Michael Levin, Scientific Panpsychism and Teleology
I’d like to draw the attention of all those interested in animal welfare to the work of Michael Levin, as I haven’t seen any mention of his work in the animal-welfarists, and I think everyone down the shrimp-mines[1] could probably benefit enormously from reading him.
tl;dr Michael Levin has a great new (well, new to me—from 2022) paper out that reviews a tonne of fascinating research with important implications for animal welfare, and even goes as far as to outline suggestions for research programmes in this area. I recommend any and all who are interestd in animal welfare to read the paper, it is open access here.
“It is imperative that we abandon the tendency for armchair pronouncements of what can and cannot be seen as cognitive and instead develop toolkits for generating and testing teleonomic models of arbitrary systems.”[2]
Background: Dr Levin in the Morphospace
My favourite sentence ever was spoken my Michael Levin in a talk in 2018:
“We can reach regions of the morphospace not explored by evolution”
Understanding this sentence should get you started on understanding Michael Levins work.
noun The totality of possible body shapes that a species or other taxon can assume as it evolves, or that an organism can take as it develops, as defined by genetic, ecological, or other factors.
The practical applications of exploring the morphospace come down essentially to regenerative medicine. But Michael is also deeply interested in the epistemological ramifications of his work.
Another worthwhile takeaway from that talk is the metaphor of moving (in the history of computing) from direct hardware manipulation to machine code to higher-level languages:
“However, as shown clearly by the trajectory of the information sciences, this is only the beginning; the next frontier is the software of life: developing a mature science of prediction and control over system-level phenotypes.”
Michel Levin is also the developer of Xenobots, a novel life form that was discovered when asking the question,
“what would happen if skin cells were removed from a frog embryo, dissociated and given a chance to reboot their multicellularity in vitro”
to which the answer turned out to be
“Without the normal instructions from the rest of the body telling these skin cells to form a passive, two-dimensional boundary layer to keep out the bacteria (a system of low agency), the true capacities of this cellular collective are revealed; it forms a three-dimensional individual with a more exciting life of self-initiated motile behaviour. The collective intelligence of these cells is revealed as, despite a novel environment and novel internal configuration that never existed in the frog evolutionary lineage, they discover novel ways to be a coherent organism.”
Levin does a lot of interesting work, but moving on to the main topic:
Panpsychism and the Cartesian Cut
Most relevantly to welfareists is the focus on hierarchies of computation:
“B, the degree of competency and complexity that can be handled by a system in its pursuit of goal states defines major transitions along a continuum of cognition ranging from passive matter to advanced self-reflective minds, which enables comparison of highly diverse intelligences”
and the implications of this view of biology in philosophy:
“Ontogeny recapitulates phylogeny, in that all of us have made the journey across the Cartesian cut (from the ‘mere physics’ of molecular networks in a quiescent oocyte to a thinking being) not only during evolution but also during our own lifetimes. What aspects of biological structure enabled mind gradually to develop and expand?”
While Levin avoids outing himself as a Panpsychist, this seems to be largely for pragmatic reasons:
“In no way are we embracing panpsychist or panexperientalist approaches, as could be misleadingly inferred … we wish not to dilute our research into unending philosophical debates.
...
Consciousness is emerging as a basic and inherent property of biological organisms which is relevant for their survival and evolution. Importantly in this respect, plants and several unicellular organisms generate endogenous anaesthetics any time they are wounded or stressed. The hypothetical basic unit of consciousness in multicellular organisms, such as humans, non-human animals and plants might be represented by cellular and subcellular levels of consciousness. There are at least three possible sources of sentience and consciousness (understood as a gradual self-mapping tool) at the cellular and subcellular levels.”
~ Slime mould: the fundamental mechanisms of biological cognition
similarly in an interview with Sean Carrol, he talks about panpsychist ideas quite comfortably:
“Let’s say people who are into panpsychism or this idea that some sort of intentionality is everywhere fundamentally, one of the problems that people often have with this is that they scale down the physical system, they say, okay, now let’s consider a rock or something like that, and then they fail to scale down the intentionality, they say, well, it’s ridiculous to say that a rock has hopes and dreams, and well, of course it is.
So what you need to do is you need to proportionately scale down the cognition, so if you ask yourself what would intentionality or freedom in the sense of indeterminacy look like in the simplest possible case, I think what you get is exactly what particle physics is telling us. So if you ask what would freedom look like at the most minimal kind of instantiation, I think you would predict something that looks like quantum indeterminacy, and if you ask what would goal directedness look like the simplest possible layer of reality, I think you would get exactly what these action principles look like.
So one of the things that we’re working on now is to try to understand the scaling of goals from tiny homeostatic goals, like either chemical reactions to try to keep certain entropic principles in particular ranges, or single cells like bacteria that try to keep metabolic states in particular ranges, how do you scale goals from these very modest types of goals to something that’s much larger, like the goal of having a properly shaped hand”
The implications of panpsychism for animal welfare are obvious.
Some more relevant quotes:
“Chimaerism is a type of conceptual universal acid, dissolving existing terminology that is not based on deep concepts but is instead a relic of parochial contingencies of our familiar forms. The ability to mix any two biological systems, in any arbitrary proportion, reveals the continuous nature of terms that are often thought to be binary. This strongly extends the gradual view, already required when taking evolution and developmental biology seriously, which lends no support to any sharp line between ‘true cognition’ in complex brains and ‘just physics’ that is sometimes said to occur in their phylogenetic and ontogenic precursors.”
“A key aspect that is revealed by this way of viewing life is the multi-scale competency; the basal intelligence of many levels of organization in a given body, all of which exhibit teleonomic behaviour in their own problem space. Recognizing this behaviour, and the specific goals being pursued, is an IQ test that scientists (observers) take when evaluating the agency of unconventional agents; our ability to detect, understand and manage teleonomy says as much about our intelligence as it does about the system being studied.”
“What is the relationship or overlap between the sets demarcated by ‘life’ and ‘cognition’? If all (most?) components in living things are teleonomic agents and are thus somewhere on the continuum of cognition, are all living things cognitive? What is a useful definition of ‘life’, given that teleonomic agents can be produced by engineering with organic or inorganic parts? Although modern life is necessarily teleonomic (in order to survive in the biosphere), could there have been very early life forms that were not teleonomic? Could current efforts at truly minimal synthetic life clarify the relationship between teleonomy and physics?”
“The impact of rapidly advancing technologies will go far beyond science, encompassing many issues that have been dealt with in science fiction but have not yet been worked through by thinkers in the fields of philosophy of mind, ethics and policy. Our culture is in for an upheaval that will far eclipse the controversy that fomented around Darwin’s ‘Origin of Species’. Our ethics structures are barely sufficient to optimize intraspecies interactions; they will need a total overhaul when we are surrounded by novel agents of every possible configuration. How does one prosecute a 50% cyborg for its misdeeds? Legal systems based on one dimension of human IQ (with a cut-off for ‘competent’) are in no way ready for a multidimensional space of beings with diverse components contributing to their mind and behaviour.”
“Moreover, although Darwin’s revolution created a continuous spectrum along which to compare intelligence (and place moral categories for relationships to primates, whales, dogs, etc., in addition to complex cases, such as octopuses), a far wider reality is beginning to emerge. There is not only a large axis representing different amounts of cognitive sophistication, but also an immense space of multiple dimensions of different types of cognitive capacity. It is entirely unclear how creatures within that space will relate to each other, making essential the search for an invariant on which to base norms (such as complexity of the creature’s goal space).”
Worth noting that I don’t think this sort of panpsychism makes the work of animal-welfare researchers in any way pointless or unimportant, rather the opposite. I share this because I think the quality of the thought is exceptional, and anyone trying to construct a useful hierarchy of moral systems will do well to learn some ideas from it. I know from my brief time reading on the topic that some of the animal welfare literature already works with teleonomy, and I imagine that an expansion of this sort of work and perhaps some gleanings from economics literature on agency could provide a more useful basis than the focus on pain receptors and simple stimulus-response loops that I have seen some people pursuing.
Finally:
“What defines a self? Although immunological and evolutionary-scale theories have been proposed, these are intimately tied to (and thus limited by) the types of organisms we observe naturally in our biosphere. A more general framework, able to encompass and directly compare truly diverse agents, needs to be based on teleonomy at its core: a self is a goal-directed system, and its level of sophistication (ranging from modest inorganic systems to trans-human beings) is set by the spatiotemporal scale of the goals it can pursue.”
“Crucially, this new field suggests not only novel capabilities and advances in knowledge, but also the need for a new ethics. The frequently voiced statements that ‘living things are not machines’ reflect an outdated essentialism and a type of magical thinking that trusts in clear, binary lines separating evolved living beings from designed machines to define our moral duty to various agents comfortably. These lines do not exist, which will be made painfully clear in the next decades as we become surrounded by collections of agents that make the iconic Cantina scene in ‘Star Wars’ look tame in comparison. Significant effort will need to be made as science and society mature to include designed beings in addition to natural beings, in order to avoid the types of ethical lapses to which humans are prone: mistreatment of those who do not resemble a familiar in-group in composition or origin. The nature of teleonomy as a guiding principle cutting across contingencies of origin story and composition, and the inevitable expansion of life throughout the option space of hybrid forms, provide important conceptual tools for a path forwards to a future where we cannot simply guess the capacity of an agent to think and suffer based on what it looks like or how closely it resembles a familiar touchstone species in the Earth’s phylogenetic lineage. It is not clear what a new ethics of life as it can be will look like, but some sort of golden rule about compassion towards systems proportional to their teleonomic capacity might be a place from which to start.”
Conclusion
My only intention was to encourage people to read this paper by Michael Levin, and explore the many ideas and links therein.
I would also like to wish any and all people working on animal welfare and other difficult problems where philosophy and decision theory interact in often brutal ways a good and pleasant 2023, and great success in their endeavors.
- ^
I was myself down the shrimp-mines for a while, trying to collaborate on an essay on consciousness with an ideologically fundamentalist vegan friend/molecular genetics researcher. I have less and less faith in my ability to finish even a coherent essay on the topic, though I wish those still down there all the luck in the world.
- ^
Unless otherwise specified, all quotes are from this paper: Endless forms most beautiful 2.0: teleonomy and the bioengineering of chimaeric and synthetic organisms
Thank you so much for this, it is extremely useful and interesting. I find Levin’s work really thought-provoking and I am particularly interested in his emphasis on ethics, which is how I encountered some of his articles in the first place. I guess you may know or be interested in this recent article of his: https://www.frontiersin.org/articles/10.3389/fnsys.2022.768201/full
It’s weird and stimulating to read how in his view we’re gonna be surrounded by diverse “beings”—whichever definition we give to this—that may be beyond this cut-out definitions predominantly adopted up to today (even in philosophy). Still, developing new ethical framework may really prove essential (Levin is well aware of many topics discussed in this forum, it appears, so he’s really a good source).
PS. this is my first comment ever here on EA Forum, but I really find this stuff exciting (I am trying to work on ethics / expansion of the circle beyond the common categories, for reasons quite close to Levin’s ones). I would be very interested in anything and anyone with some perspective regarding not only animal ethics, but further and, potentially, with a keen eye on anthropocentrism (for / against / shades of), maybe one day I’ll write something here too—so thanks again!
Thank you! Glad I got someone else to crawl out of the woodwork, welcome to the EA forum :). (I’m also more of a lurker.)
The paper you linked is indeed fantastic, I think it might be my favourite piece of writing ever. I read it after I wrote this hence why I didn’t talk about it. There are two other things I would point towards w.r.t to the topic of panpsychism:
- The Computational Boundary of the Self (another great one from Levin)
- If Materialism is True, the United States is Probably Conscious (by a fella called Schwitzgebel)
And w.r.t what you are saying about being surrounded by diverse beings—I think thinking about cyborgs and engineered entities is a great way to prepare the ground for this kind of moral shake-up, but the more I look into this the more I think that we are already surrounded by weird and wonderful beings—distributed group entities at all levels, ecology, economy, politics, sociology.
I am still thinking and reading about these topics a lot, and in the next months I’ll have some free time to maybe do some more writing about it.
In a way I see why animal welfarists are avoiding panpsychism at the moment—the absolute hellscape of welfare that is the meat industry definitely feels more pressing than, like, figuring out how to measure the welfare of the USA or the brazilian rainforest or so. Though maybe in the long run, seeing these entities as agents with hedonic valency will be a good and useful way of talking about their health and the way they act in the world. What is the impact of tictok on the mental health of the entity that is the USA? Maybe this is actually a very good and important framing, and useful?
Lets see!
There are certainly certain systems and structures that give rise to greater capacities in/of consciousness, and Levin knows this.
For example, the targeted chemical senses of a cell, with its direction-knowing proteins, and the dna/rna and morphogenic bioelectrical system (ion channels to “speak” to neighbouring cells, etc)—versus endocrine/hormone systems which allow for widespread (multi-cellular) internal communication (and work with chemical sense cell clusters to identify stimulus from outside the body) that we see in plants, fungi, and animals—versus nervous systems which allow for nearly instant (for practical purposes) internal communication (with enervated sense cell clusters for outside body stimulus spotting), which Levin himself conceives that evolution/mutation “learned” from the previously stated, much simpler morphogenic bioelectric system most cells (neuron and non-neuron alike) use. If Levin somehow became (some sect of a) panpsychist because he realised small stuff has intent due to its mechanics, and big stuff has more, longer-lived, deeper-thought-out intent because it built its stuff on former smaller stuff, then I guess all modern biologists fall under the panpsychism umbrella (I type, facetiously).
I agree some of his language is panpsychically suspicious—and his desire to pull in buddhistic stuff, also a common red flag, but I don’t think there are enough red flags to worry (yet).
He’s using “self” as “Agent + PoV” it seems. Philosophically (I hate how big that word is), “Self” (at least, in theory of mind) is just “consciousness” (for humans: a duel-stream of experience, filtered through the limbic system, flashlit upon the lens/desktop of sensory memory and short-term/working memory) + “memories” (long-term memory, decompartmentalize-stored, prior experiences that are haphazardly recompartmentalized (“recalled”) back into the ongoing duel-streams). Not but the result of processes. An “illusion” (if you want to really piss off dualists). Yeees, “monoist” physicalism. Which seems to function perfectly fine under Levin’s prescriptions.
tl;dr eh who gives a fuck, I’m a rule utilitarian; Levin seems to prescribe towards-the-truth research and is moral about it. Thank you for the very cool read, and I hope you write more and more on these and other topical roots of reciprocity, mirroring, and altruism.
I am very interested in teleonomy now, for sure.