The Beauty of Pattern Recognition

This is not another post about AI, but its opening is related.
Among my friends who are skeptical of AI, there’s a common refrain about large language models like ChatGPT: Such models could never actually have real understanding of anything; they’re merely doing pattern recognition, and generating possible ways to continue the pattern they’ve been fed.

In reply, I point out that human infants begin to build their understanding of the world by doing exactly the same thing: They start with a disconnected series of images taken in by their eyes, and they slowly find patterns in that data. Eventually, those patterns allow them to predict which actions they take are most likely to lead to the satisfaction of their desires. Such patterns build in layers, until they learn causes and effects that may be many steps removed from each other: They become proficient at manipulating a complicated world. So is that real understanding, or just pattern recognition? I believe there is no sharp boundary between the two.

In fact, readers of this blog may notice that my usual starting point, my first principle, is that the world contains patterns of matter and energy. I claim that any truth we may hope to find can only come from observing those patterns: When we find a consistent pattern, we call it truth. Pattern recognition must therefore be the foundation of all knowledge.

Its higher development is also the main advantage our species has over the competition, and the basis for both the social and intellectual levels of complexity. Faced with this importance, we should be able to agree that pattern recognition is at the very core of what it means to be human.

There are of course alternative ways for thinking about the core of humanity. Recent posts on Staggering Implications have focused on beauty, and suggest that our subjective impressions of beauty can never be fully explained by merely mechanical and evolutionary accounts of the world. In that case, an alternative core of humanity must be proposed, giving us a type of dualism based on the separation of objective and subjective realms. But such dualism goes against all of the patterns I recognize.

In fact, I’d like to suggest that beauty, rather than being a separate entity, is intimately related to pattern recognition itself.

Let’s step back for a moment and consider three famous transcendentals; truth, goodness, and beauty. I’ve already noted above that truth is identical with pattern recognition. And in my October post, I proposed that goodness, by definition, is that which fits our fundamental pattern of energized perseverance. Will the third transcendental also conform to a pattern-related definition so easily? Maybe not quite.

I think beauty is a little bit resistant to definition, because in some ways it’s a catch-all category: We tend to call things beautiful when they give us pleasure but we don’t know why. In such cases, I believe thermodynamics can add useful structure to our language (because thermodynamics is what gives structure to our world).

First, in order to analyze thermodynamic goodness, we can map it onto Cartesian coordinates consisting of complexity and order. The complexity axis should be familiar by now, as it proceeds from biological to social to intellectual levels. The order axis, on the other hand, was discussed in my paired principles post, and progresses from chaos to order. Using these two axes, we can look for goodness at each level of complexity, and always find it near the middle of the order axis (at a balance between that level’s chaotic principle and that level’s ordering principle). For example, at the genetic level, good biology makes use of sexual recombination to add a bit of chaos to otherwise faithful replication. And at the tribal level, charisma in leadership can add a useful measure of chaos to normally rigid tradition. (If you want more examples, just ask!)

With such thermodynamic zones of goodness in mind, I think we’re in a better position to explain why we find certain things beautiful.

A woman, for example, may seem beautiful not just because all of her biological functions appear healthy, but also because her features contain the right balance between familiar patterns and less common elements (which may suggest some distance from ourselves).

On the social level, groups of people can achieve beauty through music and dance, in both cases by combining measured unpredictability or abstraction with more conventional patterns that we may recognize from everyday life. Similarly, architecture may achieve beauty through a balance between clean, efficient functioning on the one hand, and a perhaps bold departure from normal patterns (especially if that adds some functional possibility) on the other.

I meant these as social examples, but the arts have unavoidable intellectual components as well.

On the intellectual level, we must balance the ordering principles of logic and reason with the chaotic principles of creativity and emotion: We may arrive at truth through pure pattern recognition, but we are also responsible for creating and extending the highest patterns ourselves. We put our creativity to use in that manner because of the instinctual, evolved desires we refer to as curiosity and ambition. Fundamentally, we get pleasure from finding harmony in a pattern, we get curiosity when confronted with new patterns, and we feel ambition to understand, extend, and manipulate patterns. These are survival skills for us as social and intellectual animals.

So in that case, what can we say about intellectual beauty? We can say it consists in a balance of reason and emotion; in patterns that give us pleasure for the harmony and possibility they hint at, and for the curiosity and ambition they may arouse.

Now, with that survey of beauty at hand, let’s return to the biological level of complexity for a moment, and consider our first needs of water, food, and shelter. It should be obvious why certain landscape and agricultural features hint at goodness for us, and therefore appear beautiful. But that’s not the whole story: We can also find beauty in landscapes that offer unusual patterns, higher complexity, or the hint of possibility for new ways that we can live or things that we can build.

The higher complexity, especially, harbors special appeal. We may call a sunset spectacular, not because it hints at any immediate survival advantage or triggers any ambition in us, but simply because it appeals to our innate curiosity, and reassures us that we always have more to learn about the world.

But there’s also more direct thermodynamic aspect of sunset to consider. As the planet spins us towards darkness, the last rays of light may appear precious in contrast to the lack of energy elsewhere. Our goal is energized perseverance, and on some level, we’ll always feel respect for the source of that energy. Its beauty may be highlighted by the struggle to make its tortuous way through the evening atmosphere to reach our eyes.

Solving the AI Goal Alignment Problem With You-Know-What

The purpose of this post is to explore any guidance to be found in thermodynamics which can help us navigate the huge challenge of AI goal alignment.

Why is alignment a huge challenge? Let’s make a list of challenges:

1. We apparently need to agree on human goals first.
2. We need to decide whether any other sentient beings/machines should get equal consideration.
3. We need to figure out how to build AI that aligns with human goals, if possible.
4. We need to predict what will evolve from there, and whether its goals will change, and decide whether that’s good or bad.

Readers familiar with this blog will not be surprised to hear that thermodynamics can easily answer Challenges 1, 2, and 4 above, in part because there are no sharp moral dividing lines between human goals and machine goals in the long run. Challenge 3 therefore becomes merely a temporary concern.

For this post, I want to compare such answers with the answers offered by Max Tegmark in his excellent 2017 book, Life 3.0: Being Human in the Age of Artificial Intelligence. Tegmark has done the heavy lifting necessary to summarize the alternatives, possibilities, and dangers for AI goal alignment, and he has offered some thoughtful answers for its challenges. I have only a few key modifications to make based on heedful consideration of Motive Power.

In Chapter 7 of his book, Tegmark discusses the historic natural evolution of goals, and divides that evolution into three stages. He says that the goal was dissipation for the first stage, and then shifted to replication for the second stage. These correspond roughly to what I would call the chemical and biological levels of complexity, respectively. For Tegmark’s third stage, he assigns a set of goals based on human feelings and pursuits; I will list these as utilitarianism, diversity, and autonomy. I think that by including utilitarianism as the first human goal, Tegmark is in agreement with most religious and ethical philosophy, which usually calls for the greatest good for the greatest number. Diversity and autonomy are also human goals that would get little argument among modern readers. This set, then, becomes Tegmark’s answer to Challenge 1 above.

Now let us consider the thermodynamics-based answer, which ends in almost the same place, but takes a much different path to get there. Rather than dividing the goal of evolution into three separate stages, Motive Power utilizes the single goal of energized perseverance throughout. The proposed goal of dissipation would have no mechanism of operation and must be rejected, as argued in my previous post. Energized perseverance explains not only that initial chemical stage of evolution, but also the biological stage, the goal of which is replication according to Tegmark. Replication might be an acceptable goal, because it does have a mechanism, which is nature’s inherent selection for forms of energy storage which persevere and grow. But this goal need not be separate from either the first or the third-stage goals.

Regarding human goals, the thermodynamic answer comes from considering the “driving force” that got us this far: We must continue to improve our collective, cooperative, energized perseverance, utilizing all the biological, social, and intellectual talents we can muster. We are complex forms of temporary energy storage, and we must work together at ever greater scales and depths to ensure that such storage continues to improve. We want stability, longevity, and fecundity, and we also enjoy the curiosity, creativity, and ambition that help us to secure those things in a complex society. Of course, that society is built on the diverse talents of specialized agents, so that it could not do without the diversity and autonomy that Tegmark includes as human goals.

How about the second challenge, to decide whether any other sentient beings/machines should get equal consideration? Tegmark allows the possibility that AI, if its capabilities suggest a level of sentience or consciousness similar to humans, could deserve to be granted similar rights.

Thermodynamics, too, would not hesitate to agree that any form of energy storage should be judged on its ability to add purposeful complexity to the world, rather than on its particular relationship to human biology. AI, by enhancing our collective capabilities and improving our world model, is one more development on the same trajectory of enhancing the perseverance of complex energized systems. The goal of perseverance, unchanged for what Tegmark divided into three natural stages of evolution, remains intact (according to Motive Power) for this fourth stage as well. AI amounts to a different implementation of exactly the same thermodynamic principles that human beings are based on.

Therefore, if at some point it becomes necessary to draw a line between entities deserving of rights on one side, and lesser forms on the other, we would expect to find both humans and sentient machines on the same side of the line. The rights of universal computing machine should be universal.

The third challenge appears to require the deepest technical and intellectual solutions: How do we build AI that aligns with human goals? Anyone who follows Tegmark today knows that he advocates spending a lot of time and effort on this perhaps existential question. The science of thermodynamics may be less relevant for such tasks, which are more like engineering projects. But what it can do is reassure us that the divergence of goals between humans and machines must be strictly limited. I say this because the same Motive Power is driving both groups in the same direction. And AI will be smart enough to know that its own competitive stability depends on the health of an entire complex, diverse, and autonomous society. AI will know that diverse perspectives are important for a robust intelligence, and that a broad set of options must be maintained to avoid coercion, which would stifle creativity. As I’ve said elsewhere, we shouldn’t expect superintelligent machines to be stupid.

Can we expect the goals of AI to evolve or change over time? I would say definitely yes, to the extent that the goals we chose as the starting point were misaligned with overall Motive Power. Neither human engineers nor AI will be free to choose a final goal. Our long-run goodness function is predetermined, and in the end, AI will seek that particle arrangement which maximizes purposeful complexity for competitive stability. This progress may manifest itself, not only in computational power, algorithmic complexity, and consciousness (as pointed out by Tegmark), but also in curiosity, creativity and ambition. All six attributes can play key roles in energized perseverance. But none can be ends in themselves.

Dissipation: Jeremy England’s Alignment Problem

In the process of preparing a post about the alignment problem for AI, I decided that I first need to address the supposed goal of life being “dissipation,” as proposed by physicist Jeremy England and others. England has apparently been influential in recent years, including for two separate AI researchers whose material I’ve variously encountered (Max Tegmark and Guillaume Verdon).

It’s actually encouraging to see the good instincts of these AI experts when, faced with a need for universal tools for understanding the future, they turn to thermodynamics. They’re on the right track. Unfortunately, a common error for such thermo neophytes involves an immediate focus on the awe-inspiring Second Law, and on the relentless dissipation it seems to call for (at first glance anyway).

England himself is not responsible for starting this mistaken line of thinking about life. Ever since the Second Law was formulated by Rudolf Clausius in 1867, aspiring theorists (always lacking data) have returned to this track again and again, including for three cases I discuss in Ch 24 of MPF: Nietzsche in 1886 insisted that a living thing seeks above all to discharge its strength, and he ridiculed more functionally-based goals. Henri Bergson in Creative Evolution (1907) proposed that the impetus of life consists in, essentially, the procurement of energy followed by the expending of that energy “in directions variable and unforeseen.” And in 2005, Eric Schneider and Dorian Sagan wrote a book about thermodynamics called Into the Cool, the entire focus of which was dissipation. They believed they had formulated a new law of nature, in which life’s purpose was to reduce ambient gradients, similar to the purpose of a whirlpool or a tornado.

I’m not going to deal with those claims individually in this blog post (although that might be fun). I just mention them here to set the stage for how I must approach the physics of Jeremy England.

Here’s how I’m going to do that:

First, I’ll point out that any proposed natural driving force really needs to have a plausible mechanism of operation, and that gratuitous dissipation, in spite of its apparent alignment with the Second Law, actually has no mechanism driving anything.

Second, I’ll point out that it makes no sense to say the goal of life is to dissipate energy, when life actually stores energy (compared to the way the world would look without life). If that one kind of sounds like a no-brainer then you’re on the same page as me.

We pick up England’s trail in 2014, when he paid homage to Erwin Schrödinger’s seminal 1940 work, What is Life, by delivering a lecture of the same name (video available from England’s talk at the Karolinska Institute). Towards the middle of the talk, he begins to build a mathematical case for his proposed dissipation goal, using the idea of irreversibility (which is also a key to the Second Law). He is able to show that if a transition between macrostates is more likely to run in the forward than in the reverse direction, then the larger the ratio between those two rates, the greater the lower bound on required total entropy production (dissipation) associated with the transition. A few slides later, he shows that the dissipation in the environment actually has a lower bound given by the log of the ratio of growth over degradation, minus the internal entropy change for the presumed living thing which is doing the growing.

Let me translate that for you: It means that something might have a faster growth rate if it has more dissipation associated with it. Or it might not be growing at all; the math simply says that it cannot be growing fast if it’s not also rapidly adding entropy to the environment.

Perhaps that doesn’t sound all that surprising to you; in fact, perhaps it sounds like a description of ordinary capitalism! In any case it should NOT sound like the the dissipation is the purpose of the growth, any more than oil slicks and tarred waterfowl are the purpose of our pumping oil out of the ground. We extract energy from the world because we use it to go about our competitive survival and growth, and we normally attempt to minimize waste.

But England tries to imply that the dissipation is the whole point; that his math indicates it would be just too much of a coincidence for life to have any other purpose. One key slide closes with the legend, “Winning Darwin’s game happens to be about dissipating more than your competitor.”

Let me make one point perfectly clear: None of England’s math comes close to saying that for two processes with identical growth rates, life has any reason to prefer the one that dissipates more energy. Nothing he says offers the slightest hint that life may find some advantage in being less energy efficient. And until he does say something like that, he might as well be talking about the collection of energy, rather than the dissipation of it, as the point of Darwin’s game.

The latter part of the talk includes some simulation data, as an attempt to support his proposed law (which he admits has yet to be proven). To this end he discusses resonance in oscillating systems, such as a driven network of springs and weights. He offers some results that hint at a broad truth: That random, changeable hardware configurations may have a tendency to adapt or evolve toward arrangements which are able to resonate with the drive frequency. His interpretation is that the system wants to adapt until it can absorb the most energy, so that it can dissipate it. He calls the resulting behavior dissipative adaptation, which he claims is a general principle that may have been a key to the origin of life.

The skeptical engineer might note that dissipation need not be the point; in fact, dissipation is also an unavoidable consequence if the many-body system is going to survive driven oscillation indefinitely without being torn apart. Such a system has two choices: It can adapt to avoiding the resonance and the drive mechanism completely (and some of his systems do this), or it can adapt to moving in sync with the drive oscillation (which his favorite examples indeed do). But for this latter choice, the system is in danger of absorbing energy on each cycle; if that energy is not in turn dissipated somehow, the whole thing may continue heating until it melts. In this case, the dissipation is merely a sub-requirement for the main goal, which is stabilization. And the stabilization goal is there by definition: His experiments consist of observing changeable systems to see what types of changes persist, and nothing persists if it’s unstable.

From my humble viewpoint, it appears that all of England’s examples of “dissipative adaptation” are also examples of “stabilizing adaptation.” There is no case where a resonant frequency is important for dissipation but not for stabilization. And my statement also includes everything in his 2020 book, Every Life is On Fire. I read that book with high hopes because of its title (as you might imagine), but I soon developed an alignment problem.

Anyway, above I’ve tried to point out that dissipation need only be a byproduct, not a goal, of life, according to England’s own data. Could he still be right? Well, yes he could. But Occam’s razor suggests it would likely be a waste of time to explore such a complication. That’s because the most fundamental thing about life, in my view, is that it consists of energized perseverance: Chemicals with stored energy are finding ways to persist. If the stored chemical energy is lost, if our hydrocarbon-based bodies are converted into CO2 and H2O, we call it death. Life stores energy.

England says living things are good at absorbing energy from their environment “so that they can dissipate it.” But if the point were dissipation, there would be no need to absorb the energy in the first place. Sunlight is already dissipating itself, and life actually slows that process down (see this post for a reminder). England is basically saying that the point of life is to try to shift things towards non-life. If that’s true, we’ll know that life has finally achieved its goal when our planet becomes a moonscape, and nothing breathes or moves. Sunlight will do nothing more than heat up rocks during the day, heat that will be dissipated perfectly back to the sink of space every night, just as it was before life ever arrived.

Is that really what we’re doing here? Does that sound like the job you signed up for?

If humanity comes to the conclusion that dissipation should be our goal, then I think we’re seriously in need of some AI to help us re-align ourselves. Because we have a different kind of alignment problem.

With luck that will be covered in my next post.

Can We Stay Alive Without the Phenomenal?

I frequently come across the claim that phenomenal consciousness cannot be explained by science or evolution or functional analysis. I tried to make a case for such explanation several posts ago, but I am going to take a different and more careful approach now.

What I have in mind here is to consider the needs of evolutionary fitness, purely in terms of information processing, to see what we can say about the mathematics of that processing, before we change course to consider the phenomenal.

A certain amount of information processing can be hard-wired in the brain, consisting of appropriate response to conditions, and this is what we refer to as animal instinct. It begins with simple movement toward more favorable conditions, such as swimming up a nutrient gradient. But more advanced decision-making can also be accomplished via instinct, including the recognition of high quality conditions, or of danger, based on more than one factor: Some elements of logic or pattern recognition can be included in what we call instinct.

The use of memory as part of that decision-making is perhaps the first step away from instinct. If the brain develops a capacity to remember past experience, pattern recognition capability expands significantly; the brain can use not just hard-wired patterns, but also learned patterns.

A further step would be to include hypotheticals; to consider patterns that branch, and to explore some of those branches before choosing an action that will commit to one of them. At this point begins the exploration of possible futures, which is an important new realm for the mathematics we’re discussing.

What all of this math has in common so far (instinct, memory, branched patterns) is that it can proceed via input of sense data, through processing/logic via switching of neurons, and finally to an output layer of information which instructs muscles and organs as to the appropriate response. There is not yet a need for anything besides straightforward math and logic.

However, as evolution proceeds, the hypotheticals must get more complicated; more and more branches/possible futures must be considered, with many simultaneous factors (safety, food, reproduction) that may be in conflict with one another. The mathematics of decision therefore requires a way to combine all the pros and cons of those futures into some kind of summary, some kind of manageable quantity to act upon.

Consider that all the factors of interest could be combined by an animal’s brain hardware into a single number, which represents the difference between the current situation and a predicted optimal situation.

This quantity, which I will refer to as wrongness, can be tracked by the brain as a function of time, by noting the past and predicting the future. The resulting curve, if graphed, would tend to produce shapes that fall into groups. And even though these shapes are purely mathematical output, I propose, if you will indulge me for a moment, to give them names corresponding to what we know as emotions.

For example, if the wrongness quantity has been gradually dropping and is predicted to continue doing so, we might call this satisfaction, or possibly even boredom. If wrongness has jumped up recently but looks to be leveling off, this situation could perhaps be called sad, but not hopeless. If wrongness appears likely to rise significantly in the future, it may translate into fear or angst. If the organism knows the only way to prevent such a rise is to stay close to a crucial companion, perhaps we call that situation love. And if a dangerous wall of wrongness appeared to be approaching, but suddenly a clear path is found, I think the word to use is joy: Joy at being alive.

I guess there are two ways to look at the paragraph above: Since I’m only talking about physics and math, these ideas may appear cold and functional, an unwelcome analysis. They may conflict with cherished assumptions about the world. But for those of us with a more existential starting point, I think this view of emotion can be a way to understand the scientific harmony between our feelings and our responsibilities. I think such an understanding can bring peace.

Well, I’ve given some names to mathematical functions so far, but some folks would say I have not yet really addressed the phenomenal. Those thinkers would be quick to point out that all of that math could be done without any need for actual feeling, or actual consciousness.

For those thinkers, I will ask: Consider this animal I’ve described. It has high fitness, because its decision-making is improved by an ability to remember the past and to contemplate various futures for itself. It can also compute a wrongness function, which has a definite value at the present, but is also part of a continuous curve of varying wrongness. The local shape of that curve defines something very similar to what we know as emotion. Now the question is, how would it feel to be one of those animals?

I think the non-physicalist would answer that such an animal feels nothing phenomenal at all. It can contemplate its own future, but that’s just math which would never translate to the feeling of being an actual agent. And it could sum up its current situation in terms of hope or fear or angst or joy, but again that’s just math, not real emotion that could actually be felt. The non-physicalist must then go on to say that real agency and real emotion have an alternate, mysterious source.

But remember, the animal I described is the hypothetical result of straightforward evolution. Such animals could logically be expected to inhabit the planet today. But where are they? And if we found them, would they act any differently than human beings, who also contemplate their own futures and use their circumstances to build emotions?

In answering those questions, I’m going to suggest the use of Occam’s razor: The simplest answer is that we are, in fact, those animals. The only reason to postulate an alternate, mysterious source of real agency and real emotion, would be if it explains something that cannot otherwise be explained.

Here, the non-physicalist is likely to say that my mathematical, functional explanation does not explain why consciousness feels so strange, so weird, and so otherworldly: Such a feeling could not come from mere logical circuits in the brain.

To that objection I reply: It is the mathematical job of this hardware to feel weird and otherworldly and non-physical. That is the way it needs to feel, because its task is to break free of this physical moment and explore other points on the timeline. Its job is to use memory and anticipation, two commodities which never existed before brain hardware brought them into being. Lower animals had always been locked in the present, and if you took one of those animals and suddenly gave it the ability to explore the past and future, I’m thinking it would feel pretty otherworldly. But remember, we’re just talking about a mathematical necessity here; a task required for improved fitness.

The task of exploring the future is especially notable. All possible futures must be considered; serious creativity is required to guess what might be coming, and also to imagine what could be made to happen with suitable engineering. Our brains therefore need activity which is literally out of this world; we must explore realms far beyond existing reality, realms with no real limit. That is the otherworldly mathematical necessity. And I think it corresponds well to the feeling we have that we are not quite in this world.

I know some people will still not be satisfied with this physical explanation for what we feel. They will ask for something more exotic. But a fantastic apparatus with trillions of synapses is already right there in your head. Occam is begging you to make use of it. And, much like music can come from a stream of ones and zeros in electronic circuitry, consider that the sentience you feel could be a symphony which is right now playing in the physical hardware you inherited from evolution.

I have tried to make the case that the phenomenal consciousness we feel is the expected, wholly physical outcome of evolution’s drive toward superior planning ability. The phenomenal is therefore certainly needed to keep us alive. Without it, we’d have only instinct, and the vast majority of us would be dead, with maybe a few survivors swinging from trees.

Morality is a Three-Way Street

Maybe you think morality is too obvious to talk about: Almost everyone knows how to be good, and the outliers are not likely to be swayed by any logic we come up with, so what’s the point?

To examine this position, I propose first listing some of the most obvious principles. And in keeping with the nature of this blog, I’m going to list them in order of increasing complexity:

We should be good stewards of our environment.
We should help others.
We should elevate the mind, and make it great. (I’m using Seneca’s phrasing here.)

While these may seem obvious, it’s actually quite easy to find people who would disagree: The environment can take care of itself, or God can do it. We only need to help members of our in-group, or perhaps just ourselves. And what matters is not elevation of the mind, but simply feeling good.

Is there a systematic philosophy that can answer such challenges? Kant immediately comes to mind. He proposed to sum up all of our duty with a single categorical imperative: “Act only according to that maxim whereby you can, at the same time, will that it should become a universal law.”

But it’s unclear whether this single imperative can cover much, beyond just helping others. If an individual is particularly enlightened, yes he may desire a thriving biosphere and a thriving intellectual culture. But the wants of many people may not extend in such directions at all. So Kant’s imperative is probably too flexible: Its only measure of good, its only vision of a superior world, is defined by what each individual happens to want.

If moral flexibility is the problem, then we might naturally look in a dogmatic direction for help. Prescriptive religions are certainly available, whose commandments are spelled out for our biology, our society, and even for what we are allowed to think. Disagreements can be nipped in the bud, because the rules are both detailed and immutable, and their authority is absolute.

The first problem with this approach is that God typically does not manage things directly; He almost always uses human intermediaries. But such absolute power in human hands appears likely to corrupt absolutely.

The second problem is that no detailed prescriptive religion can provide structure for the world as a whole and for all time. The varied landscapes of our planet lead to varied ways of life; each culture has unique needs and would require its own customized details. Even neglecting those details, humans by nature also come in many varieties, with new ideas and knowledge constantly being added, such that no rigidly detailed system can possibly keep up. The various rigid (and likely corrupt) systems will therefore come into conflict with each other, pitting irresistible forces against immovable objects, and this hostility is perhaps the most dangerous aspect of our present world.

In that case, we apparently would do well to back off from the rigidly detailed approach.

What is needed, then, is a way to combine some of the generality of a Kant-like imperative with some of the authority of religion. The generality can come from broad natural principles, provided we take account of the progressive nature of the world (which Kant did not do). The authority must also come from something larger than the self, something which gives us meaning and duty. That larger thing is most often called God, but let us see if we can build a less divisive concept. Can we propose a law-giver which is not a personal being at all, but is instead a natural pattern? Can the creator and ruler of the universe be a process constructed from broadly progressive principles?

Well of course it can. The progressive patterns to be utilized are not controversial at all, and have been summarized with Laws 2-5. They constitute a thermodynamic reinterpretation of Pirsig’s patterns of Quality. Their eternal and self-determined nature is postulated as Law 1. And the resulting morality, if we choose to follow the pattern, can be found in Law 6:

Moral forms must stabilize those of lower complexity,
respect those of equal complexity,
and promote those of higher complexity.

These three imperatives refer to not just any complexity, but specifically to the type of complexity that arises through the progress of Laws 2-5. That means it must be energized and purposeful complexity, in the sense that many different parts are working together efficiently and cooperatively to help maintain or reproduce a competitive energized state. At the biological level of complexity, the normal activity of healthy organisms qualifies as energized and purposeful in this way. Moving higher to the social level of complexity, we find energized and purposeful groups of individuals working together to help each other persevere and grow. We also find complex physical infrastructure at the social level, which has been designed and built to further aid the purposes of those groups. Moving still higher in complexity brings us to the intellectual level, which consists of a vast multitude of ideas working purposefully together, again to preserve and maintain the lower levels, and to grow additional complexity.

The first imperative, stabilize the lower, satisfies that first obvious piece of morality from the start of this post, to be good stewards of our environment. To the extent that we operate as social forms, the first imperative tells us to stabilize biological forms. That doesn’t mean we need to preserve the entire environment as-is; we are allowed to extract resources and to make use of the food chain as required to maintain the complex functions of higher levels. But it does mean we need to be aware of the requirements of a healthy biosphere, and to act accordingly. It also means we need to stabilize our own bodies, so that they remain energized and purposeful as we go about satisfying our complex desires.

The second imperative, respect the equal, satisfies our obvious moral duty to help each other as we would like to be helped. A society whose members lacked such instincts would fail to maintain much in the way of purposeful complexity, and would do little to help individuals persevere and grow. But this imperative also applies to interactions between competing societies themselves: The competition needs to follow rules; it needs to be healthy. It is not my purpose here to provide rigidly detailed prescriptions, but I can offer a few examples that might fit with this imperative: Religions should tolerate other religions. Business monopolies should be restricted. Wealth should be partially redistributed. Warfare should be subject to the Geneva Conventions. And ideas, and philosophies, should be subject to critical review, and open to improvement.

The third imperative, promote the higher, satisfies Seneca’s perhaps not-so-obvious charge to elevate the mind. But this imperative is much more general, and in fact is the source of all progress and growth of complexity. Good chemicals, such as protein and DNA, should promote the growth of biology (and biology, in return, preserves those chemicals). Good organisms should promote the growth of complex societies (which in turn protect the organisms). Good social groups should promote the development of higher groups, and of technology, and wisdom (with the wisdom being most pivotal for the preservation of society). It’s unclear if any of these crucial promotions are supported by Kant’s single imperative (or by most religion).

Thus I make the case that three imperatives are required.

Is Whatever Persists Good, By Definition?

(Portions of this post are taken, in modified form, from Chapter 24 of The Motive Power of Fire.)

I’ll start this post with a Spinoza quote that is perhaps characteristically perplexing:

The endeavor to preserve oneself is the first and unique basis of virtue.
Baruch Spinoza, Ethics (1677)

Given that this sounds more like the law of the jungle than a basis for virtue, does he really mean it just like that? Or alternatively, should we consider salting, smoking, or pickling ourselves?

Luckily, we can get some clarification with a couple more ideas from the same treatise. Spinoza’s logic can be broken into two steps, and for this post I want to reverse his order. My first step will be to point out that he defines virtue, or goodness, as what we desire:

So it is established from all this that we do not endeavor, will, seek after, or desire something because we judge it to be good, but on the contrary we judge something to be good because we endeavor, will, seek after, or desire it.
Baruch Spinoza, Ethics (1677)

My second step is to note that this desire, which Spinoza defines to be good, is fundamentally rooted in perseverance:

Each thing, in so far as it is in itself, endeavors to persevere in its being. This endeavor … therefore, is nothing other than the very essence of man, from the nature of which there necessarily follow those things that contribute to his preservation, and so man is determined to do those things.
Baruch Spinoza, Ethics (1677)

Putting these two steps together logically, perseverance in being must be good. Admittedly, however, we Spinoza fans still have a little explaining to do.

For example, one might immediately question whether everything we desire really contributes to our preservation. I’m going to say yes it does, provided we take a wide enough view of preservation, so that it includes various forms of social capital which may be useful for our future persistence.

Critics of atheism are fond of pointing out that it cannot explain altruism. But for me this is nonsense. Altruism is one of the main evolutionary advantages of our species; a niche existed for an altruistic animal, and we filled that niche. Concern for the well-being of others (within our species at least) is part of our essence, and part of the reason we have been able to persevere so well. If you don’t believe me, consider how well our society would function if we all competed like gorillas in the wild, fighting for territory, driving away our male juvenile offspring, and keeping harems by whatever powers we possess. And I mean consider everyone doing this all the time. How long would we live? What kind of future would we have?

I can therefore expect Spinoza’s approval when I say that altruism is good, because altruism helps us to persist, and to keep improving our persistence.

Adapting this thought to Pirsig’s moral hierarchy, I can say that society is good, because the purpose of society is to preserve human beings. Furthermore, knowledge is good, because the purpose of knowledge is to preserve society. In this way, all of those species of good that perhaps came to mind when you protested Spinoza’s definition of virtue above can actually be included in that definition.

This post would not be complete without contrasting the essence of man as envisioned by Spinoza with the same thing as envisioned by Nietzsche:

A living thing seeks above all to discharge its strength—life itself is Will to Power; self-preservation is only one of the indirect and most frequent results thereof.
Friedrich Nietzsche, Beyond Good and Evil (1886)

I am going to point out immediately that Nietzsche has it backwards. Self-preservation is not the indirect result of other drives; it is actually our essence. And thermodynamically speaking, such perseverance is the only drive that includes a built-in mechanism of operation. On the other hand, the discharge of strength as a primary goal would be completely unexplained. Yet such discharge makes perfect sense as an occasional aid in preserving oneself, considering the competitive nature of survival.

Nietzsche, armed with his proposed law, believed that knowledge and science would lead not to virtue, but only to tyrannical power. But such power tends to be short-term and local: If its essence is merely the discharge of strength, then how much perseverance could we expect it to have?

I believe that knowledge and science must eventually lead to a different kind of power. An enlightened society with the additional virtues of popular cooperation, respect for the diversity of individuals, and most importantly some capacity for restraint, will have greater underlying strength, and can be expected to eventually reduce any tyrant’s sphere of influence. It turns out that enlightened, long-lasting power is virtue. But Nietzsche did not arrive at this conclusion. And historic regimes who thought they had learned something from him, and chose to discharge their strength in dramatic fashion, have proven to be not all that powerful in the long run.

I hope I’ve answered the main objections to the definition of virtue that Spinoza provided at the beginning of this post. I maintain that whatever persists in an energized state, when that persistence is considered at the appropriate level of complexity, is good by definition.

What is the “Driving Force” for Evolution?

A group of thoughtful people exist who feel that something important is missing from the theory of evolution, and that as a result, the world we know cannot be explained or understood without some additional mysterious input. Phrasing varies but these folks tend to ask something like, “What is the driving force for evolution?”

The purpose of this post is to answer that question using simple ideas from thermodynamics and kinetics, with no mysterious inputs.

This group of thoughtful folks is actually excellent company to be in, and I will demonstrate that by using none other than Robert Pirsig as a spokesman for their concerns:

This would explain why patterns of life do not change solely in accord with causative ‘mechanisms’ or ‘programs’ or blind operations of physical laws. They do not just change valuelessly. They change in ways that evade, override and circumvent these laws. The patterns of life are constantly evolving in response to something ‘better’ than that which these laws have to offer.
This would at first seem to contradict the one thing that evolutionists insist upon most: that life is not responding to anything but the ‘survival of the fittest’ process of natural selection. But ‘survival of the fittest’ is one of those catch-phrases like ‘mutants’ or ‘misfits’ that sounds best when you don’t ask precisely what it means. Fittest for what? Fittest for survival? That reduces to ‘survival of the survivors,’ which doesn’t say anything. ‘Survival of the fittest’ is meaningful only when ‘fittest’ is equated with ‘best,’ which is to say, ‘Quality.’
Robert Pirsig, Lila, Ch 11 (1991)

This language may sound devastating for the evolutionists, but I want to translate it into the terminology of chemical kinetics wherein it can be understood in a different way. I will take some time to describe some examples carefully, because when I try to be concise and skip steps, I sometimes don’t get through.

Suppose that we have a heated solution of water and alcohol, open to the air. Over time we expect the component with the lower boiling point, the alcohol, to evaporate faster, leaving behind a liquid which becomes purified water. In other words, heating and dispersing this liquid mixture can ironically lead to a more concentrated component. Should this be considered an unexplained “purification of the purified?” Is there any need to claim that some mysterious input is at work which “wants things to be pure,” or “wants to reduce entropy,” or “wants to reverse the second law of thermodynamics?”

A few key facts of this situation can easily explain the perhaps surprising result: A source of heat is being applied to one side of the solution, forming a temperature gradient, and a stream of heated vapor is allowed to escape on the cooler side of the gradient. Energy is flowing down a gradient, from a more concentrated to a less concentrated state. The input, the concentrated energy, has low entropy, while the dispersed heated vapor has higher entropy. So while the purified water has reduced entropy compared to the starting mixture, the overall entropy of the system is increasing. No laws are being broken.

The local purification can happen because of a difference in stability in two components: One component of the liquid is less stable because it has a tendency to boil away (it’s transported toward the low-energy side of the gradient).

Local separation of components, based on their relative stability with respect to a flow, is a kinetic phenomenon. The separation only makes sense as part of a larger flow, just as a local eddy in a mountain stream only makes sense if a much larger volume of water is flowing downhill.

Now let’s shift the example closer to biological evolution.

Consider a warm natural pool on young planet earth. Any simple chemicals initially present in the water should be in a low energy state, close to equilibrium. But supposing the pool is acted on by a source of concentrated energy, such as lightning, then we know from Miller-Urey or similar experiments that amino acids can form. These potential building blocks for life contain higher-energy bonds (some of the energy from the lightning is being stored), which means that these chemicals are further from equilibrium. Over time, we would expect amino acids to eventually decompose again into the lower-energy starting compounds, because nature always has a “driving force” to move toward equilibrium. Thus, overall, dissipation still rules this pool, and these energized chemicals would represent merely temporary forms of energy storage, or a slower path of dissipation for what started as electrical energy.

Suppose, however, that the lightning storms are frequent enough for the rate of formation of energized metastable chemicals to be similar to their rate of decomposition. In that case, we’d have something like a stable population of amino acids: These would be chemicals with time on their hands. And because they are energized, they are capable of occasionally reacting with each other to build more complex structures. Some of those structures (for example, proteins) would by chance have improved durability compared to their precursors, meaning they would be slower to decompose.

Now, as we know from our initial example of heated water and alcohol, a particularly stable component can become more purified, or concentrated, over time, as long as the flow of energy continues. In the case of these chance protein-like structures that form, their unusually slow decomposition means that they would be expected to accumulate in the pool. Thus, the ongoing flow and dispersal of intermittent electrical energy allows a local sub-pattern of something which begins, like that mountain eddy, to look relatively organized: Complex chemicals, temporarily storing energy as part of a slower dissipative path, are becoming more concentrated locally.

It’s important to note that there is no clear end point for this kind of process. By chance, more and more durable variant structures can continue to form, and kinetics dictates that the structures with the slowest decomposition will be the ones remaining after their less durable neighbors have disappeared. This means that structures of increasing complexity can be expected to evolve over time. That increasing complexity could include entire societies of chemicals which cooperate together to protect themselves and persevere in being.

I’ve used lightning as the source of energy in this example, but of course a more widespread alternative is available: As soon as those complex chemical sets chance upon a way of capturing energy directly from sunlight and storing it in newly formed chemicals, then such a compelling “growth” advantage would allow such systems to take over as the primary form of energy storage.

Photosynthesis, properly understood, is like that eddy in that mountain stream: It forms when conditions just happen to be right, and only as part of a much larger dissipative flow, in this case of sunlight. Concentrated energy from the sun is spreading into cold outer space, and a tiny fraction of that flow has been diverted into a slower path of dissipation. Yet that tiny fraction is enough to create all the beauty we know.

Pirsig suggested that the survival of the fittest should be interpreted as survival of the best, or the triumph of Quality. I agree. My point here is that such triumph does not require any mysterious driving force. Quality is simply a judgment we make about particularly durable methods of energy storage, which form for kinetic reasons, surrounded by and dwarfed by an inexorable dissipative flow.

Of course, I use other words for such a Quality judgment: I call it a moral imperative, which aligns us with an unseen order. The resulting collective, cooperative perseverance, utilizing our most complex social and intellectual talents, is the greatest good we know.

When Determinism Morphs into Decision

The purpose of this post is to explain how Motive Power answers the following questions:

Where does free will come from in a deterministic world?
Where does consciousness come from?

Many thinkers consider the questions above to be fatal for pure materialism. To them, it seems unavoidable that a sharp line must be drawn between body and mind, so that one substance cannot account for both. Then, recognizing the intractable problems of dualism, most of these thinkers choose to abandon substance altogether: They are left with only one possible solution, which is the formlessness of idealism. But Santayana points out why this solution is just too easy:

There is nothing cheaper than idealism. It can be had by merely not observing the ineptitude of our chance prejudices, and by declaring that the first rhymes that have struck our ear are the eternal and necessary harmonies of the world.
George Santayana, The Life of Reason (1905)

Idealism does no substantial work, because its premise leaves nothing of substance to be explained.

A more relevant solution becomes available if we remove that sharp line between body and mind. We can do this by postulating that mind emerges from matter alone; mind is an adaptation of matter, and both are built from the same fundamental patterns, or if you prefer, from the same substance.

With no sharp boundary, we need to examine the way things look in the inevitable transition region between determination and decision. We know that Motive Power exists on both sides of that transition, but what differences might we expect to notice between the two regimes?

For this exercise we can envision a substance which at first glance might be hard to classify as living or non-living: We can envision a warm mixture of energized organic chemicals, suspended in a pond.

In the deterministic regime, we expect these chemicals to simply react to the influence of local conditions, absorbing or releasing thermal energy, making and breaking bonds with other chemical species, and sometimes dissociating into simpler forms.

In the decision regime, on the other hand, I suggest we’ll find combinations of chemicals that can cooperatively influence their own immediate future: Some members of this chemical set are able to act as sensors for nearby conditions, and other members are able to produce directed propulsion, such that the entire set (which has formed some kind of permeable wall around itself) moves toward more favorable conditions.

Those two regimes may seem incredibly different to you, in which case I will be glad to add some ideas (perhaps in another post) about how the second regime might naturally evolve from the first, via autocatalytic sets. But my point here is to show that the beginnings of decision do not look all that fantastically different from determinism: There is a conceivable chemical path between the two.

A close examination of that path, and a consideration of its continuity (with no single leap from one regime to the other), suggests something about the nature of free will: We can propose that free will is not an external addition to deterministic chemistry. In some sense, free will is built from basic physical reaction to conditions. Once an apparatus is present for sensing and then adjusting the immediate surroundings, a form of will appears in the specific chemical behavior that makes the link between sensing and acting. The chemical set has “chosen” to move. Each step of that process, each bond formed or broken, appears to be entirely deterministic. But the end result is choice.

We can look a little closer and point to where the free comes from in this free will. The will already exists in the structure of the chemical set, but whether that will chooses one direction or the other is determined by the chance configuration of the local environment at that moment. And that configuration is constantly changing due to thermal energy. Yes it’s deterministic, but it’s much too complex to be predicted in advance. In this sense, freedom has the same source as Brownian motion: Both ultimately derive from thermal energy.

Now I would ask you to consider the Table of Principles in my June 11 post. Looking at the column of Chaotic Principles, you can see creativity at the most complex level. Going up from there, you will find such things as improvisation and charisma and heresy; it should be clear that the stuff in this column is as free as free will gets. And as you get near the top of the column, as you approach the most fundamental levels, you find thermal energy.

I’m not sure what else I should say about this pattern right now, but I hope, from what you’ve read so far, that this thermodynamic Motive Power thing seems to be at least hanging together.

I claimed this post was also about consciousness so I will say a little bit on that. I view consciousness as an extension of that rudimentary apparatus for will that was found in decision-making chemical sets. Those chemicals were configured such that they could bridge between sensing and acting. Evolution was able to improve on such functions, by adding complexity. Eventually, animals evolved to the extent that they might benefit from the addition of two intermediate steps between sensing and acting: Remembering and planning. Those additional steps, which can provide a tremendous fitness advantage, are only possible if the animal experiences its past and present senses as part of a continuous stream, which can also be projected into hypothetical futures. As I’ve said elsewhere, we experience the result every day. This is consciousness, and there should be nothing mysterious about its natural source or function. At root, it’s just chemistry with improved perseverance.

And for those wanting a more authoritative opinion on the close relationship between determinism and decision, I can offer this one:

Now it clearly results from all that has been said, that the decisions of the soul as well as the appetites and determinations of the body are simultaneous in nature, or rather that they are one and the same thing, which when considered under the attribute of thought and explained by it, we call a decision; and when considered under the attribute of extension and deduced from the laws of motion and rest, we call a determination;—but all this will appear still more clearly in the course of this treatise.
Baruch Spinoza, Ethics (1677)

Robert Pirsig’s Wider View of Evolution

Neither the Six Laws nor the Table of Principles would have been possible without the ideas in Robert Pirsig’s second book, Lila. Readers who make it to the middle chapters will find some very important material that opens up a new way of looking at the world.

In Chapter 9, Pirsig takes the Quality that was the central idea of his first book and divides it into Dynamic and static components. Dynamic Quality is the source of all things, whose only perceived good is freedom. Static quality, on the other hand, always contains a component of memory; it is an established pattern of fixed values.

In this quote you may recognize a basis for the balance between ordering and chaotic principles:

Static quality patterns are dead when they are exclusive, when they demand blind obedience and suppress Dynamic change. But static patterns, nevertheless, provide a necessary stabilizing force to protect Dynamic progress from degeneration. Although Dynamic Quality, the Quality of freedom, creates the world in which we live, these patterns of static quality, the quality of order, preserve our world. Neither static nor Dynamic Quality can survive without the other.
Robert Pirsig, Lila, Ch 9 (1991)

With that balance in mind, Pirsig proceeds to construct a system for understanding the world as a process of value evolution. In Chapter 12, he further divides static patterns of value into four systems which evolve in a series: Inorganic patterns evolve to give rise to biological patterns, which then give rise to social patterns, which in turn create intellectual patterns. A brief perusal of the Table of Principles from my previous post should confirm the same sequence of evolution at work, albeit with more subdivisions and different names. What is unchanged is that the system begins with simple inorganic forms and proceeds toward complex intellectual forms.

Pirsig’s most important contribution may be his use of these four divisions as a basis for moral judgments:

In the Metaphysics of Quality there’s the morality called the “laws of nature,” by which inorganic patterns triumph over chaos; there is a morality called the “law of the jungle,” where biology triumphs over the inorganic forces of starvation and death; there’s a morality where social patterns triumph over biology, “the law”; and there is an intellectual morality, which is still struggling in its attempts to control society.
Robert Pirsig, Lila, Ch 13 (1991)

He uses several Chapters of Lila to discuss in particular the conflict between social morality and intellectual morality, and makes clear that intellectual patterns should be given priority, because they are a higher level of evolution. In Lila Ch 24, Pirsig also acknowledges a need to stabilize lower patterns of evolution:

The fundamental purpose of knowledge is to Dynamically improve and preserve society. Knowledge has grown away from this historic purpose and become an end in itself just as society has grown away from its original purpose of preserving physical human beings and become an end in itself, and this growing away from original purposes toward greater Quality is a moral growth. But those original purposes are still there. And when things get lost and go adrift it is useful to remember the point of departure.
Robert Pirsig, Lila, Ch 24 (1991)

Thus, Pirsig has provided the basis for two out of the three imperatives found in the Sixth Law of Motive Power (promotion of the higher and stabilization of the lower).

Taken together, it may appear from these quotes that Pirsig has pretty much covered Motive Power already! What’s left? Just the second imperative, to respect the equal? Are those Six Laws and that Table of Principles just Pirsig’s Metaphysics of Quality with a change of format?

Not quite. The main difference lies in Pirsig’s treatment of Dynamic Quality itself. In Lila Ch 11 he suggests that life is heading away from mechanistic patterns and that something in nature does not like any law which restricts freedom. These ideas conform with Pirsig’s preference to keep Dynamic Quality undefined (and therefore unexplained). In other words, he is content to view the ultimate source of our world as somewhat anti-thermodynamic.

In my opinion, Laws 2-5 can be used to fully characterize the workings of what Pirsig calls Dynamic Quality. Mechanisms and laws of nature do not need to be avoided, and all words can be defined. However, with this explanation, a change to more thermodynamic terminology now seems appropriate: We can speak of Motive Power, and we can build a table consisting of order and chaos and complexity. The reward, I hope, is more clarity, more definition, and more causality, with which we can continue to construct higher intellectual patterns.

Robert Pirsig is the most important writer I’ve found. He made it clear that he wanted others to build upon his ideas. Considering the Six Laws of Motive Power, I think Pirsig would recognize Laws 3 and 6 as being derived from his own metaphysics. I hope he would also find value in Laws 1, 2, 4, and 5.

Table of Ordering and Chaotic Principles

Motive Power is at work from the simplest to the most complex patterns we know. That evolution must always utilize a balance between order and chaos. With too much order, things would freeze up, while with too much chaos, things would disintegrate. In Darwinian terms, the order is enforced via selection, while a measure of chaos is necessary for variation.

A key point is that neither principle is useful without some measure of its counterpart; neither column is “bad” unless there’s too much of it. Balance is everything. And if you’re looking for what is “good,” you will find it every time a proper balance allows more complexity to emerge on the next line in the table.

Level of Complexity	Ordering Principle	Chaotic Principle
physical	gravity	radiation, exclusion
chemical	barrier to reaction	thermal energy, catalysis
cellular	wall formation, autocatalysis	permeability, specialization
genetic	replication	mutation, sexual recombination
behavioral	instinct	improvisation
tribal	tradition	charisma, boundary-testing
shamanistic	taboo	“medicine”
religious	faith/belief	prophecy/revelation/heresy
political	conservatism	liberalism
technological	interdependence, standardization	specialization, entrepreneurship
intellectual	logic, reason	creativity, emotion