“Reductionism is well-named. It reduces what you can do.” – Michael Levin – Sentientism Ep:199

Find our Sentientism Conversation here on the Sentientism YouTube and here on the Sentientism Podcast.

Michael is a developmental and synthetic biologist at Tufts University where he is the Vannevar Bush Distinguished Professor. He is also an associate faculty at the Wyss Institute at Harvard. Michael is a director of the Allen Discovery Center at Tufts University and the Tufts Center for Regenerative and Developmental Biology. He is also co-director of the Institute for Computationally Designed Organisms. Michael’s Levin Lab focuses on “Embodied Minds: understanding diverse intelligence in evolved, designed, and hybrid complex systems” and works “at the intersection of developmental biology, artificial life, bioengineering, synthetic morphology, and cognitive science”. This work includes the bioengineering of novel living machines and has clinical applications in regenerative medicine. Michael has editing roles at a number of academic journals and has published more than 350 papers.

In Sentientist Conversations we talk about the two most important questions: “what’s real?” & “who matters?”

Sentientism is “evidence, reason & compassion for all sentient beings.” In addition to the YouTube video above the audio is on our Podcast here on Apple & here on all the other platforms.

00:00 Clips!

01:18 Welcome

– “We biologists have plenty of spherical cows”

03:50 Mike’s Introduction

04:37 What’s Real?

– A “not religious but highly spiritual” Jewish background

– Hebrew school ” I harrassed everybody with questions…. how souls are supposed to work…the hard problem of consciousness… the answers weren’t terribly forthcoming.”

– At home “an emphasis on inquiry… asking big questions… things that matter”

– “The question of ‘how do I know?’ figured prominently in my childhood”

– “Rationality is an amazing tool but one can also ask questions about its limitiations… what are the things that we’re not seeing”

– Being sceptical about the approaches you’re bringing to a problem

– “I pretty much only have one supernatural belief which is that the universe is understandable… Once you’ve taken that on everything else becomes possible… I can’t think of anything that would be truly supernatural…”

– “We are finite beings” trying to understand reality then “you have to ask yourself ‘how is this working out?… is this helping me have a more meaningful life… be a more ethical person… have better relationships with others?'”

– “I don’t really think of myself as a biologist… my fundamental commitment… has been to understand embodied mind”

– “I’m interested in cognition, intelligence and inner perspective in a wide range of diverse systems… some of which are alive”

– “I think cognition is a broader category than life”

– “It just so happens that life is, so far, our best example of how that can scale”

– “Molecular networks scaling into cells scaling into tissues scaling into organisms and beyond”

– Collective intelligence, goals… “biology is an excellent playground for these things”

– Cognition, mind etc. “All of these cognitive claims… where you think something is on the spectrum… how much mind… I don’t think these are terms describing particular systems. I think these are terms describing our intended relationship to them… an engineering interaction protocol”

– Are humans machines? “If you have an orthopaedic surgeon who does not believe you are a machine – you are in trouble… If you have a spouse or a psycho-therapist who thinks you are a simple machine – you are also in trouble.”

– The TAME approach “Technological Approach to Mind Everywhere”

– “Multiple observers can differ as to their assessment of any given system… but it doesn’t mean that anything goes… observers can then compare ‘how well did that worldview work out for you?'”

– E.g. Animism “there’s a spirit in every rock”… how does that perspective work out for you? – what does it help you do?

– “The thing about treating these things as empirical questions not philosophy is that you are often surprised… that’s what’s good about science… an opportunity to learn”

– What tools to apply?: physics & simple engineering, cybernetics & control theory, behaviourism & learning, communication & psychoanalysis?

– Surprises: Very simple gene regulatory networks (no cell, no magic) – “that system, which most people would assume has zero cognition… is capable of 6 different kinds of learning including Pavlovian conditioning”

– Sorting algorithms: “Even those things, when you probe them the right way, show novel problem-solving capacities and behaviours that are not in the algorithm themselves”

– “We really need a sense of humility about what things can do and what kinds of tools are best appropriate for them… we don’t know… we’re not good at guessing… we need a mature science… diverse intelligence research”

– Intelligence as “competency in navigating a problem space”

-“I don’t think that any of these categories are usefully binary”

– “What I think is important about inner perspective is this: for any kind of system you want to be able to answer the question ‘how important is the system’s perspective on to the world for me to understand what’s happening?’”

– E.g. a bowling ball on a hilly landscape “If you want to know what that bowling ball is going to do, your view as a 3rd party observer… of the landscape tells whole story – you don’t need anything else… to predict what’s going to happen next.”

– “If you want to now know what a mouse is going to do on that landscape – your view of that landscape isn’t really very interesting at all… The important thing is the mouse’s view of that landscape… an internal representation of the valence… rewards and punishments… attention… The inner landscape of the mouse is the actual landscape that’s going to determine what happens next.”

– “Different kinds of systems have different degrees of representation of their outside world – it’s not just brainy, smart animals… all systems represent the world to some extent. The question is how much.”

– A system’s Cognitive Light Cone “what are the events both back in time and forward… and spatially… what do you need to know?”

– “For a bowling ball the cognitive light cone is tiny… you add up all the pushes & pulls on it you’re more or less done”

– But for even slightly more complex systems (e.g. networks of genes)… “that’s no longer satisfactory… they learn from experience… things that have happened before are going to make huge differences to what happens next… you cannot gain a full understanding… without understanding their perspective.”

– “Inner perspective is to some degree wider than biology… but what biology is really good at is scaling up these cognitive light cones”

– A rock has the same level of light cone as its pieces but “we call living things any system that is good at scaling up the cognitive light cone of its parts”

– Most cognition to date has come through biology but “Evolution has no monopoly on producing cognition”

– “I don’t see any reason whatsoever why the rational efforts of other cognitive agents such as ourselves couldn’t make new minds”

– Sentience and consciousness “those are essential… my resistance to binary categories is not to reduce these… on the contrary it’s to have a better understanding of them and their potential”

– Getting pushback from “mechanistic, supposedly reductionist scientists… ‘this is crazy – you’re painting feelings on to cells… chemistry is the best explanation!’” – “they’ve picked a level” (chemistry but no deeper e.g. quantum foam)

– Also pushback from organicists… “by putting non-living things… somewhere on that spectrum… you’re undermining this battle… to preserve the magic and the importance of life and of inner perspective”

– “I don’t think understanding these things better and understanding how they scale up reduces the importance or the majesty of the obvious sentience of living beings… it helps us to understand what we are and what our potential is and what we need to do”

– “This weird obsession with chemistry and with physics… is what undermines the efforts to really understand this”

– “I often hear people say… ‘that’s not real goals… preferences… valence… real anything… it’s just following the laws of physics and chemistry.’ Well guess what, if you zoom in to your brain or the rest of your body guess what you would find…  you wouldn’t find fairies you would find physics and chemistry.”

– Even fairies “would have to be following something”

– “If you’ve got a model of the soul… explain how that works and how that solves these problems… and we’re good to go!”

– The paramecium or a single cell “Do you or do you not believe that that thing has to some small extent a preference as to what happens?”

– If you don’t – consider embryonic development where “You go very slowly and gradually from a single cell into whatever it is that we are… and there is no magic lightning flash that converts the chemistry of the oocyte into the mind of the adult human”

– “This continuity – you can’t escape it. What we owe is not a story about the magic of bright lines we owe a story about scaling.”

– The movie Ex-Machina where the protagonist cuts open his arm to work out if he’s a robot. Michael getting emails saying “I’ve read your paper. I understand now that I’m a walking bag of cells. Now I’m depressed… this is terrible.”

– “You cut open your arm and you find a bunch of cogs and gears… you’ve had 40-50 years of experience in your own skin of being an agent… exerting effort towards various outcomes… having moral quandaries… having qualia… and when you see those cogs and gears what you’re going to decide is ‘none of that was real’ because I’m committed to the idea that cogs and gears can’t support [them]… this is so bizarre to me”

– “Why are you so attached to protoplasm and proteins… over something else that you might find [cogs and gears]. What do you think you know about those things that overrides your primary experience as a human?”

– “It’s amazing to me that what people would naturally tend to say is ‘Then I guess I’m not real I’m not important…’ vs. ‘Wow I guess I just learned something great about cogs and gears… not too much more shocking than finding out that proteins and carbohydrates… can do it”

– 2 common beliefs: “If your parts obey the laws of physics then you are somehow are reduced… this allegiance to a low level of description is completely arbitrary.” Dennis Noble’s work against privileged levels of causation.

– And “The idea that we know what various kinds of materials and architectures can do… we have absolutely no idea.”

– “It’s not emergent complexity here it’s emergent goal-directedness and emergent cognition. That happens even in very simple systems… it surprises us constantly. We’re not good at predicting it.”

– “People who say ‘I know what this is because I made it’… a really dangerous idea. You don’t know… Just because you know the parts that went into it… we do not yet know what it is and what it can do… what kind of inner perspective it has… a lot of humility is needed here.”

– “Determinism is well and good if what you’re interested in doing is looking backwards… you can always do that… But we’re interested in going forwards… that is a completely different kind of test.”

– Cellular automata driven by very simple rules “It’s fully deterministic”. Conway’s Game of Life… glider patterns “the pattern itself moves”. “You could say ‘there are no gliders’… that isn’t exactly wrong… but it’s absolutely limiting… you won’t ever build a Turing machine built out of gliders – which somebody did… it closes off that level of inventiveness… higher levels enable you to have a richer relationship with a system going forward.”

– Computational irreducibility

– Opportunities for control via higher levels (vs. atomic level control) “you can just train the rat”

– “Understanding the large-scale capabilities of your system is critical in having a… predictive, powerful, productive interaction going forward”

– Limitations of current regenerative medicine “the standard operating assumption is that cells are mechanical agents… all the excitement is about the hardware… CRISPR, genomic editing, pathway rewiring… that closes off a huge number of potentially really powerful interventions… cellular training… cellular problem solving…”

– “Reductionism is well-named. It reduces what you can do. It reduces your ability to exploit what’s really powerful about these systems.”

– Training animals without knowing neuroscience “that learning interface… cells and tissues do this too”

– Concepts of cognition and sentience “are practical and essential… it’s a testable claim… it helps you do better in the empirical world.”

– “The best thing I hope my lab does… on their best day… is to take these ideas from deep philosophy… and bringing them to the bedside… to the clinic.”

– “That inner perspective… by asking what do cells remember… what do they care about… what are their goals… these are questions that lead directly to therapeutics… regenerative medicine… cancer.”

– “I don’t feel in any way diminished by the idea that there are basically an infinitude of other minds… highly diverse minds… endless sentient forms throughout the universe… also having experiences that thrive, that suffer to various degrees… It seems childish to somehow pin the quality of what we have on the fact that no one else has it.”

– “I would like to scale my cognitive capacity… humanity to scale it’s cognitive capacity… not just its intelligence but the radius of concern, of care, of compassion… I think we need to scale that up.”

51:04 What and Who Matters?

– “It’s important to not pretend that once we understand how sentient something is we are automatically good at treating it ethically”

– “Factory farming… we all know pigs value their experience… they feel pain… yet here we are… What we know and how we treat them are not the same thing”

– “Assuming we all survive the next few decades… for the flourishing of humanity, of ecosystems, other beings on earth… really requires the deep lessons of diverse intelligence”

– “There is no such thing as the magical standard human”

– “Evolutionarily we have a lineage going back to single cells… all those things are continua… the same is true of embryonic development”

– Neuro-atypical humans, bodily modifications… “you’re going to have humans that are so modified in body and mind… cyborgs, hybrots with various biological and technical changes”

– “We have been, for centuries, lulled into a false sense of security… It was easy… does it speak? If it speaks then it’s like us. Even that humanity failed – we were really terrible to each other for much smaller differences”

– “Those [easy distinctions] are gone… It’s going to be essential that we learn to have an ethical… synthbiosis with other beings that are nowhere with us on the tree of life.”

– “Science fiction dealt with this a long time ago but somehow people today have lost sight of some of those things”

– AI and Large Language Models “they’re so different from us that it’s very easy for people to say ‘that’s not the things we want to care about’… what are you going to do when you’re confronted with humans that have 51% of their brain replaced by various silicon appliances… and they’re linked together with other people and also a few Ais… all of this is coming”

– “We’re going to have to develop better, more principled ethical frameworks for dealing with beings that are just very alien in their construction.”

– “What do all these beings have in common?” and the Cognitive Light Cone idea “what is the size of the biggest goals you can pursue regardless of your embodiment or implementation or origin story?”

– “Begins to give us a way to apportion our degree of moral concern to beings that are capable of pursuing goals… and suffering when those goals are not reached”

– “The diversity of who and what we think counts is going to be enormous”

Josh Gellers Sentientism episode

– JW: Naturalistic epistemology and sentiocentric compassion (Sentientism) condemns animal agriculture and exploitation… but social norms

– “In talking about these AIs instead of the problems with existing lifeforms [e.g. farmed animals] is a huge issue”

– Having kids as an analogy for the creation of and alignment of powerful AIs “we have very limited ability to make sure… that they’re aligned with our values… some of them go on to do wonderful things and some of them go on to do horrible things… that’s already an issue… in many ways these are not new problems”

– “The one unique thing about AI intelligences… they can be copied”

– “I’m hopeful that with the advent of different ways to grow food proteins… we’re going to eventually do away with animal farming”

– The ethics of animal research “In order to do the things that somebody would do when they go fishing… we have three months of paperwork and oversight by veterinarians and ethicists… as there should be”

– The risks of risk aversion in research

– “We are the process of a meandering evolutionary path… not optimised for any of our values… we are where evolution dumped us with all our ridiculous susceptibilities…”

– JW: Especially when we extend our consideration to all sentient animals… “If there’s some telos to this universe it’s a pretty brutal one”

– “The number of emails I get every day from people with the most unbelievable examples of biomedical suffering is just incredible… It is utter moral cowardice to focus on what we shouldn’t do… as opposed to our duty… we have the ability to rationally approach these matters… to try to guarantee a better embodiment for sentient beings on earth.”

– “The number of people that are suffering because of our ignorance… and also our timidity… our unwillingness to take responsibility… it is on us now to improve this for people… for ecosystems… for other beings”

– “The amount of ignorance and inaction that is letting sentient beings suffer worldwide”

– “I take animal use ethics extremely seriously… it is our responsibility to relieve the suffering worldwide… science has to go on… we have to do experiments that are going to enable us to improved embodied experience for all”

– The Buddhist idea of an “inauspicious birth… you can’t make progress towards enlightenment”… “At this point… every birth is an inauspicious birth… all of us are limited by the vagaries of random mutation and selection… viruses… all these dumb things that are in no one’s best interest.”

– A future dialogue between children and teachers… “You’re telling me you’d just have to stay in whatever body you randomly got… and if you had some sort of defect you’d just have to live that way?… and if the IQ and the lifespan that you were born with wasn’t enough to fulfil your goals then that’s it?… your bad luck… that will be unthinkable to modern humans”

01:08:53 A Better World?

– “The bigger picture?… Freedom of embodiment. Every sentient being should have… we should offer assistance… offer the ability for sentient beings to have better embodiments to fulfil their goals”

– “What that requires is complete control over growth and form… direct what it is that cells build”

– Regenerative medicine… “If we knew how to get cells to build specific anatomical structures… birth defects, traumatic injuries, cancer, ageing, degenerative disease…

all these things would go away if we could communicate our goals to groups of cells”

– An “Anatomical Compiler… you could just specify the shape of the thing you want”

– Vs. genetic engineering / CRISPR “I think we have a much more practical path to it… the genetics specifies the hardware… but programming at the level of hardware is really hard”

– “Honouring the agency of the cells and tissues and being a collaborator with them”

– “I think CRISPR and those kinds of things are going to flatten out soon… after you’ve picked the low hanging fruit of single-gene diseases… what genes do you edit?”

– “Initially… reprogramming tumours and regrowing limbs and growing new organs… all of which we’ve done in various model systems… but ultimately it leads to freedom of embodiment… there is no reason any of us should be stuck in the embodiment that we got… those things were not chosen with your welfare in mind… shaped by forces that don’t honour any of our values”

– Transhumanism

– The risks of Michael’s work: “For sure there are going to be risks but I’m pushing in the opposite direction of hubris… my argument is that things are so bad right now for so many that it is a moral imperative… the opportunity cost of doing nothing is absolutely massive.”

– “Let things go how they are now – that’s simply not acceptable.”

– “What does humanity look like 100 years from now? Are we still getting lower back pain and bad knees and dying into our 80’s?… That’s it?… that would seem such a waste of this incredible gift of intelligence and compassion that we have”

– “People should spend time painting the future that they do want”

– Science fiction: “Are you saying it’s impossible or just that it’s going to take a while?… Everything that’s possible is going to be done.”

Follow Michael




Sentientism is “Evidence, reason & compassion for all sentient beings.” More at Sentientism.info.

Join our “I’m a Sentientist” wall using this simple form (scroll down a little!)

Everyone, Sentientist or not, is welcome in our groups. The biggest so far is here on Facebook.

Thanks to Graham for the post-production and to Tarabella and Denise for helping to fund this episode via our Sentientism Patreon.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.