### What is a Self? This question is intimately related to the question [[What is consciousness?]]. The question ultimately is asking us to consider what _kinds_ of systems could be conscious. Is any collection of atoms conscious? Or is there some special kind of property that a system must exhibit in order for us to consider it having experiences? What we know for sure is that humans are conscious. But what is it about the brain/neurons that give it the capacity to have subjective experiences. Why isn't our pancreas the seat of consciousness? (We know it isn't the seat of consciousness because we can get it removed and still retain the capacity to have subjective experiences. While we can't have our brain removed and still be conscious). Two papers by Michael Levin, Chris Fields Research and their colleagues seem to provide a good starting point in thinking through the question of Self: - [The Computational Boundary of a “Self”: Developmental Bioelectricity Drives Multicellularity and Scale-Free Cognition](https://www.frontiersin.org/articles/10.3389/fpsyg.2019.02688/full) - [How do living systems create meaning?](https://www.mdpi.com/2409-9287/5/4/36) - [Minimal physicalism as a scale-free substrate for cognition and consciousness](https://academic.oup.com/nc/article/2021/2/niab013/6334115) There's also [a non-technical essay](https://aeon.co/essays/how-to-understand-cells-tissues-and-organisms-as-agents-with-agendas) that elaborates on the same ideas. #### What can we learn about the Self from biology and evolution? **What we know for sure is that: - Humans are biological organisms - Humans are multicellular organisms. We are a collection of billions of cells, each of which is apparently "non-conscious" (according to our intuitions) - Even if we doubt other creatures are conscious, it's hard to doubt that humans are conscious - At the extreme even if you doubt that other humans are conscious, you can't doubt that you're conscious (if you're doubting, simply prick yourself with a pin) From the study of life and tracing evolution back in history, it is obvious that **_nothing_ is discontinuous in biology**: - That is, between two distinct features of life, with enough effort, it's possible to find intermediate forms. - For example, here's [how the eye evolved](https://en.wikipedia.org/wiki/Evolution_of_the_eye). - As elaborated in [[Sentient beings that are capable of suffering]] and [[Evolution of brains, consciousness and suffering]], we can limit the capacity to have an experience to living beings that have evolved senses, sufficient neural capacity and structures to integrate sensory data into an internal map and show proactive preferences. - But the [bioelectric basis of neural communication is actually continuous with bioelectric based sensing in single celled organisms](https://www.frontiersin.org/articles/10.3389/fpsyg.2019.02688/full) and communication in simpler multicellular organisms such as plants or fungi. - Our sense of taste depends on ion-channels that we share with bacteria who sense their environment via the same ion-channels. So, we literally sense via the same mechanism. - In this way, even if the experience of non-neural living things isn't as rich as those with a brain, perhaps beings w/o a nervous system have a different kind of (simpler, slower or disjoint) experience. The main question with regards to [[Sentient beings that are capable of suffering]] is whether they suffer or not. I'm confident that they don't suffer in the same way as organisms with a brain/central nervous system do, but whether they're devoid of suffering entirely is something I'm not sure. - Since nothing is discontinuous, even [multicellular organisms](https://en.wikipedia.org/wiki/Multicellular_organism#Evolutionary_history) like humans are continuous with single celled eukaryotes, that are themselves continuous with extremely simple single celled prokaryotes (LUCA - last common ancestor) - If we can find intermediate forms between complex multicellular organisms such as humans and simple single cell organisms, the likelihood that consciousness suddenly popped up at some boundary seems arbitrary and misanthropomorphic bias (where we deny agency to non-humans because they don't seem human-like) - Even the transition from non-living things to living things isn't discontinuous. Life evolved from a system of chemicals that somehow came to be good at avoiding dissipation. - There are many scientific theories about the origin of life and all of them show the smooth continuity from a non-living system to the first living thing. - If that is indeed true and we conclude that the first living thing had some primary conscious (because it is continuous with multi-cellular beings such as us who are definitely conscious), we have to extend the ability to experience to (traditionally defined) non-living systems as well. - This brings us fairly square to Panpsychism. As [Chris Fields says](https://www.youtube.com/watch?v=XGulRS2IyF8), there's nothing special about a biological observer. Any physical system that derives information from its environment is a (quantum mechanical) observer and since biological observers are conscious, we have to conclude that all physical systems are conscious (in some sense). This is obviously not suggesting that my thermostat can reflect on its life and can have existential crisis, but this opens the possibility of the thermostat to have a very simple (possibly one-dimensional) experience. #### Reasons to believe that Self exists at multiple levels in biological systems ##### Insights from multicellularity Due to evolutionary pressures to survive and thrive, biological agents are inherently selfish. The selfish nature of genes is described beautifully in Dawkins' book [The Selfish Gene - Wikipedia](https://en.wikipedia.org/wiki/The_Selfish_Gene). But selfishness doesn't just remain with the genes. All biological systems at all levels **have** to be selfish because otherwise they will get dominated by a system that selfishly prioritizes its own survival and reproduction at the expense of a system that doesn't. If all biological agents are selfish, how do we make sense of the co-operation seen in multicellular organisms? The way to resolve this apparent paradox is to assume that cells in a multicellular organism are as selfish as independent single-cell organisms such as bacteria. Because of entropy and randomness in the environment, systems dissipate over time. The only way to survive in dissipating environment (that comprises of other adversarial systems such as predators) is to have an informational advantage over others. If you can out-predict what is out there to kill you, you can survive. This value of information is the basis of co-operation in biological systems. Cells within a multicellular organisms are living in a symbiosis with each other because they share information and resources with other. Over time have become so dependent on each other that they can't live outside of the microenvironment they've created for each other. **So, cells in multicellular organisms selfishly organize into a bigger system (the organism) that _helps_ all of them survive.** Here's how this happens. All living beings acquire information from the outside environment to sense food and danger. They have to actively maintain a state of homeostasis, otherwise they die. For small, independent cells the span of information they can acquire is limited to space in immediate vicinity to itself. Similarly, the capacity for "memory" of such a cell is limited to whatever it can store inside its cell. However, by banding together in a multicellular organism, the scope of information acquisition expands in both space and time. Now, the system of cells - the organism - can acquire information about food and danger from far away and communicate to the entire system. Similarly, with some cells happily taking the exclusive role of memory (which is sort of a "hidden layer" for persisting information over time between sensors and actuators), the system can remember a lot more about the past and also predict far into future. So, by co-operating, a system can expand its cognitive capacity and hence have a better chance of survival for all its constituents. While multicellularity helps the collection acquire more information, it simultaneously helps make the microenvironment for most cells predictable. Unlike a bacterium that can expect to encounter a range of environments (and their corresponding dangers), a cell in the heart tissue can expect to only encounter other heart cells. This surprise minimization helps in maintaining homeostasis because the heart cells then expects a very predictable (but nourishing) environment (that's due to offloading complexity handling to the bigger system that it is part of). At the cellular level, greed exists. It only looks like co-operation at a larger level. (This is sort of like free-markets where entrepreneurs compete for profits but the society benefits from net-new innovations to make life of everyone better than before) Because of expanded cognitive capacity and derived-goals of the larger system, we can truly say that **the sum is larger than the parts**. That is, the larger system is able to solve problems and pursue goals that none of the smaller parts can. **What's striking in this picture is that a multicellular organism - like us, humans - is best interpreted as a collection of selfish agents, each with their own goals.** But because this collection of selfish agents itself is selfish (i.e. as humans we instinctively prioritize our survival and reproduction), we must conclude that this _selfishness_ is going on at multiple scales in the body. In fact, even within a single cell, there are selfish systems such as mitochondria, cytoskeleton compartments and so on, each of which selfishly pursuing its own goal by co-operating to expand its "cognitive light cone" (to borrow a phrase from [Michael Levin](https://www.frontiersin.org/articles/10.3389/fpsyg.2019.02688/full)). This cognitive light cone (or cognitive boundary) is the most distant (in time and space) set of events a system can measure and hence use to influence its goal directed behavior. The cognitive boundary is extremely small for bacteria and its somewhat larger for bacterial biofilms/plants and for humans it is extremely large. A dog doesn't know or care what's happening in the adjacent town but we humans are concerned with the heat death of the universe. But cognitive boundary is a continuum. It doesn't suddenly "pop" out of nowhere at an arbitrary level of complexity. As per Michael Levin: >"Information processing and spatio-temporal integration required to make complex bodies arise from capabilities of single cells, which evolution exapted and scaled up as behavioral repertoires of complex nervous system familiar as Selves" Hence, we can define Self as: >Any goal-seeking system that acquires information and energy from the environment is a Self. > >In fact, a Self is defined by the goals (states of worlds that wouldn't exist without the system) a system pursues. This Self develops because we see what's meaningful to us and what is meaningful to us is reproductive success. In other words, environment is parsed into meaning according to whatever we've found useful in our evolutionary history. **This meaning is the sensory qualia, we can't perceive anything else from the environment** It is interesting to note that as per Buddhism, if you detach yourself from having expectations (about what you prefer and not prefer), your self dissolves. This is exactly what this line of thinking also suggests. ##### Insights from cancer and developmental biology The _agentness_ of different constituents within our bodies is best seen in the case of cancer. Normally, cells keep each other in check by communicating the state of the entire system (organism). But if these communication channels are blocked for any reason, a cell can revert to its single-celled state and starts optimizing for its survival and reproduction. It even seeks environmental niches in body that best supports its proliferation (this is what metastasis is) During development of an organism, different cells _migrate_ to different spatial locations to give rise to the eventual morphology. The independence of these agents can be made apparent through experiments like cutting the embryo into two, which results in two independent organism. If there was a strict genetically encoded plan, this cutting experiment would have disrupted the development. But if we model development as subsystems doing problem-solving, then it is easy to see how these subsystem agents can seek their own goals (i.e. target morphology) independently. The most famous example of this comes from Michael Levin's lab. [In this talk](https://www.youtube.com/watch?v=L43-XE1uwWc), he describe Picasso Frog where if they create conditions for an eye to grow near the leg of a frog, such an eye actually is able to see. Apparently, the optic nerve from this eye in the leg is able to seek the spinal cord and attach itself to it, hence making the frog see through this weirdly located eye. All this suggests that the best way to model a multicellular organism such as human is to see it as an environment that enables simple subsystems to pursue their goals (of survival - a state they prefer over non-survival) effectively. **Since we're pretty sure a human is a Self because it pursues its own goals in the environment it's located in, by extension, all those subsystems that comprise a human are a Self too.** This sense of Self extends to systems at all levels - subcellular, cellular, tissues, organs, organisms. Interestingly, we don't have to stop at an organism. Since organism is collection of subsystems communicating with each other to pursue a common goal, systems that organisms comprise where they pursue common goals can be thought of a self as well. The most notable example of this is an ant colony, which is being [studied for consciousness](https://twitter.com/paraschopra/status/1423584179715858447) and shows various illusions). But human organizations such as govts, families or companies can be thought of a self as well (with slightly different goals than its constituents). #### Insights from use of bioelectricity in cognition There's an interesting parallel between how an organism develops in morphological space and how an organism moves in our regular 3D space. Both processes require problem solving between different units (cells, or organs) and both processes use similar mechanism for communication: cell-to-cell bioelectric signals. This also shows why most mutations aren't fatal. This is because if an agent comprises of proto-cognitive micro-agents, then a mutation (leading to slightly different expectations of what these micro-agents will find around them) from their point of view is just another hostile environment (something they can work around, like most of the life). Hence, it's agents all the way down and up at different levels in the universe. Since the first cell of the embryo is a continuous, unbroken chain from the first cell, it further suggests that the entire process of evolution - the all life forms - constitute a Self in itself, #### What does it feel to be a Pancreas? If a Self exists at all scales (cells, tissues, organs and super-systems), why do we particularly feel our sense of self in the brain? We know that the brain is responsible for generating the sense of self and not pancreas because we can remove our pancreas and still retain the self. But because of continuity arguments, there's no reason why _only_ brain has to have a Self and no other system at similar, smaller or larger scale. If all goal-seeking agents have an experiencing Self and our organs are goal-seeking agents, then all our organs besides the brain should be experiencing a Self as well. Perhaps the particular Self we experience in the brain is differentiated by its capacity from other organs. The brain is rich enough to integrate multiple types of sensory data, represent the environment in an internal world and have enormous bandwidth for memory. Most importantly, perhaps it can communicate to other similar Selves in language. In [split-brain patients](https://en.wikipedia.org/wiki/Split-brain), we can actually sense the existence of two Selves where different hemispheres have slightly different preferences. **Other subsystems of humans - pancreas, heart, liver, individual cells - may have limited form of experiencing Self, independently existing with the Self represented by the brain.** Even the unconscious reflexes (such as flinching away from hot stove or riding a bicycle) may be implemented by a rudimentary kind of system with a simple Self. We can imagine our Self as we experiencing it comprising of any other Selves, each maintaining their own goals which is to co-operate and help each other survive. But why aren't we conscious of everything that is going inside the body? Being (reflectively) conscious and having a continuous awareness like we do requires energy and capacity to hold and retain information for a while (say couple of seconds to couple of minutes). This requires energy and hence evolution ended up ensuring only the **relevant** information gets bubbled up to the the organismic Self. Rest of the actions can be taken by sub-systems which may have a much simpler Self, requiring less energy and capacity to represent. #### How does the question of Self matter? Without getting embroiled into meta-physics, we can consider Self as a metaphor whose usefulness should be judged in two respect: - Scientifically, by whether it leads to new, robust research programs - Does it gives rise to new questions, experiments or explanations? If yes, it's a useful path to pursue - Morally, by whether it helps us being mindful of our interactions with the world - Does it give insights into how to live a life that doesn't unintentionally increase suffering for different sentient systems?