> For the complete documentation index, see [llms.txt](https://lifevisor.gitbook.io/book/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://lifevisor.gitbook.io/book/07-life.md). # 7. Life: Entropy and the Pursuit of Complexity

“Inner Life of the Cell” - Animation developed by Harvard University showing a motor protein carrying a vesicle inside a living cell.

*** *“The total disorder in the universe, as measured by the quantity that physicists call entropy, increases steadily over time. Also, the total order in the universe, as measured by the complexity and permanence of organized structures, also increases steadily over time.”* **— Freeman Dyson, Physicist and Mathematician, in "From Eros to Gaia"** *“An agent does not 'have' a model of its world - it 'is' a model. In other words, the form, structure, and states of our embodied brains do not 'contain' a model of the sensorium - they 'are' that model.”* **― Karl J. Friston, British neuroscientist, in "Behavioral and Brain Sciences" (Journal)** *** ## The Mystery of Life Human beings have been on a quest to explain the mysteries of life for a long time. We have been haunted by questions like: What is life? How is it different from inanimate matter? How did it emerge from inanimate matter? How does it work? And, of course, what is its meaning and purpose? And for a long time, all we could do was talk about it in religious or philosophical or poetic terms. Even today, most people prefer to talk about life in those terms rather than taking a scientific approach. (Except when they are sick or something.) We acquired our longing for magic, as we saw earlier, while playing incredibly fun games like peek-a-boo as children. But when we figured out what was really going on, we couldn't experience that magical feeling anymore. So now we are afraid of going too deep into other things we find magical, like life, the universe and everything. But luckily, not everyone feels the same way. We have had a constant stream of scientists going after the aforementioned questions of life much more methodically and rigorously, via the sciences of biology, chemistry, and even physics. And, we have made some amazing progress there. Robert Hooke’s discovery of the living cell, Charles Darwin’s theory of Evolution, and Francis Crick and James Watson’s discovery of DNA arguably mark three of the most important milestones in this quest. These milestones and many other discoveries have helped us understand the physical organization of life, how and why it has been changing over time, and the underlying code of life that controls a lot of it. And learning about it has been extremely enlightening, not just because we are literally made of this stuff, but also because we are still governed by the same fundamental principles and processes that gave birth to us, even though we may not always realize it. We will look at a few of these principles and processes in this chapter, along with how they help us build the MSE Framework. Note that since we are looking at life through the lens of science and engineering, we will use the more general term “Living Entities” to refer to it. This not only avoids the potential baggage that comes with the heavily loaded word "life", but also generalizes it to other potential forms of life besides biological life, such as organized groups of living organisms or alien lifeforms or even artificial or virtual lifeforms. ## Dissipation and Self-Organization, Again Let us start with where we left off in the last chapter, discussing the Physical Tendencies of the universe. As we saw there, the overall entropy of the universe has been increasing in accordance with the Second Law of Thermodynamics. The flow that drives this, Dissipation, means that most things have a natural tendency towards falling apart. But not always. We can see ubiquitous and ever-present evidence of the opposite movement too. As we saw, the tendency of Self-Organization channels the very flow of Dissipation, through webs of Interaction among differentiated things, into ever more ordered and complex structures. These processes that bring things together continue until eventually we see living cells emerge from complex chemical soups, complex organisms emerge from simpler ones, and self-organized complex societies of organisms emerge from individuals, and an entire biosphere emerges from everything on the planet. While we understand a lot about all of these processes, we still don’t have a full account of the first step: How life got started at all. We have some pretty solid hypotheses, but still no definitive account. Luckily, we don’t need to know that in order to define the MSE Framework, because we follow Present-Bounded Rationality as our methodology. We only care about what is happening with the processes of life right here, right now, to the best extent possible today, while keeping our eyes open for future developments. We have discovered some processes associated with the phenomena of life that are continuously occurring in the present and everywhere, and have formulated some very good theories about how or why that may be occurring. One of them is American physicist Jeremy England’s theory of Dissipation-Driven Adaptation, which could be described as a “Physics Theory of Life”. Let us take a quick look at it. ## A Physics Theory of Life Let us continue with the “ink drop in water” thought experiment that we went over in the last chapter. So, once again, imagine that you place a drop of ink into a glass of water. As we already know, you would expect to see it slowly disperse until the ink reaches an equilibrium with the water, and the entropy of the overall system would be higher than before. But now, imagine that this blob of ink didn’t dissipate into the water. Instead, it expanded a little, then contracted back, and kept doing that, continuing its “blob-ness” for a while. You would immediately feel that there is something very unusual going on. It would almost look like the ink blob was “alive” in some sense. This intuition is perfectly reasonable because that is exactly how Living Entities behave — they may expand and contract or move or change over time, but they don’t completely dissipate into their environment. They maintain their “blob-ness”, at least while they're still alive. Every living cell can be said to be like a blob of some extremely complex chemicals that resists dissipation as dictated by entropy. Another way of saying that is that it keeps itself far from reaching a state of equilibrium with its environment. The way a living cell does this is by absorbing low entropy energy from some energy source in its environment and using that energy to keep itself in a low entropy state. In the process, it radiates high entropy energy into their environment in the form of heat. The technical term for this is "dissipation". Jeremy England's theory of Dissipation-Driven Adaptation generalizes this idea to any complex "chemical soup". It states that in a complex environment with an energy source, complex entities inevitably and automatically emerge to help dissipate that energy more efficiently than it would otherwise be. Not only that, but if the energy source or the environment fluctuates from time to time, these self-organized complex entities will try to adapt to those fluctuations in order to continue to dissipate the energy efficiently. In other words, they will resist reaching an equilibrium with their environment, or alternatively, maintain their far-from-equilibrium state, as long as the source of energy and the complex chemical soup continue to exist. Such self-organized complex entities are known as Complex Adaptive Systems and the process is called Dissipation-Driven Adaptation. Note that such conditions existed on early Earth for extremely long periods of time. There were various bodies of water with complex molecules in them, constantly interacting with each other. There were also energy sources, such as the sun or the hydrothermal vents at the bottom of the ocean. So, these conditions were perfect for the emergence of such self-organizing Complex Adaptive Systems. It is possible that life emerged as a result of some such mechanism. That is why this insight is being called a Physics theory of Life. It potentially explains the emergence of life-like processes directly based on physics. Jeremy England's insight itself is based on two well-established theories in thermodynamics known as the Fluctuation-Dissipation theories of Christopher Jarzynski and Gavin Crooks. In other words, this theory has a strong foundation of well-accepted theories in physics. By the way, note that we are not claiming that this is how life actually got started on Earth. We do not know that for sure yet. We also haven't managed to create new life using this theory. But we don’t need that for our framework. As I have stated earlier, in the MSE Framework, what we care about is not what happened in the distant past, nor what happens in the distant future. What we do care about is that we can demonstrate the process of Dissipation-Driven Adaptation occurring at any time. In other words, irrespective of whether this process actually led to life, the process itself is real. As a result, we can say that the universe appears to have a natural tendency to spontaneously create higher and higher levels of self-organized complexity, as long as we have a complex set of simpler entities interacting with each other, such as those found on planets like the earth, and an energy source, such as stars like our sun or other geothermal sources. Needless to say, this is another and even stronger evidence for the natural tendency of the universe toward Self-Organization that we identified in the last chapter. But wait, there’s more! Karl Friston, a British neuroscientist, has an even more interesting theory of life, based on statistical physics. ## Friston’s Genius Formulation of the Phenomena of Life As mentioned above, Karl Friston is a neuroscientist. So, his original goal was to come up with a fundamental mathematical model of how brains work. Using various techniques in statistical physics and machine learning, he produced a mathematical model that captures the essence of how brains function. This model has been incredibly successful at explaining many aspects of brain function. He calls it the Free Energy Principle. But he didn’t stop there. He discovered that this principle could easily be applied to explain the behavior of all Living Entities in general, starting from a single cell to a large living organism or organized groups of such organisms. We will not go into the math here, because it will take an entire book of its own to explain. But we will try to understand the intuition behind the principle, as applied to Living Entities. Let us start by taking a look at the following diagram, which shows a simplified version of the principle. It shows a Living Entity (big box on the right) interacting with its environment (the picture of the earth on the left).
Let us start by observing that every Living Entity we come across in the world is enclosed inside a semi-permeable boundary or “skin” of some sort. Mathematicians have a name for this boundary— it is known as a Markov Blanket. And, just like the name suggests, it separates and protects the contents of the Living Entity from its environment. For example, the cell wall of any single living cell is such a Markov Blanket. The skin that surrounds our whole body is also such a blanket. If one thinks of a country as a Living Entity, then its border can also be thought of as such a blanket. It is easy to see that the Markov Blanket plays a crucial role in preventing the contents inside from being dissipated into its surroundings. But it goes well beyond simply protecting the contents inside. It also has the capability of allowing chemical compounds and electrical signals in, which can be called perception. And it can also act on its environment, again in the form of chemicals or electrical signals or even physical movement. Let us go over various parts of the diagram above to understand what is going on in a little more detail. We will need to do that in order to understand the essence of the most critical aspects of life. Each of the sections below corresponds to one of the terms in the diagram above. ## Internal States The first thing to realize is that any living entity can only survive in a limited range of states. These are the states that are conducive to its continued existence. If the system goes outside this range, it is in danger. For example, your body has certain states, consisting of its temperature, blood pressure, blood sugar level, and so on. Your body constantly tries to keep the values of these states within a certain range, failing which, you do not function properly. Too high or too low a value of your body temperature or blood pressure or sugar level, and you start having problems. These states are called “Internal States”, i.e. states internal to the Living Entity, to distinguish them from the states of the environment, which are external to the Living Entity. Basically, what we are saying is that, in order for a Living Entity to sustain itself, it needs to maintain its Internal States within a limited range of values that are conducive to its survival. ## Hidden States Analogously, the environment also has a large number of states. For a simple system like an amoeba swimming in water, the environmental states may be characterized by simple parameters like water temperature, pressure, salinity, acidity, distance from a predator and so on. And since these states are external to the Living Entity and thus unknown to it, we call them “Hidden States”. ## Perception and Action Now, the environment keeps changing constantly, which is to say that the Hidden States keep changing constantly. For example, in the case of the amoeba, the water temperature, salinity, distance from predators etc. keep changing. Any such change to a Hidden State could impact the Living Entity. There is always the danger that some of these changes will force the Living Entity to go outside the safe range of its own Internal States, potentially causing it to not function properly or even die. It goes without saying that, in order to sustain itself, the Living Entity needs to be able to: 1. Perceive the changes to the environmental or Hidden States, (for example, the water temperature on its right is becoming too cold) 2. Analyze them to figure out how to respond (for example, determine that it needs to move where it is warmer and realize that means moving to the left, where it is warmer), and then 3. Take appropriate actions (for example, actually move to the left). As hinted at earlier, the Markov Blankets of Living Entities are capable of performing functions 1) and 3) above. These are depicted in the diagram above with the labels Perception and Action, respectively. The second part, analyzing the input and deciding how to respond, is accomplished by the Living Entity creating and maintaining an internal “Model” of the environment, as we will see below. This Model is built by the Living Entity, based on its learnings from its previous experiences of interacting with the environment: What changes occurred in the environment, what actions it took, and what happened as a result. If the results were good, the relationship between those environmental changes and the decisions the Model came up with would be reinforced. If not, they get weakened. Over time, the model gets better and better at making the right decisions. (Of course, most Living Entities have ancestors, and a lot of this learning gets passed down from generation to generation via genes. But we don’t need to get into that detail yet. We will get to it eventually. For now, let us just say that it learns this by itself.) Here is an example to illustrate this further. Let us say you are looking at your social media feed and you come across something outrageous or triggering. Your eyes perceive it and notify your nervous system. Luckily for you, you have experienced this before. Your internal model has learned, from your past actions, that clicking on that item will simply lead to your feed getting inundated with more and more outrageous and triggering items. You will just get more and more triggered and outraged. Not only will you waste your time, but you will ruin your whole day. So you stop yourself from clicking on that item. Instead, your internal model tells you to switch to the far more rewarding activity of reading books about engineers finding meaning in their lives. \[ Good for you! You have proved that you are an excellent Living Entity with a great future! ] ## Generative Model and Active Inference As we have mentioned above, when a change in an environmental Hidden State is perceived by a Living Entity, it relies upon an Internal Model to decide how to respond to it. But it can do even better. The Internal Model isn’t just reactive, it can also be proactive. So it doesn't just respond to changes in the environment, it can even get ahead of them by imagining things that could happen and being ready for them. Think about what the Internal Model needs to do to accomplish this. The Model needs to be able to make reasonable predictions about what the environment might do next, and prepare a set of responses to deal with them. But, in order to figure out what responses might be appropriate, it has to be able to actually experience those conditions in the environment and respond to them to see what happens. This means it has to actively seek such conditions out. In order to accomplish this, the Living Entity needs to do two things: 1. Its Internal Model needs to be “generative”, i.e., capable of generating possible futures about what the environment might do next. These are known as “counterfactuals”. And 2. It has to take appropriate actions to seek these counterfactual conditions out in the environment. This process is called Active Inference. Active because it is not just passively inferring the Hidden States of the environment, but actively imagining what they could be and then actively seeking evidence for their existence. ## Surprise! Note that the ultimate goal of this exercise of Active Inference is to avoid surprises! The recent Covid-19 pandemic provides an excellent case study for this. The global health system can be thought of as a Living Entity with a generative model for world health, with the WHO and affiliated organizations being its home. Such a model can be expected to be able to predict, based on its past experiences, as well as its generative ability, that pandemics of various types may occur at any time. And it needs to be ready for them with appropriate actions such as early warning entities, isolation protocols, vaccine research, educational campaigns and policy prescriptions. A lot of this planning and strategizing involves imagining scenarios that could unfold (including ones that may have never occurred in the past) and preparing for them. Of course, the scenarios they generate cannot be completely random. They need to be within the realm of possibilities given their knowledge of how the world works. Typically, organizations such as the WHO would come up with realistic scenarios, conduct simulations, mock exercises, strategic and tactical planning exercises and so on to actively infer what types of responses would work in those situations. The ultimate aim of this exercise is to ensure that if another pandemic were to break out in reality, even if it is of a different character, it would not be a surprise to the global health system, and it would be able to respond immediately and effectively. ## Minimizing Surprise As mentioned earlier, the Free Energy Principle involves a lot of complicated math. The end result of that math is a formula that calculates the amount of surprise that a Living Entity might experience when its environment changes in some way. The formula further states that this surprise needs to be minimized in order for the Living Entity to continue to exist. It should be intuitive that minimizing unpleasant surprises should result in the Living Entity being able to respond to the change in the environment effectively, keeping itself intact. In mathematical terms, minimizing surprise is also called “minimizing the free energy” (hence the name, Free Energy Principle). (Note that the “free energy” mentioned here is not related to the concept of energy from physics. This is a separate concept from statistics.) Another mathematical term for the same process is “maximizing Bayesian Model Evidence”. Essentially, the intuition here is that the Living Entity wants to collect evidence that its Internal Model is robust against changes to the environment. Once again, it goes without saying that the better the evidence for the model, the better chance the Living Entity has of continuing to exist. The main takeaway here is that when you update your internal model to minimize surprise, or improve its “Model Evidence”, you are basically becoming better at predicting what the environment might do next and responding effectively to it. ## Birthday Surprise Here is another interesting example to illustrate the idea better. It is a common practice to surprise people on their birthdays. And very often, the person having the birthday ends up suspecting that something is up and isn't really surprised. How does that happen? All of us have models of our family members’ or friends’ behaviors inside our minds, based on their past behavior. We can also imagine the ways they might behave if they were trying to hide something from you. So, when they start acting in a weird manner, we notice that it is similar to how we had imagined they would behave if they were hiding something from us. Based on all of that, we make a prediction that there may be some kind of a surprise coming our way, though we may not know all the details. Then, we may start asking probing questions and see how they respond to them. Or look for evidence around the house of unusual activity. This gives us more clues and reduces our uncertainty further. All of this is another way of saying that we instinctively use our Perception, Internal Generative Model and Actions to perform Active Inference and minimize our surprise! Ok, that pretty much sums up the essence of the Free Energy Principle without going any deeper into the math. ## Modeling The main takeaway from all the discussion above is that every Living Entity, in order to persist, must build and maintain an internal model of its environment. It does this by perceiving its surroundings and continually updating its model accordingly. And the more complex Living Entities go a step further. They also build a model of themselves, which sets the stage for them becoming self-aware. Together, this modeling is the basis of all learning, all prediction, and, as we will see later, all meaning-making. Best of all, this happens all on its own. Every Living Entity, from a single cell to a complex living organism to an organized group of organisms, is doing this automatically, all the time. So it looks like we have discovered another natural tendency of the universe, the first one that is specific to living systems: the tendency of living systems to build internal models of themselves and their world. This is the first of what we will call the Biological Tendencies, which build upon the four Physical Tendencies we collected in the previous chapter. *** #### ***Universal Tendency #5: Modeling*** *All living systems build and continuously refine internal models, or generative models, of themselves and their world. These models allow them to perceive, predict, plan, and navigate their environment in order to continue to exist.* *** Having looked at the two main ideas connecting life and physics that have been developed fairly recently, let us now turn our attention to another important aspect of life, namely, intelligence. ## Intelligence It is interesting to note that Friston wasn't looking for a principle that governs living entities. He was looking for the way a brain functions. In other words, he ended up discovering a fundamental principle of life while looking for a fundamental principle of intelligence. This was probably not an accident. It is common to think of the phenomena of life as separate from that of intelligence. Many people think that primitive life, like bacteria and algae etc. are unintelligent and one has to get to larger, more complex creatures before we can call them intelligent. But, as we just saw, every living cell necessarily has to have some amount of intelligence just to be able to maintain itself in a dynamic environment that might surprise it. It has to have the ability to perceive its environment, create and maintain a generative model of the environment based on past experiences, generate counterfactuals, look for evidence for them, use the model to make decisions, and then act on those decisions. All of these activities look very much like intelligence. For example, the Merriam-Webster dictionary defines it as “the ability to learn or understand or to deal with new or trying situations.” Google defines it as “the ability to acquire and apply knowledge and skills.” Even the simplest bacteria can be said to have such capabilities, albeit at a very basic level appropriate for their limited range of activities. In fact, when you look at all the things that happen inside a living cell, one could argue that the most intelligent entity in the universe, per unit of volume, is a living cell! The human brain may be far more intelligent overall, but it pales in comparison to the complexity and capability that a living cell packs in such a small volume. All you have to do is watch some of the living cell simulations (or even videos of actual living cells under a microscope) to see the incredibly complex and varied activities occurring inside living cells. These include various molecules being ferried around in very purposeful and intelligent ways, complex structures being built or dismantled, cell walls allowing or rejecting various chemicals from entering the cell and so on. Watching a movie of a cell dividing into two, including how the DNA gets duplicated, is simply awe-inspiring. White blood cells hunting down an intruder is nothing short of a nail-biting drama. All of this occurs at the cellular level like clockwork, in an environment full of chaos. Way more chaos than most of us experience in our daily lives! It is simply amazing to realize that all of that is going on inside every cell in your body, right here, right now! (As an aside, anyone who thinks that science kills magic, or awe or even gratitude from our lives, should really watch such videos and then close their eyes and imagine all those things occurring inside their own body billions of times. They will instantly recover all of what they may have felt they had lost and more.) I see no reason to think of all of these activities occurring inside living cells as unintelligent. Of course, this isn’t abstract or theoretical or artistic or other forms of intelligence that we typically associate with human brains. A living cell isn’t going to beat you in IQ or chess or music composition. But it can absolutely beat you in terms of the amount of complexity and variety it handles in such a small volume. Now, you might think that I am about to add another item to our growing list of universal tendencies. But I decided against doing that here, because, as we just saw, intelligence and life seem to be inseparable, and we have already captured the essence of it in the Modeling tendency above. But no worries. This universe we have ended up with is so awesome that we still have quite a few more tendencies to discover and appreciate. Before we get to the next of them, though, let us reflect on one idea that has been quietly at work behind everything we have discussed so far. ## Emergence While describing various phenomena in the chapters on Physical Reality as well as Life, we have come across quite a few cases where larger entities, patterns, or properties can be seen to arise through complex interactions among a large number of smaller or simpler entities, even though these outcomes are not apparent from the individual components alone. For example, the phenomenon of the temperature or pressure of a gas emerges from the complex interaction of a large number of its molecules. Or the phase of matter, like liquid or solid, emerges from the arrangement and interactions of atoms. Or Complex Adaptive Systems, and eventually, Life, emerge from complex interactions of a large variety of molecules. Or murmurations of birds emerge from a flock of birds flying together or traffic patterns emerge from a large number of cars. Or how galaxies emerge from a large collection of stars and other matter in space. In each of these cases, it is extremely difficult or even seemingly impossible to determine how some of the properties exhibited by the large-scale or emergent phenomenon could be derived from its components. In fact, often it is the case that we can model the emergent phenomena quite well, and we can model the constituent parts quite well, but we can't explain how the former emerges from the latter. This phenomenon of emergence is so common, occurring in so many contexts and scales, that it is worth naming explicitly. But we won’t add it as a separate tendency, because it is a meta-tendency in the sense that is shows through various other tendencies at various levels and domains. It is one of the aspects of Differentiation as well as Self-Organization. Emergence is simply what these tendencies look like from the point of view of the new properties that appear at each level, properties that the lower level alone could never have predicted. With that idea in hand, let us return to the Biological Tendencies and pick up where Modeling left off. ## Exploration A model is only as good as the information that goes into it. So living systems don’t just sit back and wait for the world to come to them. They actively go out and probe it. This is the deeper meaning of Active Inference that we encountered earlier. It extends beyond passive perception to the active seeking of new information, in order to reduce uncertainty about the environment, as well as about the entity’s own model of itself. Living systems behave as if they are driven to close the gap between the predictions made by their internal models and what reality actually does. In conscious beings, we recognize this tendency by a more familiar name: curiosity. A Living Entity going out and seeking evidence for the counterfactual scenarios its generative model has dreamed up is, quite simply, being curious about its world. And in sufficiently complex entities, this curiosity can also turn inward, toward modeling and exploring their own behavior, so as not to be surprised by themselves. Either way, exploration, both outer and inner, is a biological imperative built into the very architecture of life itself. So let us add it to our list. *** #### ***Universal Tendency #6: Exploration*** *Living systems actively probe the world, and themselves, seeking out new information and experiences in order to improve their models and reduce uncertainty. What we typically call curiosity is this tendency expressed in conscious beings.* *** ## Generativity Exploration improves a living system’s model of the world. But living systems do something even more remarkable. They generate new counterfactuals and act on their basis to change the world itself. Extending Active Inference one step further, living systems don’t merely adapt to their environment, they alter it. They generate novelty and bring genuinely new forms into being. This can be seen as self-organization that the system itself is actively driving, rather than just passively undergoing. We see it everywhere in nature. Birds build nests, bees build hives, and beavers build dams. All of these can be seen as innately creative acts that those creatures have learned through evolution. And evolution itself is the grandest example of all. Recall that not every behavior of a Living Entity is learned within its own lifetime. A lot of it gets passed down from generation to generation, through small variations in the information passed on, which we call mutations. One can think of a mutation as yet another way for the generative model of a Living Entity to try out a counterfactual: it brings a slightly altered offspring into existence, which then seeks evidence for the viability of that change. If the evidence is good, the change persists; if not, it dies out. This is Evolution by Natural Selection, and it can be seen as Active Inference playing out across generations, with the DNA, along with replication and mutation, serving as the generative model. Everything you see in the biosphere, from the simplest bacteria and algae to complex creatures such as ourselves, as well as the self-organized packs and societies of social organisms, was generated by this relentlessly creative process. It is no accident that another word for the entire biosphere, or even the whole universe, is Creation. And it is this same process that has filled the universe with its mind-boggling diversity of forms. In conscious beings, this tendency reaches its most self-aware expression. Human creativity, including all of art and engineering, is generativity that knows itself to be generative. So let us add it to our list. *** #### ***Universal Tendency #7: Generativity*** *Living systems act on their environment to alter it, not merely adapting to it but actively generating novelty and bringing new forms into being. What we typically call creativity is this tendency at work, and evolution is its grandest expression.* *** ## Coupling So far, the Biological Tendencies we have described, Modeling, Exploration, and Generativity, can all be carried out by a single Living Entity on its own. But living systems rarely stay solitary. Conscious organisms model not just their physical environment, but each other. And, crucially, they deliberately bind to one another, as well as to their environment. Recall the Physical Tendency of Interaction from the previous chapter, where differentiated things affect each other physically. Coupling is what happens when living systems recognize those connections and choose to act on them. A parent’s bond with a child, a community’s care for its members or for its land, these are cases where interaction becomes personal, felt, and reciprocal. This is how individual Living Entities self-organize into families, communities, organizations, ecosystems, and ultimately the entire biosphere. In each case, the boundaries simply widen, and the Free Energy Principle and Active Inference continue to apply, now at the level of the group. A company, for instance, brings needed people in and lets others go, models its customers, suppliers and competitors, projects their future behavior, and acts to avoid nasty surprises, exactly as a single cell does. Such groups evolve too, in social systems, culture, and economics, which are forever changing. So let us add the final tendency to our list. *** #### ***Universal Tendency #8: Coupling*** *Conscious living systems model one another and deliberately bind together, into relationships, families, and communities, weaving webs of mutual recognition and shared meaning. Where Interaction is physical, Coupling is interaction that has become personal, felt, and reciprocal.* *** ## The Need for a New Law of Nature Needless to say, many scientists have noted the prevalence of these Universal Tendencies we have been collecting: Differentiation, Interaction, Dissipation, Self-Organization, Modeling, Exploration, Generativity and Coupling. While we do have the law of evolution in biology, what we really need is a deeper and broader law of nature that encompasses all of these processes in all of these contexts, starting from quantum fog to galaxies. Even the Free Energy Principle, which appears to provide great explanations of the phenomena of life and intelligence, can't really be called a "law of nature". (In fact, that's why it is named a "principle" rather than a "law".) Much more work must be done to make these concepts rock solid from the scientific point of view. We can still work within this limitation because our Present-Bounded Rationality methodology allows us to make progress as long as we are relying on the best possible explanations available to us as of now, and we keep an open mind to modify our beliefs as we learn more. That's exactly what we are doing here. Luckily for us, some scientists have started putting forth proposals to formalize a new “law of increasing functional information.” I am hopeful that some of these efforts will bear fruit and that this law will eventually become as well-known as the law of entropy. The following is an excerpt from the paper that proposes such a law. As you can see, it hints at all the Universal Tendencies we have covered so far: > *“Physical laws—such as the laws of motion, gravity, electromagnetism, and thermodynamics—codify the general behavior of varied macroscopic natural systems across space and time. We propose that an additional, hitherto-unarticulated law is required to characterize familiar macroscopic phenomena of our complex, evolving universe. An important feature of the classical laws of physics is the conceptual equivalence of specific characteristics shared by an extensive, seemingly diverse body of natural phenomena. Identifying potential equivalencies among disparate phenomena—for example, falling apples and orbiting moons or hot objects and compressed springs—has been instrumental in advancing the scientific understanding of our world through the articulation of laws of nature. A pervasive wonder of the natural world is the evolution of varied systems, including stars, minerals, atmospheres, and life. These evolving systems appear to be conceptually equivalent in that they display three notable attributes: 1) They form from numerous components that have the potential to adopt combinatorially vast numbers of different configurations; 2) processes exist that generate numerous different configurations; and 3) configurations are preferentially selected based on function. We identify universal concepts of selection—static persistence, dynamic persistence, and novelty generation—that underpin function and drive systems to evolve through the exchange of information between the environment and the system. Accordingly, we propose a “law of increasing functional information”: The functional information of a system will increase (i.e., the system will evolve) if many different configurations of the system undergo selection for one or more functions.”* > > **― “On the roles of function and selection in evolving systems” | Michael L. Wong et al | PNAS Vol. 120 | No. 43 October 24, 2023** ## Mindfulness This brings us back to one of the most central concepts in this book. A significant part of the behavior displayed by all Living Entities, as described in this chapter, which, as we saw, can be derived directly from physics, could actually be described as “mindfulness”. Allow me to explain why. Time to expand our minds. Not just our minds, but even the definition of what we mean by a mind. Let us start by saying that the concept of "mind" includes intelligent, agentic, curious and creative behavior displayed by any complex system, appropriate for its level of complexity. If we define it this way, it can be seen to apply to all Living Entities that we have described: living cells, organisms, organized communities of organisms, ecosystems and so on. Mindfulness, at the human level, is a practice that helps us clear our minds of thoughts that distract us from paying our full attention to what is going on around and inside ourselves. Through various techniques, the practice teaches us how to calm our minds, clear it of all the stuff that occupies it most of the time, and focus our attention on the present moment and situation. This allows our perception to become sharper and more wide-ranging, which helps our internal models to become better at modeling our true reality, and, as a result, makes our actions more effective. Well, as we have already seen, that’s nothing but a central aspect of Active Inference! That's exactly what every living organism is inherently doing all the time! Every living cell keeps itself whole and functioning in the highly dynamic physical world by sensing its environment as accurately as it can, updating its internal model appropriately, and acting on the environment in accordance as effectively as it can. And, over time, either via the process of its own learning or genetics or evolution, this inherent functioning of the entire phenomenon of life gets better and better. This is exactly what the practice of Mindfulness tries to teach - improving our perception of what is happening around as well as inside us, processing it without getting biased or distracted, thus improving its accuracy, and responding to it as effectively as we can. Since the simpler organisms don’t have complex minds such as ours, they don’t really need to learn any practices like Mindfulness to quiet their minds. Their "minds", such as they are, are already pretty quiet! They are mostly just performing Active Inference without any distraction generated by their minds, and as a result, one can say that they are inherently practicing mindfulness all the time. This can also be characterized as tacit or embodied mindfulness. So while we have to learn this practice, they are just born with it. It is interesting that we had to wait for human thought to evolve for millennia before we rediscovered something that was an integral aspect of all living organisms! And now that we have, it is only reasonable that we embrace it like we would a long-lost friend. I hope this also clarifies why, in the chapter on Methodology, we decided to name our methodology Present-Bounded Rationality. Qualifying rationality in this manner not only makes sense based on the reasoning provided in that chapter, but it can even be derived from First Principles. (In fact, I had originally named the methodology Mindful Bounded Rationality, but then I felt that Mindfulness was too loaded a concept to use at such a fundamental level, so I went with the more fundamental idea of the present.) One can even argue that Present-Bounded Rationality is the inherent methodology of all Living Entities. Mindfulness forms a central idea in this book and keeps popping up in multiple places, not because it is popular these days, but because it can be scientifically proven to be a critical aspect inherent to all Living Entities. Still, I think mindfulness expresses only a part of what Active Inference involves. There is another part that also needs to be discussed. ## Prioritizing Counterfactuals As we saw, the process of Active Inference involves the internal generative model generating counterfactual scenarios and then looking for evidence of their existence in reality. Now, there is no limit to how many such counterfactual scenarios the model can generate, so the process of Active Inference has to select which of these scenarios are good candidates for evaluation through action. In the case of highly complex Living Entities such as ourselves, this is where higher level concepts like reason as well as values come in. Reason is the faculty Living Entities use to rank the scenarios in the order of their likelihood of being validated. And "values" is the faculty it can rely on when reason isn't able to provide sufficiently clear guidance (possibly because there isn't sufficient evidence to do so). Values can be thought of as a priori weights that complex organisms associate with various choices so that the choices can still be ranked even when reason isn't able to do so. Over the millennia, various human societies have come up with various sets of values that help people make such decisions in the absence of sufficient evidence. But, as we have seen, very few of these values come with sufficiently rigorous justification for their validity. So, any rigorous framework such as the MSE Framework must provide clear guidance for which concepts can be considered to be "valuable" in this sense. And, in order to meet our criteria, it must also provide a sufficiently rigorous justification for suggesting those concepts. Also note that while the counterfactual scenarios occur in the future, and in some cases, even the distant future, the decision to evaluate them and act on them occurs in the present. Let us keep these thoughts in our mind for now. Just like mindfulness, "values" will become another pillar of our framework as we progress through the book. *** Of course, invoking things like "mindfulness" and "values" before we have even delved into the mind itself may appear a little premature. So let us do that in the next chapter. The mind, or more specifically, Consciousness, is the next logical layer in our framework.