In this essay I hope to convince you that what you really want—what you have always wanted, without realizing it—is something you have probably never heard of. You think that you want happiness, health, security, the esteem of others, etc. In your generous moods you want these things for other people. You think that you want a fair and just society. Freedom, knowledge, biodiversity, sustainability. I will argue that these are all shadows of a deeper phenomenon. What you actually want is adaptive complexity.
Convincing you is only part of my purpose. I also want to defend a more general claim concerning what is normatively valuable. I suspect that there is something universal about valuing adaptive complexity, i.e., that any sufficiently sentient creatures, not only human beings, should be expected to value it also.
The plan for the paper is as follows. Section 2 is a primer on complex systems generally and the concept of adaptive complexity more specifically. Section 3 examines the most common and obvious answer to what we think we want, namely happiness. I’ll discuss the subtleties of this concept and the broader idea of well-being. Section 4 challenges the idea that the well-being of conscious creatures is the ultimate basis for subjective value. Section 5 is the crux of the argument, in which I defend the claim that all of the items I listed above are examples of adaptive complexity. Finally, Section 6 presents an evolutionary psychological hypothesis about how and why this ostensibly universal preference for adaptive complexity might have arisen.
A complex system is a system whose collective behavior cannot be precisely predicted in the long term but which nonetheless reveals emergent patterns [see references 1, 2, 3, 4 below]. Many complex systems are collections of similar components that causally interact with each other locally and non-linearly through feed-forward and feedback pathways. In some systems, the components of the system are “agents,” i.e., entities that sense, think, and act. Agents could be individual organisms, human beings, or even group agents such as companies, teams, or governments. In other systems, the components might be cells or even much simpler entities such as the particles of a fluid. In general, the patterned collective behavior of a complex system is difficult to attribute to a single source. When we examine the components or agents one at a time, there tends to be little or no indication that they will do anything interesting when assembled into an interacting group. Thus the patterns are “emergent properties.”
Canonical examples of complex systems include social insect colonies and flocks of birds, both of which exhibit behavioral patterns that arise as emergent properties of the group rather than from deliberate plans carried out by each individual or coordinated by a boss. No individual ant, for instance, intends to discover and publicize the shortest route between the nest and a food source, but the colony does just that, collectively. No individual bird has a plan to organize its flock into a particular shape. In the human realm, market economies are complex systems and reveal collective properties, such as bubbles and crashes, that no individual participant intended.
The core term ‘complex’ means roughly ‘between ordered and disordered.’ This can be further clarified in terms of probability and information theory, thereby providing an abstract mathematical foundation that makes complex systems theory useful for a wide range of natural and social phenomena.1 Note that in this special use, the term ‘complex’ is not synonymous with ‘complicated.’ Jet engines are complicated, but they are engineered specifically not to display emergent properties. Their creators designed them for a particular task, and that intentional arrangement of parts makes them a paradigmatic example of a system whose patterned behavior arises because of a deliberate, centrally-coordinated plan.2
Human beings are innately interested in various types of complexity. Consider for example your level of interest in a cultivated field vs. a wild forest. The field is simpler in terms of its composition (it has fewer species), its spatial arrangement, and its temporal dynamics. Because of these simplicities, the field is easier to predict and to manage, but the forest is more interesting. At first glance, one might think that this is because the field’s predictability makes it boring, whereas the forest is unpredictable and therefore interesting. But it’s more than that: random number generators and the sound produced by a monkey banging on a piano are unpredictable but uninteresting. Our interest is captured by systems that are not entirely ordered and predictable, nor entirely disordered and unpredictable, but rather, somewhere in between.
One important feature of complex systems, inherited from the broader category of dynamical systems, is the concept of an ‘attractor.’ An attractor is a state or a set of states that a system settles into over time. Take the simple dynamical system of a ball rolling in a bowl. For some range of initial velocities, the ball will ultimately settle in the bottom. Because that stationary position is a single point, it is an example of a ‘point attractor’ for that system. And that attractor has a ‘basin of attraction’: some set of initial conditions that settle into it. If the ball’s initial velocity is large (meaning outside the basin of attraction) it might shoot past the rim of the bowl and fail to settle into that attractor. Another example is an asteroid captured by the gravity of a star. The asteroid might fall into the star, making that another point attractor. Or, like a fast ball, it might escape. But if the asteroid’s initial velocity is just right, it could settle into an orbit around the star, which would be an example of a ‘limit cycle’ attractor—not a single state, but a repeating sequence of states.
In addition to point and limit cycle attractors, complex systems also exhibit a third kind of behavior, one that looks mostly but not entirely periodic. In describing this, Wolfram [4] emphasizes the spontaneous appearance of “local structures”: small patches of order in a broader sea of either uniformity or randomness. In John Conway’s famous cellular automaton The Game of Life, entities like “gliders” and “blinkers” are examples of local structures. Both emerge spontaneously from many initial conditions and propagate through the system. When they encounter each other and other structures, they often trigger a cascade of events, sometimes creating larger patterns that are more complex and more delicate. As a rough preview of why this behavior is relevant to the topic of this paper, consider what can happen when groups of people meet. Sometimes, the interaction will be merely transient: nothing durable or interesting emerges from the crossing of paths. Other times—e.g. in great historical collaborations or conflicts—something about the interplay creates an explosion of new phenomena. Ideas, creative works, or furious contests spring forth and spread through the corresponding cultures, altering the people who encounter them and in turn being altered by them.
The last concept I need for the present work is the obvious one, the one I started with: adaptive complexity. The literature includes two versions of this idea. One version refers to a system in which individual agents learn from their experience in a way that improves their performance on a task [7]. For example, consider a group of human beings engaged in some project, perhaps a sports team or a company. As it encounters challenges, the group displays emergent properties, and these properties change as the individuals in the group learn. Clearly, this version only applies to systems composed of agents. The system is adaptive because the individual agents learn. But I am more interested in the second version of adaptive complexity. In this version, the system as a whole—which may or may not be composed of agents—is homeostatic in a particular way. Namely, it exhibits emergent properties that facilitate its continued existence as a complex system. When the system’s behavior drifts too far toward either a highly ordered regime (like the cultivated field) or a highly random one (like the monkey at the piano), some endogenous force, itself an emergent property of the system, pushes the system back toward the middle [8, 1]. The system’s complexity is itself an attractor. Unlike the first version of this concept, here, adaptation is a property of the system but not its components. To anthropomorphize terribly, it is as if the system as a whole “wants” to avoid becoming frozen in a state of perfect predictability or sublimated in randomness. In order to be a complex adaptive system, this homeostatic force must be endogenous: its source cannot be an external observer who intervenes when the system is becoming too predictable or too chaotic. There certainly are examples of the latter type, but because the corrective force is exogenous, the system itself does not possess a homeostatic force and should not be categorized as adaptive.
So far I have said nothing about how adaptive complexity relates to the list of things that I said you thought you wanted: happiness, health, security, etc. I want to begin making that case now by exploring the connection between adaptive complexity and happiness. Among all the items on the list, happiness is the most obvious and most common answer to the question “What do you think you want?”
For humans at least, happiness resists straightforward definition. It’s not just pleasure, for instance, because depending on how pleasure is arrived at, it may or may not produce happiness. The pleasure that arises from the successful conclusion to hard work is associated with happiness, but the pleasure that arises from drugs or delusions is not. Another subtlety is that most of us do not want a life composed entirely of happiness. We regard people who seem to be that way as having something wrong with them and as living colorless and superficial lives. The rewards of hard work only grow from frustrations that we have overcome; the comforts of companionship are only apparent if we’ve also felt loneliness. So rather than an uninterrupted flow of pleasure, happiness seems to require something richer and more inclusive, a path through emotion and experience that is “full” in some way. Some amount of short-term suffering is consistent with that, perhaps even necessary.
To address these subtleties and avoid the obvious confusion that would be produced by saying that happiness requires unhappiness, some writers prefer to use the concept well-being. Whereas pleasure seems to concern only how we feel in a particular moment, and happiness seems not to capture the diversity of subjective states that we deem necessary for a good life, well-being seems better able to capture a more holistic picture in which, as I suggested above, some amount of unpleasant experience is not only acceptable, but desirable. What we really want is not to feel a certain way in each moment, but for our lives, considered more fully, to have a certain property, some kind of flourishing, worth, eudaemonia.
For non-human animals, well-being may be more straightforward than it is for us, but it is often still more complex and more holistic than a continuous hedonic drip. Zookeepers and pet owners, for example, have long recognized the need to provide companionship and psychological enrichment activities for members of many mammal and bird species who, if deprived of them, suffer from boredom or, if exposed to too much variability, stress. Such animals appear to have analogs for depression, neurosis, and perhaps even suicidality [9, 10]. So even if we cannot at this moment fully articulate everything that well-being does consist of, we can see at least that it appears to be more than happiness, which itself is more than pleasure.
It’s worth noting, too, that most people (e.g., not psychopaths or sadists) want well-being not just for themselves but also for others. Despite prioritizing our own well-being above that of others—with notable exceptions driven by kin selection and empathy—we still assign some value to the well-being of other people and other conscious creatures. There are important and complicated questions about how we want that well-being distributed: would you be willing to sacrifice one point of well-being so that someone else could gain two points? Also there are questions about how to regard the well being of a group: is it just the sum of the members’ individual ratings, or is a fair distribution required? And what does “fair” mean? We must also consider zero sum games: How much does the well-being of a carnivore matter, vis à vis that of its prey? This is not the place for a full treatment of these subtleties, but none of them undercut the central notion that much of what we want appears to be related to the well-being of conscious creatures.
So far, so good. But is the well-being of conscious creatures the only thing we want? Consider plants. Do you care about them? When drought or fire destroys a forest, do you feel that something has been lost? Plants don’t consciously experience well-being, so if you do value them, it has to be for some other reason. Perhaps it is because you regret the loss of habitat and the possible impact on the well-being of conscious creatures. Perhaps, that is, you value plants but only because they have instrumental value. This seems unlikely. A thousand-year-old Bristlecone pine clinging to life on a crag doesn’t provide habitat for conscious creatures, but you still want it to survive. If we discovered a plant-only alien world, we would be careful with it. All this suggests that some non-conscious phenomena strike us as having intrinsic value. The well-being of conscious creatures is not all that we want.
Motivated by thoughts like this, some writers [11, 12] suggest that we broaden our idea of value to encompass all life, whether conscious or not. In that case, human and animal well-being would still be valuable, but so would the life of trees and algae. So would the life of organisms that we may find unrelatable or unappealing, such as fungi and parasites. This doesn’t necessarily imply that we should imbue all living things with equal value, but it does imply that we assign them some value.
The well-being of living things, in other words, seems to encompass many of the values we’ve discussed so far. But the path we’ve taken to this point might leave you with a lingering sense that there might be something else, some other basis for value. In pursuit of that possibility, consider John Elkington’s sustainability-related concept of the “triple bottom line” for a company [13]. For a traditional capitalist, the central purpose of a corporation is to generate financial returns for its shareholders. Its ability to do that is its value, nothing more or less. But Elkington notes the rising value that many people place on a company’s social and environmental impact. For advocates of this view, a company’s value is not a unitary thing, but rather, a triad of people, planet and prosperity. Each is valuable in its own way and irreducible to the others.3
If you find that view persuasive and extend it beyond corporate value, it would seem that, like Elkington, you endorse what philosophers call values pluralism. That is, you value a heterogeneous list of mutually irreducible things. Life could be one such thing, valuable for its own sake. As we saw above, even if the life in question proved incapable of consciously experiencing well-being, its mere existence would still be valuable, and your hope would be for a world with more life rather than less. If some of that life could experience conscious well-being, then that would be a second kind of value. So what you really value is a world with as much life and also as much conscious well-being as possible. You might say that well-being depends on life but that the two are not identical: living things are valuable for reasons that have nothing to do with well-being. You might even entertain the possibility that a non-biological entity, such as a digital artificial intelligence, could experience conscious well-being despite not being biologically alive. If so, that would cement the irreducibility of the two values.
In principle, this values listing project needn’t stop there. You might value non-living phenomena as well. Which universe would you prefer: an empty cosmos consisting only of vacuum, such as we are told may exist in the far future, or a heterogeneous cosmos with stars, rocky planets, etc., but still no life? Even more abstractly, which string of digits do you prefer: 000000000000000, or 011010110101101? If you prefer the latter in either case, perhaps you value diversity per se, thinking along the lines of “If we can’t have conscious well-being, if we can’t have life, at least let’s have something interesting happening.”
Despite appearances, I am not trying to make a case for values pluralism. I will argue in the next section that the apparent diversity of seemingly valuable things is superficial. All the examples I gave above—conscious well-being, the life of plants, a non-vacuum universe, diverse binary strings—can be expressed as manifestations of adaptive complexity. Their value differs only in scale, not in type. Adaptive complexity is a supervalue. This unified theory of value is at once more parsimonious and more actionable: if you want many independent things, you can be stymied when forced to weigh one against another.
To make the case for general reducibility, I could try to demonstrate that each thing that we value has two properties: that it is complex, in the sense of lying somewhere between predictability and chaos, and that its complexity is adaptive, in the sense that an endogenous force acts to maintain that complexity. But my task is simpler than that. I do not have to prove that each thing we value actually has those properties. I only have to show either that we wish that it did, or that our desire for it is proportional to the extent that we think that it does.
Let me begin by establishing that pleasure is a manifestation of adaptive complexity. From there I will work back up to well-being.
In a straightforward evolutionary sense, to experience pleasure is to be in a situation conducive to survival. When an organism experiences physical pleasure, this is in fact a signal that, in the interests of self-perpetuation, the organism should continue doing whatever it is presently doing. Now consider the range of such behaviors that organisms display, and the connection that these behaviors have to our intuitions about value. Even many microorganisms possess a behavioral repertoire that includes movement toward suitable environments and food, and escape from various types of physical threat. When we see that an entity has these self-preserving mechanisms, we recognize a rudimentary form of agency, and it earns a place on the list of things whose well-being we seek to maximize, along with our own. We begin granting it moral status. Consider the attention that would be captured by the discovery of Paramecium-like organisms on another planet. Even if we knew beyond a reasonable doubt that they were not sentient, their self-preserving activity would carry moral weight as a matter of fact.4
Although they do not experience pleasure, plants are significant in this way because they too possess self-preserving homeostatic mechanisms such as photo-, geo-, and hydro-tropisms. These serve plants’ interests by maintaining their access to light and water. We notice these and feel some measure of guilt when we kill a plant, especially through neglect, in which case there is nothing to offset the loss, rather than through, say, eating, which supports our own life. Plants would probably be even more compelling carriers of value if the timescale of their behavior more closely matched our own. For example, if plants perceptibly turned toward the sun as we watched, or extended their roots toward moisture as we extend our cold hands toward a fire, we would automatically regard them as more morally significant. These reactions of ours are hints about the inner workings of our value calculators. Apparently, these do not entirely depend on an entity’s ability to experience well-being, but they do seem to depend sensitively on a complex entity’s ability to preserve itself.5 On this basis, we can predict that attributions of value will be greater for, say, trees, which possess all three of the homeostatic mechanisms listed above, than to algae, which do not.
And finally let’s consider larger animals. No plant can do as much to preserve itself as even the laziest vertebrate. And again, adaptive capability and moral status (itself a proxy for value) seem to be correlated. Informally speaking, more passive animals—sea cucumbers, for instance—command less moral attention. Many octopus are smaller than sea cucumbers, so in some brute arithmetic sense an octopus is less alive than a sea cucumber, but octopus more easily attract our attention as being worthy of protection and preservation. The critical independent variable here is the exhibition of a behavioral repertoire that preserves well-being…that is to say, the capacity to go on existing as a complex adaptive system. Indeed, those behaviors, and the mental processes upon which they depend, are the source of our regard for consciousness. The upper limit of moral value extends even further if the animal’s inner life is rich in various ways—especially play and sociality—and if it acts to maintain psychological as well as physical complexity.
If we look again at our own mental life, we find another clue that adaptive complexity lies at the root of pleasure, well-being, and our value system. That clue is the psychological concept of flow. Flow states occur when a person is richly engaged in a task that requires the creative application of skills which they possess just barely at the threshold required. As the person overcomes one challenge, another immediately presents itself, which is overcome in turn—but not too easily—and so on. This dynamic relationship between an individual and the world is consistently associated with satisfaction and a subjective sense of well-being that reaches beyond the domain of the task and comes to characterize the individual’s sense of life in general [14]. And critically, for my purposes here, flow states are not well-described as an uninterrupted flow of happiness. Instead, they require a steady flow of problems and resolutions within a domain that is not entirely predictable.
Here is a thought experiment. In world A, you are the captive of a wealthy and benevolent dictator. You have superb housing, food, education, and entertainment, but no autonomy over your movements or schedule. In world B, you are free but face the responsibilities, insecurities, and constraints of normal life. Which would you prefer? I suspect almost everyone will choose world B, even if it is less comfortable than world A, suggesting that freedom may hold some amount of intrinsic value. But let’s explore further the relationship of freedom to well-being and to adaptive complexity.
Berlin [15] distinguished two types of freedom: negative and positive. You are negatively free when “there is nothing stopping you”: i.e., your actions are unabridged by external forces or people. Handcuffs, threats, and the captor in world A are impediments to negative freedom. In contrast, you are positively free when your actions are guided primarily by your own conscious will. Compulsions, deep fears, and neuroses are impediments to positive freedom. While both types of freedom bear upon how things turn out for us—and therefore how much well-being we enjoy—we are particularly keen to preserve our negative freedom. So eager, in fact, that we are willing to sacrifice some well-being to preserve it. At first glance this is surprising: one might have thought that the reason we like freedom is that it makes is happy. If that is right, why sacrifice happiness for freedom? One way to understand this is to realize that it is rooted in a deep desire for open-endedness. We want to do as we please and to experience whatever may happen. As long as it is not too terrible, i.e., as long as the complex interplay between our actions and the world doesn’t result in an inability to continue living and striving, we prefer that unpredictability.
Of course, life in society also constrains our freedom in certain ways, but we accept this because the protections and provisions of the state tilt the field toward more positive outcomes. Unpredictability is nice…as long as one is protected from violence and theft. Friedrich Hayek [16] observed that modern societies, seen as overall systems, function the way that they do because of a combination of “organization” imposed by government and “spontaneous order” that arises as an emergent property of many independent people expressing their positive liberty. The emergent order, he argues, is more sophisticated than any that could be imagined or imposed by a government. General rules of conduct can and should be set by the state, but the best outcomes for individuals will occur when they are allowed to use their knowledge for their own purposes. The emphasis Hayek places on emergent complexity is noteworthy. When he writes that “the value of freedom rests on the opportunities it provides for unforeseen and unpredictable actions,” [16, Volume 1: 56] and that “[t]he most important of the public goods for which government is required is thus not the direct satisfaction of any particular needs, but the securing of conditions in which the individuals and smaller groups will have favourable opportunities of mutually providing for their respective needs” [16, Volume 2: 2] (my emphases), it is natural to interpret him as valuing adaptive complexity at least as much as well-being itself.
Here is another way of saying the same thing. Like an organism, a whole society can be a complex adaptive system. The parts of a society—individual people, cultural groups, formal organizations such as companies, laws, the natural environment and built infrastructure, etc.—all interact and display emergent properties [17]. And, as with individual organisms, certain conditions facilitate the survival of that society and its attendant complexities and emergent properties. Those conditions are what we value.
Another of those conditions, and one that we value alongside freedom, is justice. A just society is one governed by laws crafted and enforced to ensure its continued existence. A just set of laws and enforcement mechanisms is a rule book which, when diligently executed, ensures that the power of the state will be used to protect the interests of individuals and groups, serve the public good, and prevent bad actors from taking advantage. Injustice is a sign that something has gone wrong: either the rules are inadequate or imperfectly enforced. This deficiency decreases the well-being of individuals and threatens to disrupt the dynamics of the society as a whole. Perhaps some part of the community or even the society as a whole will evaporate. This is a close analogy to the physical well-being of an organism, which can be compromised by (among other things) genetic defects and immunological insufficiency. A society is already a complex adaptive system. We value justice because it is part of what ensures that a society can continue to be one.
Some theories of justice, such as that proposed by Rawls, argue that justice must go further than ensuring the ongoing survival of society [18]. In particular, Rawls argues that justice is served by raising the well-being of the least well-off. This is easier said than done, largely because the level of that “floor” is an emergent property, difficult to predict even when one has the power to set society’s rules [19]. Rawls suggests that people with such power should exercise it while imagining that they have no knowledge of their position in society. Behind this “veil of ignorance,” policy makers would not know, for example, their wealth, race, gender, age, or religion, and so would be blocked from creating laws that allowed them to gain systematic advantage on the basis of a personal characteristic over which they had no control. Although this procedure does not guarantee equality or a “high floor” for the least well-off, it does avoid the problem of individuals or groups feeling (justifiably) frustrated that they have been treated unfairly.
Rawls’s approach, like Hayek’s, can be cast comfortably in terms of adaptive complexity. Notice the attention he gives to the use of differences that could form the basis for structural advantage. Without the veil of ignorance, lawmakers might engineer a system that works to their benefit and to the detriment of others. The threat, in other words, comes from the ability to engineer the system in a particular way. But complex systems are not engineered. The signature behaviors of complex systems are emergent rather than intentionally designed. What Rawls’s veil of ignorance does, implicitly, is ensure that any inequality in outcomes is an adaptive emergent property of a complex system rather than a maladaptive intentional property of an engineered system.
The last two values that I want to place in the framework of adaptive complexity concern ecosystems. The first of these is biodiversity, and the second is what is sometimes called intactness.
Biodiversity, like happiness, is deceptively subtle. In particular, it is more than the number of species found in a region, for the obvious reason that the “presence” of a species depends in some non-trivial way upon the number of individuals and their distribution [20]. Having a single individual of a species in a certain region should not count for much, in terms of that region’s biodiversity. Also, what we value in an ecosystem is not just a “first-order” property—biodiversity itself—but also a second-order property pertaining to the resilience of that property, i.e., the ability of biodiversity to recover after an insult of some kind.
When invasive species displace endemic species, the disruption tends to percolate through the ecological network to threaten other species. This constitutes a decrease in complexity and an increase in a particular kind of predictability: namely, disproportionally more energy will be stored in the tissues of the invading species. One can imagine a hypothetical invasion scenario in which the overall amount of life, even conscious well being, could hypothetically remain constant or even increase. But even then, I suspect that most ecologists would regard invasive species negatively because of the loss of complexity.
Over the long term, however, natural selection is adept at discovering mechanisms to exploit predictable sources of energy. If that evolutionary search succeeds—if a species evolves to eat the invader—the exploiter gains a fitness advantage. Plenty of food, and scant competition. This pushes the system back toward greater diversity and the baton is passed, as yet other species evolve to exploit the exploiter. This is how natural selection, an endogenous property of the system, acts homeostatically. At the end of section 4 I said that we value systems like this. This claim can be tested by asking how our attitude toward invasive species would change if evolution were facilitated or suppressed. I leave this as an exercise for the reader.
Intactness, the property of being wild or unmanipulated by humans, adds a further complication, because our valuing it can oppose our desire for biodiversity. Consider a mostly barren patch of mud on the sea floor. If a ship sinks there accidentally, or if we intentionally deposit a pile of debris, we create conditions for increased biodiversity. The introduced objects become substrate for corals, anemones, and other creatures, create shelter for fish and crustaceans, etc. From a biodiversity standpoint this is a good outcome, but in another sense our intervention may be seen by some as an unwelcome disruption of wilderness. This can be drawn into even higher relief by imagining a situation in which the initial conditions are a little richer. Intentionally flooding a desert, for example, will create opportunities for greater biodiversity, but at the expense of desert-adapted ones and, again, of intactness.
Let’s acknowledge the likelihood of differing views on the ethics of the sea floor and desert interventions. Rather than defending one position or the other as “correct”, I would like to see whether we can attribute the difference to intuitions about adaptive complexity.
Human-designed systems tend to be less adaptively complex than evolved ones. Significantly so. Even when adaptive complexity is our explicit goal, as it often is with the miniature systems of aquaria, terraria, biodomes, wildlife parks, and the like, we lack the know-how needed to create sustainable multi-scale ecologies. Only evolution, acting over evolutionary timescales, has that power. I suspect that those who favor the sea floor and desert interventions do so because they think that they increase local complexity, even if only for a while. They would presumably favor those interventions even more strongly if the effect lasted longer, and still more if the interventions sparked truly self-sustaining local ecologies. Those who disfavor the interventions, I wager, share the desire for long-lasting local complexity but place greater weight on their awareness of how often human management, even when well-intentioned, degrades the sustainable complexity of evolved systems. Neither group actually attaches value to intactness per se. They both want to maximize adaptive complexity, but differ in their intuitions about whether a particular intervention serves that end.
Suppose I am right that our values are reducible to a preference for adaptive complexity. The natural followup question is: why? The best way of answering that is to sharpen the question into this: How might it have come about that we have this supervalue? As I explain below, a plausible answer is that it arose through evolutionary selection effects. More specifically, I will argue that we value adaptive complexity (and therefore well-being, justice, biodiversity, natural intactness, etc.) for the same reason that we prefer non-poisonous food: had our forebears felt otherwise, we would not be here.
Consider a hypothetical community of ancestral hominins whose preferences are satisfied by monotony, situations that we would find appallingly boring. These creatures do not seek, and indeed actively avoid, novelty in any form. One consequence will be the suppression of geographical dispersion, because leaving home would bring them into contact with new and unpredictable landscapes. Resource depletion might force some emigration, but in general population density will rise to the maximum supportable by a habitat. This raises the population’s vulnerability to natural disasters such as fires and drought, as well as to socially transmissible infectious diseases. The creatures’ novelty avoidance will also manifest as a lack of curiosity and experimentation. They won’t try new foods, for example. This narrow dependency is a risk. Even culturally-transmitted practices observed among modern non-human primates, such as termite fishing in chimpanzees, may not occur in these dull hominins, because they would have no drive to understand why a conspecific was acting weird. Compared with mutants who preferred a more complex world, members of this community would be at an evolutionary disadvantage.
On the other end of the spectrum, suppose these hominins preferred chaos, situations that we modern humans would find stressful. Chronic unpredictability is not just stressful: it is dangerous. From an evolutionary point of view, the purpose of a brain is to control behavior in real-time, avoiding threatening circumstances and seeking those favorable to survival and reproduction. That kind of control requires that there be detectable patterns of threats and non-threats, as well as a dependable relationship between behavior and outcome. A brain configured to seek out chaos would yield a creature continually thrust into potentially life-threatening circumstances, and therefore one unlikely to pass on the genes that encode its chaos-seeking nervous system.
A corollary of this argument is that the psychological capacity to value adaptive complexity rests on the behavioral capacity to act on that value. Values just are behavioral dispositions. Plants and sessile filter-feeders will probably not evolve much of a value system at all, let alone one tuned to the things we ourselves care about, because there is not much they can do to act on their preferences one way or the other. That being the case, there is no net selective pressure on them to evolve brains that value predictability, complexity, or chaos. Entities that do have such a behavioral capacity (and we may include individual persons, animals, and entire societies in this class) will experience the selective pressure described above and tend to value adaptive complexity. To oversimplify a bit: the more adaptively complex one is, the more one tends to value adaptive complexity.
[1] Kauffman, S (1993). The Origins of Order. Oxford University Press.
[2] Langton, C (1991). “Computation at the edge of chaos.” In Forrest, S (ed.), Emergent Computation. MIT Press.
[3] Mitchell, M (2011). Complexity: a guided tour. Oxford Univ. Press.
[4] Wolfram, S (2002). A New Kind of Science. Wolfram Media.
[5] Gros, C (2011). Complex and Adaptive Dynamical Systems: A Primer. Second Edition. Springer.
[6] Anand, K and Bianconi G (2009). Entropy measures for networks: Toward an information theory of complex topologies. Physical Review E, 80(4), 045102. https://doi.org/10.1103/PhysRevE.80.045102.
[7] Holland, JH (2006). Studying complex adaptive systems. Journal of Systems Science and Complexity, 19, 1–8.
[8] Funtowicz, S and Ravetz JR (1994). Emergent complex systems. Futures, 26(6), 568–582.
[9] Anisman, H and Matheson K (2005). Stress, depression, and anhedonia: Caveats concerning animal models. Neuroscience & Biobehavioral Reviews, 29(4-5), 525–546. https://doi.org/10.1016/j.neubiorev.2005.03.007.
[10] Wang, Q, Timberlake MA, Prall K and Dwivedi Y (2017). The recent progress in animal models of depression. Progress in Neuro-Psychopharmacology and Biological Psychiatry, 77, 99–109. https://doi.org/10.1016/j.pnpbp.2017.04.008.
[11] Fromm, E (1980). The Heart of Man: Its genius for good and evil. Harper & Row.
[12] Wilson, EO (1994). Biophilia: The human bond with other species. Harvard University Press.
[13] Elkington, J (1999). Cannibals with Forks: the triple bottom line of 21st century business. Pbk. ed. Capstone.
[14] Csikszentmihalyi, M (2014). Flow and the Foundations of Positive Psychology: The Collected Works of Mihaly Csikszentmihalyi. Springer Netherlands. https://doi.org/10.1007/978-94-017-9088-8.
[15] Berlin, I (1969). Four Essays on Liberty. Oxford U.P.
[16] Hayek, FA von (2012). Law, legislation and liberty: a new statement of the liberal principles of justice and political economy. Routledge.
[17] Page, SE and Miller JH (2007). Complex Adaptive Systems: An Introduction to Computational Models of Social Life. https://doi.org/10.1515/9781400835522.
[18] Rawls, J (1999). A Theory of Justice. Rev. ed. Belknap Press of Harvard University Press.
[19] Davis, JB (2012). Rawlsian individuals: Justice, experiments, and complexity. Journal of Economic Issues, 46(3), 729–743.
[20] Scholes, RJ and Biggs R (2005). A biodiversity intactness index. Nature, 434(7029), 45–49. https://doi.org/10.1038/nature03289.