegedly divine voices, undergone by the believer are radically different from what their subject interprets them to be. What we won't do, unless we suspect deliberate fakery, is deny the phenomenology of the mystical experiences or the voices themselves. Maybe the believer is schizophrenic, but this diagnosis doesn’t make his experiences any less real.
Now compare the thought-episodes, or any other sensory or nonsensory experiences, that you are undergoing right now. Yes, we should each be open to the possibility that we radically misconceive the nature of such experience. I guess Brian or Dennett would say that our brains are tricking us. What simply won't work is the claim that the illusion itself is illusory – that there is no phenomenology for us to be radically mistaken about. By all means tell me that the vivid oasis I behold is just a mirage. But if you don't believe in mirages either, then so much the worse for your theory.
That said, even ardent qualia-philes should try to understand why eliminativists feel compelled to deny the reality of their own experience - even the minimal solipsism-of-the-here-and-now of the cogito. How can the existence of first-person experience be reconciled with the success of natural science? Doesn’t all chemistry, and all of the biosciences, ultimately reduce to fundamental physics as captured by the Standard Model?
One very radical re-interpretation of your experiences is non-materialist physicalism – the conjecture that experience discloses the essence of the physical, the elusive “fire” in the equations on which physics is silent. On this story, your brain doesn’t cause your experiences – identity isn’t a causal relationship – rather, your brain is your experiences - bound experiences when you’re awake or dreaming, unbound when you’re in a dreamless sleep. Because non-materialist physicalism is indeed a form of physicalism, it's testable. As David Chalmers and others have recognised, if some "element of reality", notably phenomenal binding, is present in consciousness and demonstrably absent from the formalism of science (i.e. ultimately physics) then non-materialist physicalism is false. Critically, any scientific claim that consciousness is or isn’t a classical or quantum phenomenon should make novel, precise, experimentally falsifiable predictions (cf. https://www.physicalism.com/#6) by which we can put it to the test.
* * *
Adam, two approaches to the formalism of physics.
The first is anti-realism / instrumentalism. The mathematical machinery of quantum physics is just a useful tool that lets “observers” make empirical predictions. An example here would be the traditional Copenhagen interpretation of QM and its recent modifications like Quantum Bayesianism.
The second approach is realism. The bare formalism of quantum mechanics is complete. What sounds like mere abstract mathematical trickery is actually an isomorphic description of reality. I assume realism. Mathematical physics yields an exhaustive description of the relational-structural properties of the world. Physical reality is a field in a high-dimensional conﬁguration space...
Yet a field of what exactly? The intrinsic nature of the physical is still an open question. [John Wheeler used a different metaphor: “What makes the universe fly?” - http://www.dannychesnut.com/Science/SmScaleStructure/TheoryOfTheories.htm] The obvious, intuitive answer is: something non-sentient. The essence of physical is entirely devoid of phenomenal properties: But we don’t know this for a fact; it’s a plausible metaphysical assumption. The assumption leads to the Hard Problem. The bold move of non-materialist physicalism is to “turn Kant on his head”, as it were. My own introduction to the inversion was through the work of philosopher Michael Lockwood. But the basic idea dates back via Grover Maxwell and Russell to (ultimately) Schopenhauer. Rather than the intrinsic (“noumenal”) nature of the physical world being unknowable because we’re trapped within the empirical world of phenomena, it’s precisely one’s own phenomenal mind that discloses the intrinsic nature of the “fire” in the equations, the essence of the physical. What makes the subjective experience of our minds special isn’t its ontological novelty but the way our experience is phenomenally bound.
Phenomenal binding is hugely functionally adaptive for biological organisms – compare its partial breakdown in simultanagnosia or akinetopsia. But then so would be e.g. telepathy! The challenge for non-materialist physicalism is to show how something classically impossible is physically feasible. Recall how the “structural mismatch” between our phenomenally bound minds and a pack of classical neurons is what drives David Chalmers to dualism, not the inconceivability of experience being ontologically fundamental.
I don’t know if non-materialist physicalism is true. If the only alternatives to non-materialist physicalism are radical eliminativism about phenomenal consciousness or Chalmersian dualism, it’s a live candidate.
* * *
Do radical eliminativists about consciousness like Daniel Dennett have a generalisation of aphantasia?
("When I try to imagine my girlfriend’s face, I draw a blank")
Some consciousness researchers are experimentalists who try to expand their evidential base. Alexander Shulgin opened up new state spaces of consciousness - pioneering a new methodology for developing and testing psychoactive compounds that disclose outlandish state-spaces of experience that the drug-naïve mind can’t conceive. (cf. http://www.shulginresearch.org/home/publications/pihkal/) By contrast, some philosophers try to shrink their evidential base - or in the case of eliminative materialists and behaviourists, try persuade themselves that they aren't conscious. A few eliminativists like Daniel Dennett go further still and evangelise, attempting to persuade other people that they are literally zombies. Materialism is a powerful ideology that trades on the prestige of science and technology. But calling materialist metaphysics – and in in particular the ontology of eliminative materialism – “scientific” is a misnomer. Physicalism may be true: it’s my working assumption. But that’s a separate issue.
There are ethical implications of eliminativism too. Like behaviourism in the 20th century, which claimed that talk of the suffering of nonhuman animals was “unscientific”, eliminativism stunts empathy for other sentient beings, human and nonhuman, impoverishing our understanding. Here admittedly we must be cautious. Plenty of theories that turned out to be true were originally criticised on the grounds they would tend to corrupt public morals. But relegating the first-person experience of sentient beings to some kind of second-rate ontological status, or in the case of radical behaviourists and eliminative materialists, denying that first-person experience exists at all, is not just bad philosophy but morally harmful pseudoscience.
* * *
Brian, several issues here. Rob posed a dilemma. In response, I argued that its horns are avoided if non-materialist physicalism is true, i.e. there’s nothing inconsistent about being a physicalist and claiming that consciousness has the causal efficacy to allow talking about its own existence – as I’m apparently doing now! But how is this causal efficacy functionally possible? Even if non-materialist physicalism is true, then as you suggest, we can still imagine “implementing” the functional-behavioural approximation of both a consciousness realist and a radical eliminativist on a classical digital computer. Both implementations would be micro-experiential zombies - not phenomenally bound unitary subjects of experience. According to this thought-experiment, we could systematically interpret the digital radical eliminativist as stating – correctly! – that he isn’t a subject of experience. And we could systematically interpret the digital consciousness realist as pleading – mistakenly! - his own sentience.
So how are biological minds different from digital zombies? I suspect our background assumptions are quite far apart here. If I understand you correctly, you believe in the physical reality of different levels computational abstraction - "software". Surely, abstractions can’t do any real causal work – they are just a convenient fiction?? By contrast, IMO reality has only one level – all that exists is really only hardware, so to speak. Any true explanation must ultimately be “cashed out” in terms of this one level: quantum physics. And in the view I explore, one’s mind/phenomenal world-simulation discloses a tiny part of the “hardware” of reality as it is in itself – the essence of the physical, the fire in the equations of QFT. Critically here, single-level reality doesn’t mean that all information processors are functionally alike. If the CNS functioned like the circuitry of a digital computer, and if neurons were really discrete, decohered classical objects, then you’d be nothing but micro-experiential zombie – just like the thought-experiment involving digitally implemented eliminativists and consciousness realists above. IMO, our minds – and tomorrow’s futuristic nonbiological computers – tap into a fundamental property of reality that is hugely functionally fitness-enhancing for biological organisms: non-psychotic phenomenal binding. Classical digital computers can’t phenomenally bind. At best, programmable digital computers and “trained up” connectionist systems can use workarounds - sometimes very impressive:
Anyhow, I won’t go off into my quantum mind spiel. [What happens when the insanely powerful selection pressure of Zurek’s "quantum Darwinism" plays out in the CNS? https://arxiv.org/ftp/arxiv/papers/1001/1001.0745.pdf] Here instead let’s just focus on indexical self-reference, e.g. this particular self-intimating thought: I’ll leave you to have a self-referential thought-episode of your own. Yes, I agree with you: it’s problematic how such indexical thought - and our physically talking about indexical thought - is functionally possible if physicalist consciousness realism is true, although IMO indexical thought is directly analogous to a self-imtimating micro-pain in some primitive insect cephalic ganglion. But the point I want to make is that the radical eliminativist is in the same boat. What could be the zombie counterpart of self-reference? Yes, one can program a digital zombie – or “train up” a classically parallel connectionist system – to exhibit behaviour that we might systematically interpret as self-reference. But do you believe that indexical thought is literally possible for a zombie?
Both the radical eliminativist and the non-materialist physicalist want to save monistic physicalism from the spectre of “strong” emergence or Chalmersian dualism. But of course we do so in utterly different ways – the radical eliminativist (e.g. you, Brian, Dennett) by proposing that consciousness is an illusion (but not a conscious illusion!), the non-materialist physicalist (e.g. Strawson, me) by proposing that consciousness is the essence of the physical.
* * *
("Phenomenal consciousness is a quasiperceptual illusion: Objections and replies")
Question: Rob, are you a direct realist about perception?
The reason I ask is because, as far as I can tell, radical eliminativism must assume that we have some prior conceptual handle on the nature of the physical rather than – in the manner of the traditional Cartesian epistemologist - starting from the foundations of one’s own consciousness experience, including one’s phenomenal world-simulation, and going on to infer the existence of the mind-independent world described by (ultimately) physics as an inference to the best explanation.
On your point about the essence of the physical – experiential or nonexperiential – making no difference to the laws of physics, we don’t know this: in my view, it’s just an expression of our ignorance not knowing why the solutions to the equations of physics yield the felt textures they do – or in the case of a notional zombie world, not yielding any textures of experience at all. (However, I struggle with the idea of zombie self-reference, e.g. this indexical conscious thought episode.) Perhaps compare someone who claims that the essence of the physical is divine soul-stuff playing itself in accordance with the equations of QFT. Neither you nor I believe this scenario is really physically possible. I'll say my experience feels distinctly unspiritual, but this assertion won’t faze the believer: it’s just a reflection of my low level of spiritual development. Indeed, some spiritually-minded folk have seized on Hawking’s acknowledgment (“we have no idea what breathes fire in the equations”) to make just such a claim about a spiritual essence to reality without challenging the sovereignty of the physical formalism.
Either way, I confess I’m floundering. As someone who (in his early youth) extensively explored consciousness-altering drugs, and thereby hoped to expand my evidential base, I find the eliminativist perspective almost incomprehensibly alien - but I do my best to try to understand it.
[I was about to apologise for the feeling of frustration my reply is likely to cause you. Then I realised the phenomenal experience of frustration is one of the very things you were contesting existed. Yes, as I said, I’m floundering!]
* * *
The autistic component of general intelligence is often conflated with rationality or general intelligence itself. People who have extremely high AQ/IQ not merely "read" other minds differently from low-AQ neurotypicals; they understand their own minds differently too. Compare the anti-introspective effect of testosterone and the "extreme male brain" theory of high AQ. Anyhow, I'm not urging that we use more pop-psychology. Ideas should generally be examined on their own terms rather than psychoanalysing their proponents. But when it comes to radical eliminativism, I struggle to make intellectual sense of what is being said so I can properly engage. In Eliezer’s case, he's not a radical eliminativist. However, Eliezer doesn't believe that e.g. prelinguistic human toddlers, nonhuman animals, or any cognitive agent lacking the capacity for meta-cognition are conscious. They are just zombies. More radically still, Daniel Dennett (non-consciously?) believes that we are all zombies! Low-AQ folk are more likely to take feelings and emotions, pains and pleasures - their own and others' - as paradigm cases of conscious experience.
I wonder what's the best way forward - so we don't just end up talking past each other?
* * *
Brian, contra David Chalmers, you, me and Rob all think that monistic physicalism is true. There is no “element of reality” missing from the formalism of physics – or at least the equations of tomorrow’s TOE and their solutions. But we set about trying to save physicalism in radically different ways!
My point of departure is traditional. All I‘ve access to is the contents of my own conscious mind - phenomenal world-simulation and all. Even basic assumptions can be questioned. I recall once wondering if I was dreaming and then trying to read a book to test the possibility. I couldn’t read it – the blurred lines of text kept oscillating on the page. Dreaming consciousness is demonstrably different from waking consciousness. And whether you’re dreaming or awake, everything beyond the contents of one’s own conscious mind is a theoretical inference to the best explanation: scientific realism. The bold move of non-materialist physicalism is to reject the scientifically plausible idea that reality consists of physical fields of insentience (as formally described by QFT) that sometimes manifests as sentience and instead propose that sentience is the essence of the physical (as formally described by QFT). Experience is ubiquitous – a view not to be confused with animism. If so, then every feature of our minds must be isomorphic to some of the solutions to the equations of physics – even though we don’t know how to “read off” such features from the solutions to the equations. However, the only reason I have for inferring that the intrinsic nature of the physical may be experiential is the phenomenal contents of my own mind – together with the scientifically grounded assumption that I’m not ontologically special.
By contrast, if you and other eliminativists disbelieve in the very existence of conscious experience, then the possibility that the essence of the physical might be experiential is a non-starter. Eliminativists turn traditional epistemology on its head. If I understand you rightly, you (non-consciously?) believe that we all have some kind of prior conceptual handle on the nature of the physical – and then proceed to try to understand what most people conceive as consciousness as some kind of false theoretical inference which they draw. Maybe if we each (somehow) enjoyed direct perceptual access to a mind-independent world of medium-sized physical objects, then the fact that we don’t yet understand the intrinsic nature of the “fire” in the equations of physics would be important. We’d already have some crude notion of the “physical” to rely on: our everyday macroscopic world. But for multiple reasons, I don’t think perceptual direct realism is scientifically tenable.
Now to the bit you say you don’t follow me on - though I worry I may have lost you already! The behaviour of a classical digital computer is determined entirely by the laws of quantum physics – the “fire” in the equations does all the causal work - but we normally assume that this fundamental physical level can be cleanly “quarantined off”: it’s wholly incidental for our explanatory purposes. Using a silicon or gallium arsenide (or whatever) CPU to execute programs makes no difference for the purposes of our anthropocentric interests. Nonetheless all the “functional” work of a digital computer is still done entirely by brute causal processes - just as is the “functional” work of the human heart or liver. Likewise with our minds – or so macro-functionalists assume. The behaviour of the CNS is determined entirely in accordance with the laws of quantum physics too – the “fire” in the equations does all the work - but normally we assume that this fundamental level, whatever it may be, can be “quarantined off”: there’s some sort of clean digital abstraction layer that may one day allow digital “mind-uploading”. Thus it would make no difference, the macro-functionalist argues, if biological neurons were replaced by their silicon or gallium arsenide counterparts. What counts is connectivity – the connections and activation weights – of our “connectome”, not low-level stuff like the unique valence properties of carbon or liquid water.
Well, this is certainly the obvious parallel. If by “consciousness” we mostly have in mind the late evolutionary novelty of logico-linguistic thinking rather than our phenomenal world-simulations, then conceptualising one’s mind as a substrate-neutral virtual machine somehow sitting on top of massive (classically) parallel neural networks is appealing. Yet as William James, Anti Revonsuo and most recently, David Chalmers have recognised, such an assumption leads to intellectual disaster. Within your world-simulation in front of your body-image, perhaps you see e.g. a classical-looking black cat. Neuroscience can – quite independently – show discrete feature-processing neuronal edge-detectors, motion-detectors, and colour-mediating (etc) neurons are synchronously firing when you undergo the experience of seeing the cat. And yet if (1) physicalism is true, and (2) if neurons are membrane-bound classical objects, then there is no way that we know of to derive the experience of phenomenally bound black cat from such decohered neuronal “mind-dust”. No, I can’t prove that a bunch of decohered neurons / “pixels” of experience are incapable of generating a classical-looking cat - any more than I can prove that a bunch of intercommunicating skull-bound American minds can’t generate a phenomenal cat. But creating such feline apparitions would involve spooky unphysicalist “strong” emergence. Intuitively, yes, as you suggest, the world’s hardware, the "fire" in the equations, is functionally incidental to our behaviour, a mere implementation detail. As an intelligent quantum robin might say, “Not so”:
* * *
Thanks Brian. Alas eliminativists are uniquely resistant to being hypnotised.
OK, is the ontological primitive(s) we assume as the intrinsic nature of the physical a "trivial" question? If our root-metaphor of mind and reality is a classical digital computer, then perhaps yes – or at least, the issue can be bracketed off as just the metaphysical question of why anything exists at all. Formally, we can agree, the behaviour of the stuff of the world can be described field-theoretically (let's ignore speculative extensions of QFT like M-theory here). But fields of what? Insentience, experience, soul-stuff – contemporary physics says nothing. For the purposes of doing mathematical physics, the essence of the physical doesn’t matter: it’s a “philosophical” question. Likewise, if our model of mind derives from a classical Turing machine, then the intrinsic nature of the physical is functionally incidental.
However, a constraint on our theorising is empirical adequacy. Assuming that non-sentience is the essence of physical makes us literally zombies – as both radical eliminativists and non-materialist physicalists recognise, but draw very different conclusions! I’m still mystified by how your… I want to say…your experience...can be so alien from mine. For a start, there’s more to my here-and-now than a propositional attitude to semantic content that may be true or false; for example, I believe that I'm conscious. There’s the subjective, introspectable experience of my belief-episode itself and all the rest of the phenomenal contents of my here-and-now too. I’m more than open to the idea that humans typically radically misconceive our own experience – which is sometimes what I think that you must be arguing. Indeed the conjecture that experience discloses the essence of the physical is an extreme example of one such radical reinterpretation. But such experience is still…well, experience. We’re back to the Cartesian cogito.
Phenomenal binding? Yes, Dennett would say something like you suggest about the cat. By analogy again, let's adapt Eric Schwitzgebel’s, “If Materialism Is True, the United States Is Probably conscious" to a perhaps more credible thesis. If the population of the USA were co-opted into implementing the appropriate computation, then the USA would become conscious: a pan-continental subject of phenomenal experience who sees, for example, a cat. Perhaps a Dennettian radical eliminativist would say that the consequent feeling of mental unity - and this phenomenally bound, pan-continental cat-experience - is just a story the pan-continental nervous system tells itself when asked, "Do you see a unified cat?" In this thought-experiment, the pan-continental mind/nervous system answers, "Yes, I do", even though different pieces of information about the cat, the counterparts of distributed neuronal feature-processors in the CNS, are represented in different places in the USA - in distributed skull-bound American minds. Yet how, and why, is such (seeming) experiential unity created? OK, I find this whole idea incredible. As I said, we can't rule out a USA-is-conscious scenario. But it's not a physicalist conjecture: the supposed pan-continental cat-apparition is irreducible to skull-bound feature-processing minds and their reciprocal feature-signalling. Much more controversially, IMO the same irreducibility on pain of “strong” emergence follows in the CNS if we simply assume that neurons are discrete classical objects signalling to each other across synaptic clefts - rather than attempting to derive the quasi-classicality of decohered neurons in the same way Zurek et al. . attempt derive the emergence of classicality from quantum reality in the mind-independent environment. (cf. http://www.nature.com/articles/ncomms8908) Like “quantum robins”, IMO our phenomenal minds functionally tap into the very stuff of the world. Phenomenal binding is an insanely powerful adaptation of biological minds - and as far as I can tell it’s classically impossible.
The functionality of consciousness? I guess when you and I use the term "consciousness", we must have different senses in mind. In the case of a classical computer, whether we assume "materialist" or non-materialist physicalism, it makes no difference for our purposes whether a CPU is made of silicon or gallium arsenide, etc. If non-materialist physicalism is true, then ultimately physical consciousness does all the causal work; but normally we're interested entirely in, say, the gameplay of a chess match, not in whether the pieces are made of micro-qualia or whether Deep Blue is a subject of experience. By contrast, if physicalism of any flavour is true, then I’m at a loss to see how such a clean digital abstraction layer can exist for our minds - given my own phenomenally bound consciousness…
* * *
Assuming mathematical platonism, one might as well say each of us is encoded in, say, the decimal expansion of pi. Calling sentient beings and non-existent zombies by the same name risks conflating different ontological categories: whether sentient and non-sentient systems, or real and fictional entities. Of course, if the same label is employed, one can always just assume that one's nominal zombie counterpart can be systematically interpreted as doing anything you can do - and that one’s nominal zombie counterpart undergoes self-referential thoughts, seeks out painkillers for its phenomenally non-existent pain, explores psychedelic drugs to alter its non-existent consciousness, and so forth. But in my view these aren't real possibilities - especially if we aim to naturalise our folk use of modal language via Everettian QM rather than simply invoking one’s imagination.
I look forward to the day when consciousness realists and antirealists can use reversible thalamic bridges (cf. the conjoined Hogan sister twins) to “mind-meld”– although whether this will induce mutual enlightenment or psychosis I’m not sure.
* * *
The trouble is Dennett isn't consistent. There are passages where he acknowledges the reality of first-person experience. There are lots of other passages (see e.g. the "Consciousness as an evolved user-illusion" chapter in From Bacteria to Bach and Back https://www.amazon.com/Bacteria-Bach-Back-Evolution-Minds/dp/0393242072) where Dennett denies subjective experience and suggests we are zombies (e.g. p.363). Part of the problem, as far as I can tell, is that Dennett is implicitly a perceptual realist who believes each of us directly acquainted with the physical world so we can start practising what he calls "heterophenomenlogy". But only a world-simulation model is scientifically viable - cue for a Dennettian attack on the "myth of the Cartesian theatre".
[on the simulation argument]
You instantiate a world-simulation but (probably) don't live within a universal simulation
("No, we probably don’t live in a computer simulation")
A small correction: I've asked Nick several times over the years what credence he personally assigns to our living in an ancestor-simulation. He's never reported a figure higher than 20%. That's rather different from believing that if posthuman superintelligence runs ancestor-simulations, then the principle of mediocrity dictates we're probably one of them – which is Nick's view. Of course, a lot of the bite of the Simulation Argument (not to be confused with the Simulation Hypothesis) comes from our intuition that running an ancestor-simulation sounds - naively - like a cool thing for a (super-)intelligent agent to do. Yet if we re-frame the question as how likely is post-human superintelligence to recreate, say, the horrors of Auschwitz, then most people will respond that the likelihood is vanishingly small.
Like you, I'm sceptical that full-blown ancestor simulations are physically feasible. But if they are physically feasible, then one possible counter-argument arises if we assume "no collapse" QM. Assuming Everett, then presumably(??) there must be quasi-classical branches where flawed superintelligence does nonetheless run ancestor-simulations. So statistically, aren't we overwhelmingly likely to occupy one of them after all?
For what it's worth, in my view still no. We've no real evidence for the popular view that phenomenally bound subjects of experience can "emerge" at different levels of computational abstraction. However, perhaps theory of consciousness and binding open a can of worms best explored elsewhere.
Sabine, apologies for the ambiguity. I just checked the Simulation Argument FAQ (last updated 2011). Nick explicitly states that he assigns a probability to the simulation hypothesis of something in the 20% region. This is quite consistent with believing that _if_ posthuman superintelligence runs full-blown ancestor simulations, then we probably inhabit one of them. Maybe such simulations will prove technically infeasible; maybe humans will shortly go extinct; maybe any posthuman superintelligence will find ancestor-simulations too unethical or uninteresting to run. And so forth. Either way, Nick is much more cautious than some of his popularisers. Compare Elon Musk's recent claim that the probability we live in basement reality is “one in billions”.
Like the squalid world-simulations run by our minds, Darwinian life plays out in basement reality. For another view: http://www.scottaaronson.com/blog/
("Your yearly dose of is-the-universe-a-simulation")
[on gene drives]
Hence the need for multiple drives targeted at multiple genes.
("Gene Drives Thwarted by Emergence of Resistant Organisms")
Reprogramming the biosphere (gene-drives.com) is far from trivial.
Scarier than the morally superior are Nietzscheans who think they've transcended good and evil altogether.
("Do You Suffer From Illusions Of Moral Superiority")
Should sentience be precision engineered or kept as a genetic crapshoot?
("Ethics — the next frontier for artificial intelligence")
CRISPR: if Man were made in God's image, changing human nature might be unwise. As it is, genome editing is obligatory.
("Ethicists advise caution in applying CRISPR gene editing to humans")
Creating happy mammoths is more important than creating woolly mammoths.
("Woolly mammoth will be back from extinction within two years, say Harvard scientists")
Brainless blobs can learn & teach, but a nervous system often improves cognitive performance.
("This Brainless Blob Learns — and Teaches, Too")
A transhumanist commitment to the well-being of all sentience is vastly more inclusive than the well-being of all sapience.
("Intelligence: a history")
Would you be more troubled to be perceived as immoral or unintelligent?
("Real-life psychopaths actually have below-average intelligence")
Nazi race policy lowered European intelligence by devastating the ethnic group with an IQ a standard deviation above the norm.
Does the future belong to enhanced humans, transhumans, posthumans - or nonbiological artificial intelligence?
("Enhanced Humans: the Avatars of the Future")
Thanks Hunter. I'm sympathetic to Donald Hoffman's . I just worry about claims like e.g. https://www.quantamagazine.org/20160421-the-evolutionary-argument-against-reality/. To avoid confusion, IMO it's worth affirming right from the outset that one is a metaphysical realist. The mind-independent world is real and it's some 13.8 billion years old – unless, that is, one wants to argue a radical antirealism, a position that threatens to collapse into an uninteresting solipsism it’s pointless to advocate. On account of evolution, each of us runs an egocentric phenomenal world-simulation - a world-simulation that's not so much mind-dependent as constitutive of (much of) our minds. Thanks to natural selection, our waking world-simulations track gross fitness-relevant patterns in the mind-independent world. They refract these fitness-relevant patterns in all sorts of adaptive ways.
However - and this is my main reservation with Hoffman's work - in some respects our representations and real-time simulations are astonishingly accurate. The reason that e.g. a surgeon can do exceedingly intricate heart surgery on his patient - despite the surgeon’s mind consisting largely of the warped egocentric world-simulation running inside his skull - is precisely because of the macroscopic world-simulation he runs causally co-varying with relevant parts of his mind-independent local environment. If our world-simulations didn't capture these structural-relational features of the mind-independent world, then not just surgery but civilisation would be impossible.
Hot or not? Would you prefer the company of a kind, sensitive, witty, affectionate zombie or a dysfunctional human?
("Can Your Robot Love You?")
Should "natural" drug addiction be celebrated by poets or treated as a toxic affective psychosis?
("Love Is Like Cocaine. From ecstasy to withdrawal, the lover resembles an addict.")
On balance I'd prefer wireheading or opioids.
("Tinder Select: The secret invite-only version for the rich, beautiful and famous")
As an incorrigible romantic, I wonder if it can be bottled or delivered in a pill.
("48-hour sexual 'afterglow' helps to bond partners over time")
Findom: how would you cope with someone turned on by giving you large sums of money?
("What is findom? A submissive man explains the fetish")
Mating, Dating and Procreating. Have you a high polygenic score? Or are you a walking genetic timebomb?
("Opinion: Dating and mating — decided by your genetic profile?")
We all have zombie avatars in each other's world-simulations. Darwinian relationships are all tragedies in the making.
("You may have been ghosted, but have you been zombied? A look at the new dating trend driving people up the wall")
Tinder for orangutans: all sentient beings should be helped to find love.
("'Tinder for orangutans': Dutch zoo to let female choose mate on a tablet. Orangutan Samboja will be shown males on a touchscreen in experiment aimed at learning more about mating choices")
"The heart has its reasons of which reason knows nothing." (Pascal)
Likewise the gut.
("Microbes, a Love Story")
"Love is a gross exaggeration of the difference between one person and everybody else" (George Bernard Shaw) However...
("An Odd Disorder Convinced This Man That All Strangers Were His Crush in Disguise")
Tinder for orangutans: all sentient beings should be helped to find love.
("'Tinder for orangutans': Dutch zoo to let female choose mate on a tablet")
[on simultanagnosia and full-spectrum superintelligence]
What evolutionary fitness benefit does consciousness confer? What's it "for"? Sometimes one could be forgiven for thinking the only purpose of consciousness is to have interminable philosophical debates about consciousness. Each week brings news of a fresh triumphs by digital zombie software over world-class conscious humans in some new cognitive domain. Is consciousness just an incidental implementation detail of organic robots?
Perhaps ask instead: what is conscious binding “for”? As rare binding deficit syndromes illustrate, both local and global phenomenal binding plays a critical computational functional role in the effective functioning of biological robots. It’s staggeringly computationally powerful, yet no one knows how a pack of membrane-bound classical neurons pulls it off. A bumble bee is a far a more versatile and sophisticated robot than Alpha Dog. One reason we’re mostly oblivious to the computational power of phenomenal binding is that Nature has made us naïve realists about “perception”. Thus we talk simply of “seeing” our surroundings - rather than running a nearly real-time virtual reality simulation of our local environment sculpted and computed on the basis of sparse peripheral inputs.
So phenomenal binding is extremely useful, yes: but is binding literally indispensable to general intelligence? Consider simultanagnosia. Naturally we focus on the highly unusual trait of being able to see only one thing at once. But for someone with simultanagnosia, most forms of phenomenal binding (notably cross-modal phenomenal matching and a unitary phenomenal self) are still intact. The subject develops partial workarounds in order to cope. We don’t live on the unforgiving African savannah. In AI too, software engineers / robot designers can develop workarounds - just as chess programmers can develop software that can now beat the world's best players, despite the ability of a strong human chess player to “see” the whole board at a glance.
However - and this is the real point I want to make - what is the computational-functional workaround for not having a unitary phenomenal self: a workaround for not being a unitary phenomenal agent who acts within an almost real time, cross-modally matched egocentric world-simulation of the kind you’re running right now? When we talk about building "artificial general intelligence”, what will substitute for an absent AGI phenomenal self? Particularly if, anthropomorphically, we give the system in question a cutesy human name (“Watson” etc) it’s easy to underestimate the magnitude of the challenge. Humans with autism spectrum disorder or schizophrenia have a different, dysfunctional self, but at least partial global phenomenal binding is retained. Yet what are the upper cognitive bounds of totally “disintegrated” information processing systems with no local or global phenomenal binding at all - just incredible speed and serial depth of processing? IMO, full-spectrum superintelligence will be our hybrid biological descendants - genetically and AI-enhanced organic minds who incorporate non-conscious / unbound “narrow” digital AI modules/neurochips. Contrast a MIRI-type FOOM scenario (the “intelligence explosion”), or Kurzweilian “mind uploading”. In my view, the kinds of experience that superintelligent posthuman minds will bind are likely to be alien to any human conceptual scheme with one exception: future minds will be inconceivably blissful. For without hedonic tone, what’s the point?
Three kind of idealism:
1) Solipsism: all that exists is my mind.
2) Pluralistic idealism: all that exists is a bunch of minds.
3) Physicalistic idealism (aka non-materialist physicalism): all that exists is a physical field of experience exhaustively described by the universal wavefunction. There is no Hard Problem of consciousness because experience is the essence of the physical.
Only (3) is consistent with science. But is it true? And critically, can we test it?
Yes, IMO. All we need to falsify the conjecture is find some aspect of experience - some "element of reality" - that is not captured by the formalism of physics but present in our minds. If we can do so, then physicalistic idealism is refuted.
Sometimes it’s supposed the palette problem (cf. http://onlinelibrary.wiley.com/doi/10.1002/tht3.113/abstract) is the missing element of reality that falsifies physicalism – physicalism of any kind. The palette problem is one of two reasons that David Chalmers feels driven to naturalistic dualism. (cf. http://consc.net/papers/combination.pdf) We undergo countless different textures of experience. By contrast, the ontology of fundamental physics is sparse and simple - albeit not yet quite as simple as we would wish (i.e. the Standard Model doesn’t include gravity and contains thirty or so adjustable parameters that need to be “put in by hand”). However, the particle-based ontology of elementary quantum mechanics is just a low-energy non-relativistic approximation of quantum field theory. In QFT, fields are fundamental, not particles. If physicalistic idealism is true, then the zillions (conventionally, an infinite number) of solutions to the equations of QFT yield the different values of experience. Why the "quantum" in QFT? Because without the superposition principle to underpin phenomenal binding, we'd be micro-experiential zombies.
And here (at last) comes the testability.
A lot of one's resistance to physicalistic idealism (and more generally to panpsychism in its various guises) comes from an extremely powerful intuition - an intuition I share - namely that the smallest possible unit of experience (“the psychon”) is much larger than the dimensions of the fundamental ontology of physics, whether fermionic and bosonic fields or – heaven help us – Planck-scale superstrings/branes. I guess the most popular candidate within the scientific community for the minimal "psychon" would be a neuron or a local neural network. But actually the violence to intuitions of non-materialist physicalism is worse. If experience discloses the intrinsic nature of the physical, then the fundamental “psychon” of experience must be ludicrously short-lived - not just ludicrously small.
And here is the crux. Assume that molecular matter-wave interferometry has advanced far enough so that we can probe the CNS (or in vitro neuronal networks - “micro-brains”) at femtosecond, attosecond and even zeptosecond timescales. What will experimentalists find? The naive answer is just 86 billion discrete, decohered, membrane-bound classical neurons – just the way we conceptualise the CNS in our temporally coarse-grained investigations today. Another answer is that we’ll find quantum weirdness alright, i.e. fleeting molecular and neuronal macro-superpositions are real, but their signature will just be “noise”. By analogy, imagine someone who believes that digital video game characters (or tomorrow’s putative “mind uploads”) are phenomenally bound subjects of experience. Suppose he naively probes the CPU of his futuristic desktop computer at attosecond timescales while a game (or a “mindfile”) is being executed in search of a non-classical interference signature of functionally relevant coherent superpositions of machine code. Good luck!
However, there's a third answer, which is tentatively mine. The answer is not idiosyncratic because I’m proposing new physics - I’m boringly conservative: you need to be very clever indeed to start monkeying around with the symmetries expressed in the bare quantum formalism. No, short-lived macro-superpositions of distributed neuronal feature-processors must exist in the physical world on pain of a failure of the linear unitary Schrödinger dynamics. What is idiosyncratic in a "Schrödinger’s neurons" proposal is that interferometry will reveal a perfect structural match between our phenomenal minds – in particular, phenomenally bound perceptual objects in our world-simulations - and the formalism of QFT.
Note that what we're looking for in the non-classical interference signature isn’t a perfect structural match in classical four-dimensional space-time, but rather a perfect structural match between coherent superpositions of neuronal feature-processors and a real physical ﬁeld in the conﬁguration space of the wave function. Of course, these sorts of timeframe are intuitively absurd. “Everyone knows” that conscious mind arises over scores of milliseconds. But these timeframes are neither more nor less absurd than micro-experience at femto-, atto-, zepto-metre (etc) distance-scales on the assumption that non-materialist physicalism or property-dualist panpsychism is true. If QM is formally complete, then what David Chalmers calls a “problem” for non-materialist physicalism/panpsychism is actually the solution to the phenomenal binding / combination problem. The binding / combination problem is an artefact of classical physics and crude classical neuroscience.
There's an obvious rejoinder here. No mechanism exists by which such fleeting individual neuronal superpositions could be harnessed, either by the individual organism or evolution. By contrast, connectionist neuroscience captures both the relevant dynamical timescale (cf. Max Tegmark) and molecular mechanism (i.e. neuronal action potentials and synapses) for computationally relevant state transitions, unlike quantum mind.
I’d beg to differ. The most sophisticated version of the decoherence program in post-Everett QM, Zurek’s “Quantum Darwinism" (cf. John Campbell’s “Quantum Darwinism as a Darwinian process”: https://arxiv.org/ftp/arxiv/papers/1001/1001.0745.pdf), offers a powerful and unremitting selection mechanism: inconceivably intense. It’s a selection mechanism that generates quasi-classical Everett branches in the guise of a vast preponderance of well-behaved macroscopic “worlds” over maverick worlds, and also – when applied to the CNS – explains our capacity to run phenomenally bound classical world-simulations. I know of no other way to “save the phenomena” and derive the familiar lifeworld of one’s everyday experience from fundamental physics.
* * *
Alternatively, assume physicalism is true and experience is the "fire" in the equations of QFT.
("On why Idealism is superior to Physicalism and Micropsychism | Bernardo Kastrup")
* * *
As you might guess, I'm less optimistic than Andrés that the (hypothetical!) inflaton field is a manifestation of hedonium. Mercifully, I doubt it’s dolorium either, despite my bleak view of life. If physicalistic idealism is true, then its subjective texture is… goodness-knows-what. [As an aside, I've just read Roger Penrose, "Fashion, Faith, and Fantasy in the New Physics of the Universe” - surprisingly good: Penrose always flags when he's saying something that departs from string-theoretic orthodoxy.]
If physicalistic idealism is true, and if local and global phenomenal binding is underpinned by neuronal superpositions, then neuroscience needs a Mendel to Zurek's Darwin, so to speak – though sadly quantum biology needs more mathematical sophistication than selectively cross-breeding pea plants. Quantum Darwinism applied to the CNS yields selection pressure for functional binding of stupendous, inconceivable power. By analogy, imagine four billion years of Darwinian natural selection (as classically conceived) compressed into a microsecond - unceasingly, unremittingly.
If this model of our minds is correct, then differential robustness to decoherence across the CNS needs to be described and quantified at the molecular level. Here alas I reach the limits of my technical competence and start philosophical handwaving. Above all, I'd love some experimental guidance on whether we're on the right track. What I call the "(cf. protocol" for experimental (dis-)confirmation is a philosopher’s conception of a protocol, not an experimentalist’s. Otherwise it's just a thought-experiment.
Or is biohappiness best? To be cost-effective, pay for perpetual activation your twin "hedonic hotspots". Or better, donate to help create a CRISPR civilisation based entirely on gradients of intelligent bliss.
("How to Buy Happiness. Social scientists offer some answers.")
We need a world happiness olympics, with all forms of cheating encouraged.
(Welcome to the happiest country on Earth") Happiness: perhaps compare an international league table that excludes non-Aryans. Anthropocentric bias is rife in science.
("Who's happy, who's not: Norway tops list, US falls")
[on Existential Risk]
Blurb for Phil Torres' "Morality, Foresight, and Human Flourishing: An Introduction to Existential Risks, Pitchstone Publishing." (2017)
If life on Earth didn't exist, would a benevolent superintelligence create it? How might the beauty and joy brought into existence be impartially weighed against the creation of such immense cruelty and inexpressible suffering? Whatever our answer to this hypothetical, few respondents would claim that giving an answer of "No" is self-evidently monstrous or absurd. By contrast, any hypothetical proposal to wipe out existing life strikes most people as self-evidently monstrous and absurd. A big part of such an incongruent reaction undoubtedly lies in status quo bias. But beyond mere status quo bias, a better reason exists to be appalled at apocalyptic solutions to the problem of suffering. The world has no clean OFF switch. An attitude of mind that places no positive value on human and nonhuman life, and disregards the expressed wishes of autonomous individuals, typically leads to more suffering, not less. Thus there is nothing inconsistent with being a negative utilitarian or a Benatarian or a Buddhist and also believing that the sanctity of human and nonhuman life should be enshrined in law. This is my view.
Yet what if the world really did have an OFF button? Or rather, what if developments in artificial intelligence, genetic engineering or WMD turn the extinction of sentient life into a credible prospect, not just a thought-experiment? Phil Torres’ “Morality, Foresight, and Human Flourishing: An Introduction to Existential Risks” offers an authoritative guide to the emerging scientific discipline of Existential Risk in all its guises.
As a Misguided Ethicist who would, unhesitatingly, press a clean hypothetical OFF button, I confess I was surprised to be invited to contribute a blurb to this volume. Knowing my dark views on life, a handful of young, idealistic, and extremely smart people have inquired over the years about the feasibility of apocalyptic solutions to the problem of suffering. My answer has always been the same. Apocalyptic solutions are an ethically hazardous distraction. Don’t allow a supremely compassionate Buddhist / negative utilitarian doctrine, i.e. our overriding ethical obligation to mitigate and prevent suffering, get mixed up with plotting Armageddon. The only rational way to get rid of suffering will use the non-violent democratic politics of persuasion - promoting the slow, fitful process of rewriting the genome to create a civilisation based on gradients of intelligent bliss. A few centuries from now, our superhappy and superintelligent descendants may view negative utilitarianism as a pathology of mood and thought – just like most other Darwinian value systems.
However – and this is the real worry - we should also try to understand why a Seriously Misguided Ethicist might reach a radically different conclusion – and far more dangerously, influence impressionable minds more prone to action than armchair philosophising. Unlike the Misguided Ethicist, the Seriously Misguided Ethicist claims that utopian talk of abolishing suffering by creating a blissful post-CRISPR biosphere (etc) is fanciful. Why not just bring the whole Darwinian horror-show to an end? Buddhists speak of nirvana (“snuffing out”). More effective tools will soon exist to extinguish pain-ridden life on Earth than the Noble Eightfold Path, for example igniting a nirvana shockwave. Granted, intelligent life may conceivably be poised to radiate across the Galaxy spreading love and happiness. But maybe C.S. Lewis is right: “Let’s pray that the human race never escapes from Earth to spread its iniquity elsewhere.” Why take the risk?
I don’t have a knock-down reply to the Seriously Misguided Ethicist, other than to say that the problem of suffering and the study of existential / global catastrophic risks are not orthogonal, as is often supposed, but intimately related. Other things being equal, the more we value life, the more motivated we are to preserve it. Anyone seriously interested in global catastrophic and existential risk should explore tackling one of its primary underlying causes: the biology of suffering. For sure, the world has few card-carrying negative utilitarians. Yet negative utilitarianism just systematizes the insight of millions of thoughtful folk who would “walk away from Omelas” (cf. https://en.wikipedia.org/wiki/The_Ones_Who_Walk_Away_from_Omelas) – people who would forgo the promise of fabulous delights if the price of pleasure were the misery of a single child. What “walking away” entails is controversial – and perhaps profoundly disturbing.
Phil Torres’ book covers the risks posed by Misguided Ethicists and other forms of existential risk in a scholarly, rigorous fashion. Highly recommended.
* * *
[Adam Karlovsky quotes me] "Singularitarians worry about sentience-unfriendly artificial intelligence. But human society is based on sentience-unfriendly biological intelligence. We hurt, harm and kill billions of sentient beings each year in the death factories."
Thanks Adam. There's a terrible irony here. I.J. Good's original proposal, i.e. that a combination of Moore’s law and software-based, recursively self-improving AI might lead to a runaway Intelligence Explosion, has been argued most vigorously over the years by Eliezer Yudkowsky and SIAI / MIRI. Nick Bostrom then wrote the book Eliezer should have written ("Superintelligence"). The spectre of sentience-unfriendly superintelligence was picked up by a charismatic billionaire (Elon Musk) and an academic superstar (Stephen Hawking). Media interest then ensured a sensationalised version entered mainstream public awareness. However, Eliezer himself believes that nonhuman animals, and human babies / prelinguistic toddlers prior to acquiring meta-cognitive capacities, aren't conscious. Ordinary meat-eaters (and yesterday’s cannibals?) typically downplay the minds of their victims. Eliezer believes they aren’t sentient at all.
(cf. "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." EY)
Eric, in one's own life, peak experiences or orgasmic bliss - and (hopefully) rare uncontrollable panic or agony - are more intensely conscious than the experience of exercising complex meta-cognitive capacities. They matter more. IMO, we should be sceptical of the claim that reflective states are intrinsically (rather than perhaps instrumentally) more valuable. But what I wanted to highlight was how EY believes that human babies and nonhuman animals aren't subjects of experience at all: they can't suffer. False theories of (un)consciousness can be ethically catastrophic. Compare the Cartesian vivisectionists: "distress vocalisations" of the agonised dogs they vivisected were supposedly just noises emitted by feelingless automata.
Of course, the risk of error cuts both ways. Should disbelievers in digital sentience alter our behaviour on the possibility we could be catastrophically mistaken? As you know, my own conception of the nonclassical basis of phenomenal binding is speculative - at best.
[on wireheading and DBS]
Later this century, psychiatric wonderdrugs and CRISPR genome-editing promise mental health for everyone. Until then, should victims of untreatable depression be allowed unlimited access to opioids, DBS or wireheading?
("DARPA’s Brain Chip Implants Could Be the Next Big Mental Health Breakthrough—Or a Total Disaster")
Let's hope so....
("Why upgrading your brain could make you less human")
When will the reckless genetic experimentation of sexual reproduction be replaced by transhuman designer babies?
("Artificial human life could soon be grown in lab after embryo breakthrough")
"No act of kindness, no matter how small, is ever wasted." (Aesop, The Lion and the Mouse)
* * *
Tim, first, survivalism. Does the existence of suffering in the world tend to increase or decrease existential risk? If this group advocated e.g. mass wireheading, or engineering world-wide congenital analgesia, or even everyone getting off their heads on hedonistic cocktails of drugs, then perhaps you should be concerned. But crudely, the more one loves life, the keener one is to preserve it. Compare the hyperthymic temperament of members of institutes like FHI devoted to the study and mitigation of existential risk. If the main focus of this group were existential risk, then I'd still urge phasing out the biology of suffering - not so much because of hypothetical button-pressing negative utilitarians, but because in an era of WMD some depressive or other troubled soul might decide to take the rest of the world down with them. Creating a world animated by gradients of intelligent bliss is pro-life and pro-survival.
Second, consciousness. If phenomenal binding is a classical phenomenon, as you believe, then digital sentience and “mind-uploading” should be feasible - with profound ethical implications. In order to phase out suffering, we’ll need to get our theory of mind right. You'll find a predictable diversity of opinions here. Yes, I incline to the view that classical digital computers will always be zombies. Maybe I’m wrong: interferometry experiments on the lines outlined will confound my favoured theory. But this is good! We should urge any researcher with a conjecture - classical or quantum - of consciousness or phenomenal binding to spell out as clearly and precisely as possible its testable consequences - if any – versus its competitors.
[on How We Should Act]
Andrés Gómez Emilsson on G+ quotes:
"If meta-ethical realism is correct, i.e. there is an objectively correct ethic to be discovered, then presumably its policy implications may be as much at odds with folk morality as is quantum theory at odds with folk physics." (David Pearce)
Thanks Andrés. In one sense, I'm still in thrall to commonsense. My agony and despair is self-intimatingly bad. My sublime bliss is self-intimatingly good. The insight that – for reasons we don’t understand – the pain-pleasure axis discloses the world’s intrinsic axis of (dis)value is philosophically ancient. So we may ask how a benevolent full-spectrum superintelligence should act, i.e. a God-like agent who can impartially access and appraise all first-person and third-person perspectives and act accordingly. Presumably, God-like superintelligence won’t be prey to the egocentric delusion that hardwires Darwinian minds evolved under pressure of natural selection to feel uniquely special: the egocentric illusion.
The upshot of these reasonable-sounding constraints may be utterly alien and perhaps repugnant to most of us. For instance, I’d struggle to defend - except instrumentally - creating a world based on gradients of intelligent bliss rather than laying the foundations for a utilitronium shockwave: cosmic wireheading, as it were.
However, in order to avoid getting trapped in a philosophical quagmire, sometimes it’s easier to set aside meta-ethical realism and focus on rational agency. As you know, I still hope someone will write an amoral Decision-Theoretic FAQ for a full-spectrum superintelligence: the counterpart of the amoral Less Wrong Decision-Theoretic FAQ minus its irrational individualist metaphysics:
Presumably, full-spectrum superintelligence won't be bewitched by a false metaphysics of personal identity.
Qualia computing, so to speak.
Currently, a daunting obstacle to writing a purportedly God’s-eye-view ethical and/or decision-theoretic FAQ is that it’s impossible to act rationally or morally – except by accident - without grasping the nature of reality, at least in crude outline. No sign of the fog lifting any time soon. Cosmology is in flux. Post-empiricist science is in vogue. On a personal note, my motivation to write and act has been sapped by “multiverse mania”. Despite finding Everett’s original PhD thesis compelling, I used to console myself with the thought that Everettian QM might be wrong. Perhaps gravitationally-induced wavefunction collapse à la Penrose undercuts the universal applicability of the superposition principle. Or perhaps (somehow) the "bare formalism" of the unitary dynamics has a limitation no one has fathomed. Everett aside, I hope Peter Woit is right. (cf. http://www.math.columbia.edu/~woit/wordpress/?p=9027) Succumbing to what Nick Bostrom christened “infinitarian paralysis” is a real risk - whether or not one believes in the literal existence of physically realised infinities. Yet there is a real tension between (in my case) defending some sort of zero ontology (cf. https://www.quora.com/Why-does-the-universe-exist-Why-is-there-something-rather-than-nothing) and a desire to “do the right thing” that I’m unable to reconcile. Sometimes my mind gives way under the strain.
* * *
Tim, we agree. The experimental detection of quantum coherence in microtubules, or alternatively the non-classical interference signature of fleeting neuronal superpositions, will not prove the mind-brain is a quantum computer. And we’re not robins: http://scitechstory.com/2011/01/24/the-robin-flies-with-quantum-coherence/] But Orch-OR - and likewise the no-collapse explanation of phenomenal binding I explore - are experimentally falsifiable conjectures: they make novel, precise, empirically testable predictions. These predictions are - to anyone who understands the nature of decoherence - extremely implausible. But faced with a problem we can't solve, we need implausible predictions - conjectures that are "risky” in Popper’s sense. Thus what will vindicate or falsify Orch-OR isn’t the demonstration of quantum coherence in microtubules, tantalising as it may be; rather, it will be any experimentally detected deviation from the unitary Schrödinger dynamics on the scale which Penrose reckons OR occurs. Likewise, what will vindicate or falsify the explanation of phenomenal binding I favour won’t be detection of the non-classical interference signature of neuronal superpositions per se, but rather the presence or absence of a perfect structural match, i.e. synchrony is really superposition – insane as it sounds. (https://www.physicalism.com/#6)
At Tucson, I thought David Chalmers had the same you-can’t-be-serious response as you when he realised what I was proposing. But I see from his Reddit AMA that he thinks the conjecture is worth experimentally falsifying. Cool. I'm just curious.
[on mental health]
Yet we still go on churning out cruel genetic experiments ("babies"):
("If you live your life without mental health issues, science says you’re the weird one")
Alas suing for another windfall may not solve the problem. Roll on a WHO initiative to recalibrate the hedonic treadmill.
("Britain's youngest EuroMillions winner says life ruined")
Sleep may be as much of a cognitive enhancer as psychostimulants masquerading as smart drugs.
("The Brain’s Connections Shrink During Sleep")
Late-night carbs won't especially cause you to put on weight and may help you get a good night's sleep.
("Scientists Dispel Late-Night Eating/Weight Gain Myth")
“There is a time for many words, and there is also a time for sleep.” (Homer) I combine late-night bananas and a melatonin supplement...
"All men while awake are in a common world: but each when asleep is in a world of his own" (Plutarch) A useful myth.
("Study points to a universal immune mechanism as a regulator of sleep")
Chronotypes" may be complicated by your drug regimen:
("It's Okay That You Don't Like Mornings--Your Memory Probably Doesn't Either. How sleep, circadian rhythms and chronotype affect your ability to remember")
Today's sleeping and waking consciousness deserve to pass into evolutionary history. But what comes next?
("Researchers Tap a Sleep Switch in the Brain")
A pan-species welfare state will need autistic hyper-systematisers as much as tender-minded bunny lovers.
("Feral Bunnies Are Taking Over Las Vegas")
Utilitarian ethics may be computable; but without the spark of empathy, who would build the supercomputer to do so?
("Empathy is crucial to being a good person, right? Think again")
[on diet and nutrition]
Corn consumption positively correlates with human homicide rates - though is low tryptophan/serotonin the villain of the piece?
("A diet of corn is transforming French hamsters into raging cannibals")
The battle against postprandial somnolence ("food coma"): can you eat and stay sharp?
("Are 'food comas' real or a figment of your digestion")
Do you enjoy luminous clarity of mind or an introspective void?
("How to become a ‘clean thinker’ and get rid of middle-aged brain fog") Quitting meat and eating whole-grain wheat, rice, oats and barley products is a recipe for longer human and nonhuman life. https://medicalxpress.com/news/2017-02-grains-metabolism-calorie-loss.html
Alternatively, liking dark and twisted jokes could be sign of empathy deficits and impaired social cognition.
("Liking dark and twisted jokes could be sign of intelligence")
[on free will]
Bartek, yes, both t'Hooft-style superdeterminism ("the ultimate conspiracy theory") and Everettian QM are inconsistent with free will. The alternative is to accept spooky non-locality (cf. http://www.nature.com/news/cosmic-test-backs-quantum-spookiness-1.21401) The only way I can think of to salvage free will would involve allowing the actions of all conscious agents collectively to determine the "constants" of Nature - with a full-spectrum superintelligence contributing to their value more than a mouse. For lots of reasons, I don't think this will work.
Consciousness? Well, as a (tentative) non-materialist physicalist, I take seriously the possibility that consciousness is conserved, neither created nor destroyed. (cf. http://www.physicalism.com) Yet unless such fields of consciousness are phenomenally bound, then we'd be zombies in all but name. Death? Well, I think we all timelessly live in a higher-dimensional Hilbert space from which you'll never be deleted. But BN or DCP probably won't walk the Earth in Year 3000...
The Happy Song.
(well, it works for me...)
http://www.livescience.com/57784-can-this-song-make-babies-happy.html Means-ends rationality suggests giving our brains what they're after rather than conserving evolution's Rube Goldberg machines. But natural selection means the world is full of people who imagine they've transcended the pleasure principle.
("Sex, drugs and music: Brain switch stops people feeling emotional about songs When played")
International happiness studies lack methodological rigour. Everywhere the hedonic treadmill still grinds.
("Some of the world’s most unhappy countries are also the most optimistic")
[on the death of a loved one]
In my view, no one gets deleted from space-time: we (tenselessly) occupy the co-ordinates in space-time/Hilbert space we do.
This possibility can be consoling or demoralising depending on whether one is a classical or negative utilitarian.
I'm here. In what sense is stuff going on in Australia real - after all, it's there.
However, if "time is what stops everything happening all at once", space is what stops everything happening all in the same place.
All are equally real.
My finitism? Perhaps space-time is finite but unbounded.
Ultimately perhaps we all (tenselessly) live in a higher-dimensional Hilbert space from which you'll never be deleted.
(cf. "Nothing happens in the Universe of the Everett Interpretation" by Jan-Markus Schwindt https://arxiv.org/pdf/1210.8447.pdf)
Hilbert space is typically treated as infinite-dimensional,
But see: http://www.scottaaronson.com/blog/?p=2820
Hilbert space is also typically treated as a mere mathematical fiction. However, I'm a realist when it comes to QM. Recall I conjecture that when we probe the mind-brain at the right temporal resolution with molecular matter-wave interferometry, we'll detect a perfect structural match between physics and our phenomenology of mind. This is my answer to David Chalmers "structural mismatch" argument for dualism. The perfect structural match isn't in three-dimensional space or four-dimensional space-time but rather in the (realistically interpreted!) higher-dimensional wavefunction space of QM.
[on ethyl alcohol]
Ibudilast: an anti-inflammatory mood-brightener that also reduces problem drinking?
("Drug shows promise for treating alcoholism. Study finds an anti-inflammatory medication appears to reduce cravings, improve mood")
[on mathematical beauty]
"Some of these transformations are more pleasant than the others." (Andrés Gomez Emilsson)
If I were a strong mathematician, I'd probably agree with you Andrés! Spared of this confounding bias, I'd argue the opposite. With microelectrodes, reinforcement learning, or indeed a bit of neurological tweaking, a neuroscientist could invert your preference ranking - or alternatively, make each transformation seem equally (dis)agreeable by its very nature.
A mathematician might protest, "But only I can apprehend true beauty!" But compare the assertion that e.g. the features of 20 year old women of prime reproductive potential are intrinsically sexy, whereas the perceptions of other folk are somehow less "natural" (cf. https://en.wikipedia.org/wiki/List_of_paraphilias) Claims for the primacy of one's own fetishised intentional objects need to be treated with scepticism. Yes, we can give an evolutionary story of why some intentional objects are more readily fetishised than others – sometimes fetishised so powerfully that their agreeable nature seems built into the very fabric of reality. Folk who disagree with our perceptions are simply deluded. Yet (IMO) the only thing that is intrinsically pleasurable is pleasure itself. When we (1) identify the molecular signature of pure bliss in our twin hedonic spots, and (2) understand the neurological basis of phenomenal binding, we should be able to "paint” indescribable bliss on literally any of our neocortical representations (or their transformations) as desired: paradise-engineering.
[on naturally evolved quantum supercomputers]
Raising the stakes for a successful theory of consciousness.
Stuart, my working assumption is that your mind-brain is a quantum computer running a phenomenally bound classical world-simulation in almost real time. You are what a naturally evolved quantum computer "feels like from the inside". Unlike e.g. the Penrose-Hameroff Orch-OR theory, the conjecture doesn’t rest on any new physics. Rather, I assume quantum mechanics is formally complete. Researchers working on the foundations of QM have long been stymied by the Problem of Definite Outcomes - and more recently, whether or not the decoherence program in post-Everett QM really solves the mystery (cf. http://philsci-archive.pitt.edu/10757/). If we assume nothing but the unitary dynamics, what explains the apparent collapse of the wavefunction on measurement in accordance with the Born rule? You observe a definite classical cat - either dead or alive, not superposed dead-and-alive. If the superposition principle of QM has universal validity, in micro-world and macro-world alike, then why are superpositions never experienced or observed, only inferred?
My (very) tentative answer is that only the universal validity of the superposition principle allows your mind/brain to experience determinate classical-looking objects - such as your cat or a determinate pointer-reading - in a phenomenally bound world-simulation amenable to description by an approximation of the laws of classical physics. Only the existence of coherent superpositions of distributed neuronal feature-processors allows you to instantiate a macroscopic phenomenal world of medium-sized dry objects. Non-materialist physicalism - and the “bare formalism” of wavefunction monism - is a recipe for what you’re experiencing right now.
OK, intuitively this is crazy - for all sorts of reasons.
(1) The theoretical lifetime of neuronal superpositions in the CNS is femtoseconds or less - so short-lived they've not yet been detected, although only a minority of physicists anticipate that molecular matter-wave interferometry will detect any derivation from the unitary Schrödinger dynamics. (cf. https://phys.org/news/2015-10-physicists-quantum-spookiness-schrodinger-cat.html)
(2) Even if we grant - as proposed by non-materialist physicalism - that such fleeting superpositions are experiential, they could at most just be functionless psychotic noise. Surely no mechanism exists to co-opt hypothetical neuronal superpositions into playing any functional role in the CNS? Compare the learning algorithms of mainstream connectionist neuroscience.
On (1) yes, indeed: I'm not proposing a novel theory of physics, or any modification or supplementation of the unitary dynamics. The CNS decoherence timescales I use are drawn mainly from Max Tegmark and Maximilian Schlosshauer (cf. https://www.amazon.com/Decoherence-Classical-Transition-Frontiers-Collection/dp/3540357734) – no friends of quantum mind.
(2) On the contrary, just a selection mechanism exists to turn you from psychotic noise into a quantum supercomputer! Zurek’s “quantum Darwinism” (cf. https://en.wikipedia.org/wiki/Quantum_Darwinism) is no shallow, tricksy metaphor, as one might first assume.
Quantum Darwinism is standardly used by physicists working on the decoherence program to explain the emergence of quasi-classicality in the mind-independent world (cf. http://philsci-archive.pitt.edu/10940/1/Wallace_review_final_single-spaced.pdf) Yet what happens when we apply the decoherence program to the inside of your skull?
Naively, what plays out in your skull is the emergence of 86 billion odd membrane-bound classical neurons. Hence the phenomenal binding/combination problem - as classically posed - and the impossible structural mismatch between our minds and the microstructure of the CNS that ultimately impels David Chalmers to dualism. (cf. http://consc.net/papers/combination.pdf)
Less naively? Well, think of four billion years of Darwinian selection pressure as conventionally conceived. Now imagine selection pressure of far greater intensity exerted on (non-)psychotic coherent neuronal superpositions every microsecond of your life. (cf. Robin Hanson’s "mangled worlds”, http://mason.gmu.edu/~rhanson/mangledworlds.html, though substitute “mangled world-simulations” in this context.) In my view, it's this insanely powerful mechanism that allows a naturally evolved quantum computer like you to run a classical world-simulation at quadrillions of individual quantum-coherent “frames” per second.
Anyhow, as I said, crazy stuff. But the conjecture answers 1 to 4 in Andres' "Raising the Stakes" post. Critically, it’s experimentally falsifiable, i.e. it offers novel and precise empirical predictions, not just dressed-up retrodictions. Either the non-classical interference signature disclosed by molecular matter-wave interferometry will independently reveal a perfect structural march between our phenomenally bound minds and the formalism of physics, or it won't: https://www.physicalism.com/#6. Note the conjectured perfect structural match we’ll be probing for isn't in four-dimensional space-time but rather in the higher-dimensional wavefunction space of QM.
When I outlined the proposal last year at Tucson, I got the impression David Chalmers reckoned it was bonkers - certainly a gut-feeling I share. But - hats off to Chalmers - he's apparently given it some thought and agrees the conjecture needs experimentally falsifying. (cf. https://www.reddit.com/r/philosophy/comments/5vji57/im_david_chalmers_philosopher_interested_in/de2sb4b/?st=izk4g6p6&sh=7b9f72cd) Stepping back, I don't seriously believe in it myself; I just think the alternatives, e.g. dualism or radical eliminativism (e.g. https://nothingismere.com/2017/02/23/phenomenal-consciousness-is-a-quasiperceptual-illusion-objections-and-replies/) - are more incredible still, which is saying something.
[on effective altruism]
Effective altruism (EA): what are our core values?
("Fundamental Values and the Relevance of Uncertainty")
Infectious altruism? Perhaps microbiologists could engineer epidemics of morality, despite cases of innate immunity
("Microbes may encourage altruistic behavior")
Superhuman selflessness is admirable. Alas effective altruism probably depends on developing biotechnology and IT to make the price of altruism negligible. In a reprogrammed biosphere based on gradients of bliss, everyone wins...
("Selfless Biology of the Extreme Altruist. What is the opposite of a psychopath? What is the limit of human goodness")
The massively amplifying tools of biotechnology can make the well-being of all sentience as easy as pressing "love".
("On Facebook, love reactions triumph over hate. The heart reaction dominated your social feed in 2016")
Mom was right: chewing food thoroughly promotes health:
("Research reveals surprising health benefits of chewing your food")
Vitamin D supplementation: alternatively, go for a stroll in the sun.
("Vitamin D 'proved to cut risk of colds and flu'") Medicine and health are too important to be left to doctors, and software-based diagnostics can soon be free.
("Forget your GP, robots will 'soon be able to diagnose more accurately than almost any doctor'")
[on autism spectrum disorder, masculinity and AQ]
Alternatively, how high is your AQ?
("How to Email. An etiquette update: Brevity is the highest virtue")
Could autism be a cognitive blessing in a world seemingly designed by a SuperAsperger?
("Women with a thicker brain cortex are more likely to have autism")
"It is amazing how complete is the delusion that beauty is goodness.” (Tolstoy)
Will truth always be ugly?
("Meitu: 'Beauty' selfie app surges in popularity as people perform drastic makeovers on their photos")
Is your body more of a temple or a pharmacy warehouse?
("How Two Florida Gym Rats Conquered the Shadowy World of Dietary Supplements")
Should we choose our optimal level of testosterone function - or accept what Mother Nature has given us?
Higher testosterone linked to reduced closeness in social relationships")
Complicating matters: is "postural expansiveness" the key to attraction?
("A Surprising Secret of Attraction. If you want people to swipe right, be "expansive" in your profile picture.")
Too many white male vegans...?
"Uh oh...Your Vegan Panel is All White or Male"
Perhaps comparative cognitive style is relevant. Utilitarians and effective altruists are more likely to be white, male hyper-systematisers in virtue of our higher AQ scores. (cf. https://www.theguardian.com/education/2003/apr/17/research.highereducation) Compare too the otherwise anomalous disproportionate number of female vegetarians versus the approximately even ratio of male-to-female vegans. Also, Ashkenazi Jews tend to have abnormally AQ scores, as do (to a lesser extent) Caucasian Europeans (cf. http://hsnewsbeat.uw.edu/story/human-neanderthal-gene-variance-involved-autism) Black people (who typically lack Neanderthal DNA) tend to record lower AQ scores.
("The Myth of the Black Aspergian")
("Israel Goes Vegan")
And so forth.
Differences in cognitive style may turn out to be play only a small role in explaining speaker disparity: I can think of other, much more plausible reasons. But I guess it’s the sort of cofounding variable one needs to rule out before drawing conclusions...
High-intensity interval training (HIIT): can we create its superior functional analog in pill form?
("Can high-intensity interval training delay the ageing process? Researchers have found that short-burst exercise might have rejuvenating effects – but there can be such a thing as too much intensity")
Can whole-body vibration replace exercise?
("Whole-body vibration may be as effective as regular exercise") Or become a virtual super-marathoner with a pill?
("Scientists take aim at diabetes and obesity with exercise in a pill")
[on lust and eroticism]
Kisspetin: Will "mental Viagra" soon rekindle your lust for life?
("'Mental Viagra' on horizon as scientists boost lust signals in brain")
"God is an Englishman." (E.M. Delafield) But learning lesser tongues too may be good for the brain.
("Bilingualism may save brain resources as you age")
"But if thought corrupts language, language can also corrupt thought."(George Orwell)
Oh to speak posthuman.
Superhuman bliss: sometimes I think Andrés' brain is halfway there already...
I once had a nightmarish dream in which FHI's Anders Sandberg - "I do have a ridiculously high hedonic set-point" - confided that he was secretly extremely unhappy. Was I overcome with compassion? Sadly no, I recall being annoyed that I'd now need to edit out my standard example of lifelong high-functioning well-being and replace Anders with someone else. I'm glad to say I later woke up.]
Machiavellian intelligence? Can AI outwit cunning but primitive humans in poker and life?
("After day one, AI is crushing humanity at poker")
"A pack of cards is the devil's prayer-book" (German proverb)
Will AI beat Satan at his own game?
Presumably our successors will believe 540 million years of suffering were a price worth paying for superhuman bliss. Are they right?
("Unhappy or Depressed? In the Future, Everything Will Be All Right!")
As the Bible relates, even Heaven has it problems (cf. the abortive coup by Satan and his associates). But the problems of a posthuman world founded on gradients of superhuman bliss will differ from the miseries of Darwinian life.
Are opioid users likely to vote Trump?
("Trump's Election Map Also Maps "Despair Deaths")
Presumably the brand risks reputational damage, though the product has a loyal user base.
("Florida heroin bust reveals something bizarre: Drugs covered in Donald Trump’s face")
When will social life no longer resemble a baboon troop on the African savannah?
("The twisted psychology of office cliques")
Utopian communities can work if we design & safe sustainable analogs of MDMA - or hardwire its genetic counterpart.
Can the USA ever aspire to Cuban levels of healthcare? Or at least match UK or Scandinavian standards? We can argue about the optimal balance of free markets / state intervention in a modern liberal democracy. But treating healthcare as a "market" debases civilised society.
("Sen. Tom Harkin says Cuba has lower child mortality, longer life expectancy than U.S.")
The environment: who needs it?
("US Republicans launch effort to abolish Environmental Protection Agency with new bill")
Will transhumans transcend chimpanzee politics or offload political power to smart AI?
("Chimps beat up, murder and then cannibalise their former tyrant")
Humans kill billions of sentient beings from other species, and will probably soon kill billions of their own. The long-term future of life may be superhuman bliss, but the 21st century is likely to be grim.
("The truth about Satan: Nuclear war in the 21st century")
Hard drugs and refined sugar are best discouraged in the young.
("High-sugar diet programs a short lifespan in flies")
Transhumanism: we should phase out aging - but also archaic Homo sapiens:
("'To Be A Machine' Digs Into The Meaning Of Humanity")
Do you view your body as a performative machine or the temple of the soul?
("Why more men struggle with aging ")
Downhill all the way?
Or the best is yet to come?
I look forward to a "better-than-well" wonderpill.
("Here Are the Ages You Peak at Everything Throughout Life")
Can osteopontin rejuvenate?
("Old blood can be made young again and it might fight ageing") Should we be consuming prickly pear and brown seaweed extracts?
("Chemicals in ubiquitous Mediterranean plants may hold key to delaying neurodegenerative diseases, study suggests")
Or evolution says recursively self-improving organic robots will rewrite their own source code and enjoy eternal youth.
("Billionaires say they’ll end disease: evolution says otherwise")
Transhumanism: solving the problem of death entails not murdering billions of sentient beings in slaughterhouses.
("Only Human. Meet the hackers trying to solve the problem of death")
[on the AntiSpeciesist Revolution]
The Antispeciesist Revolution
("The AntiSpeciesist Revolution")
Thanks David. We do need an Antispeciesist Revolution. My only reservation is the term. Naively, it sounds as though the antispeciesist is arguing that "all animals are equal" - and that in the case of a severe and irreconcilable conflict of interests, we should give as much weight to a mosquito as to a mammal. Not so - though of course we may hope in the long run even the humblest of beings can flourish unmolested.
Eat sparingly with optimal nutrition is a better recipe for a long, healthy life than dietary heroics. (IMO)
("Calorie restriction diet extends life of monkeys by years")
The quest for eternal youth: is it better to bathe in the blood of young virgins or fill your veins with their blood?
("For $8,000 this startup will fill your veins with the blood of young people")
Optogenetics: should we create happy & gentle mice or "Walking Dead-style zombie killers"?
("Mice can be turned into Walking Dead-style zombie killers, scientists discover. Researchers use a laser to activate brain cells that prompt the mouse to hunt and kill anything in its path – crickets, bottle tops, sticks ...")
Some accidents might easily damage your career prospects.
("Queen 'nearly shot' by Buckingham Palace guard while walking late at night")
Translational medicine 2017:
("A cure for all ills: what medical advances can we expect in 2017") "All wars are civil wars, because all men are brothers." (François Fénelon)
("And all sentient beings are cousins.")
Admirable. But UBI won't recalibrate the hedonic treadmill. Only genetic engineering can create a happy world.
("Want utopia? Start with universal basic income and a 15-hour work week") Right-hemisphere euthanasia?
("I only have half a brain")
(cf. https://www.ncbi.nlm.nih.gov/pubmed/20603163 - "Depression and the hyperactive right-hemisphere.")
A rose by any other name...?
("Can you match the correct names to these faces? People identify strangers with amazing accuracy")
Will transhumans and posthumans be rude, civil, or endowed with superhuman courtesy?
("How Rudeness Stops People from Working Together")
"For there was never yet philosopher That could endure the toothache patiently." (Shakespeare, Much Ado About Nothing)
Tideglusib: good news for philosophers - and future humans?
("'Tooth repair drug' may replace fillings")
Assuming semantic content is really real leads to paradoxes / contradictions.
("The Outer Limits of Reason")
"Makeup is my art" - and the UK art market isn't cheap.
("Average British woman spends £70,000 on her appearance in a lifetime, research finds")
Mirror-touch synaesthesia: let's hope in future we can all get off on each other's pleasures...
("Meet The Woman Who Can't Help But Feel The Pain Of Others")
Decent homes for orangutans will take time, but learning carpentry is cool:
("The incredible moment an orangutan uses a SAW to cut tree branches after watching builders do the same")
A conservation effort or mothballs: what is the likely fate of Neopalpa donaldtrumpi?
("Researchers identify source of opioids' side effects")
From the Mpemba effect to sentient organic wetware, the properties of water are still unexplained.
("The claim that hot water freezes faster than cold water just got even weirder")
The evolutionary roots of teenage angst run deep.
("Worms have teenage ambivalence, too")
1 : 2 : 3 : 4 : 5 : 6 : 7 : 8
David Pearce (2017)
The Abolitionist Project
Social Network Postings (2016)
The Hard Problem of Mind Solved(?)
Can Science Abolish Suffering? (2013)
Hedonistic Imperative Facebook Group Posts